EFFECTIVENESS AND COST EFFECTIVENESS OF DOSE ADMINISTRATION AIDS (DAAS) PHASE 3 FINAL REPORT 11 May 2006 Project conducted by Quality Medication Care Pty Ltd in conjunction with the Therapeutics Research Unit, University of Queensland, Princess Alexandra Hospital This project was funded by the Australian Government Department of Health and Ageing as part of the Third Community Pharmacy Agreement through the Third Community Pharmacy Agreement Research and Development Grants (CPA R&D Grants) Program managed by the Pharmacy Guild of Australia
398
Embed
PHASE 3 FINAL REPORT...Chris Doran Alison Haywood Beverley Glass Elizabeth McDowell Martyn Symons James Leslie Quality Medication Care Group, School of Medicinei , University of Queensland
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
EFFECTIVENESS AND COST EFFECTIVENESS
OF DOSE ADMINISTRATION AIDS (DAAS)
PHASE 3
FINAL REPORT
11 May 2006
Project conducted by
Quality Medication Care Pty Ltd in conjunction with the
Therapeutics Research Unit, University of Queensland,
Princess Alexandra Hospital
This project was funded by the Australian Government Department of Health and Ageing as part of the Third Community Pharmacy Agreement through the Third Community Pharmacy Agreement Research and Development Grants (CPA R&D Grants) Program managed by the Pharmacy Guild of Australia
Quality Medication Care Group, School of Medicine, University of Queensland
Project Chief Investigators: Professor Mike Roberts
Quality Medication Care Pty Ltd &
Director, Therapeutics Research Unit,
Southern Division of School of Medicine,
University of Queensland
Doctor Julie Stokes
Therapeutics Research Unit,
Southern Division of School of Medicine,
University of Queensland
Project Manager: Julie Stokes
Report prepared by: Julie Stokes, Clare Ientile and Michael Roberts
Contributors: Geoffrey Lewis
Chris Doran
Alison Haywood
Beverley Glass
Elizabeth McDowell
Martyn Symons
James Leslie
Quality Medication Care Group, School of Medicine, University of Queensland i
TABLE OF CONTENTS LIST OF ABBREVIATIONS.......................................................................................................VIII
packaging and containers). Given the limited data available, reference texts
(textbooks) related to the field, reference lists of articles and abstracts from
conference proceedings were also examined. The goal of the literature search for
phase 3 was to provide a background on physicochemical stability issues relating to
the repackaging of medicines in DAAs in the absence of specific evidence.
Quality Medication Care Group, School of Medicine, University of Queensland13
Follow-up with Professor Roger Nation on any ongoing work that may be relevant
to drugs in DAAs (Professor Nation has previously carried out work in this area).
International contacts (Professor Gordon Flynn) were also approached to ascertain
whether any other research (potentially unpublished) was known.
Contributed to the various revisions of the report and the development of
recommendations.
3.1.7 OTHER CONSULTATION
Input from several other stakeholder groups was sought prior to developing the best
practice models. Meetings were held with:
Dr Kay Sorimachi and Ms Lyn Todd of the Pharmaceutical Society of Australia
(PSA) to discuss best practice in DAAs and the current state of the review of the
PSA guidelines. PSA sets professional practice standards and is the author of
existing guidelines and standards for DAA services.
Ms Lorraine Humphreys, Director of the Quality Care Pharmacy Program (QCPP).
The professional practice standards developed by PSA were included in the QCP
program (an accreditation program for community pharmacies).
Preliminary input was also sought from Aged Care Assessors. Four individual
assessors in the south east Queensland (QLD), Northern Rivers District of NSW were
contacted by telephone.
Discussions with Dr Larry Kelly (Acting Director, Therapeutic Goods Administration
(TGA) Laboratories Branch) also addressed other aspects related to the preparation of
DAAs in pharmacies. At his suggestion an email requesting a search of the Adverse
Drug Reactions Advisory Committee (ADRAC) database was sent, seeking reports of
adverse events associated with the use of DAAs.
Key informants also identified via snowballing methods were:
The Aged Care HMR Coordinator of the South East New South Wales (NSW)
Division of General Practice. This division was identified by through following a link
provided by PSA. The Division had undertaken a project to improve links between
GPs, RCFs and pharmacists aimed at improving communications and systems.
The Queensland Health Safe Medication Practice Unit. This unit was identified a
developing systems to be used in Queensland Hospitals to improve medication
safety along the continuum of care.
Dr Ro Saxon (Director, HDG Consulting Group), who was involved in a project to
develop implementation tools to support the APAC RCF medication management
guidelines (Australian Pharmaceutical Advisory Council 2002), including those
associated with medication management. This link was identified by a community
pharmacist (and Guild National Councillor) who had taken part meetings associated
with the project.
PSA also identified that the Australian Pharmaceutical Advisory Committee were
developing a document, “Guiding Principles for medication management in the
community” that might potentially impact on DAA best practice. As at February 2006,
this document had not been approved for release. On contacting APAC, we learned
that we could not make further reference to this draft document as it had not been
approved.
Quality Medication Care Group, School of Medicine, University of Queensland14
3.2 PRODUCTION OF DRAFT BEST PRACTICE MODELS
The approach taken by APAC in the production of their “Integrated best practice model
for medication management in residential aged care facilities” (Australian
Pharmaceutical Advisory Council 2000) was taken as a model for the production of the
preliminary drafts of best practice documents that were later circulated for comment. In
the APAC document, recommendations for practices were made followed by a more
detailed discussion of the issues or potential practice problems and the solutions to
these problems suggested in the best practice model. In the DAA models, the goal of
best practice was to adopt strategies available to prevent or minimize disruptions to the
medication administration system and when the use of these strategies improves the
efficiency of the overall system. Issues/problems addressed in the models were the
factors that were identified as contributing to unsafe practices, reduced effectiveness of
DAA services for patients with special needs, and inefficiencies in service provision.
Solutions were the strategies that practitioners are using or propose to use to
overcome the barriers to safe and efficient DAA provision. Draft tools to help in the
implementation of the best practice models were produced to support some
recommendations.
Quality Medication Care Group, School of Medicine, University of Queensland15
4. DEVELOPING BEST PRACTICE MODELS -
RESULTS
4.1 CONSULTATION
4.1.1 DAA EXPERT PANEL
Since providing a DAA service involves technical activities, a DAA Expert Panel of
pharmacy stakeholders was set up to inform model development from a technical
perspective and to act as a sounding board for the research team as new information is
obtained from other stakeholders that contributes to model development. A meeting of
the panel was held at the Gold Coast in March 2005.
Several key issues for model development arose from the DAA expert panel meeting:
That “best practice” refers to the process of model development and does not imply
that the model cannot be improved further.
That the best practice model should be an implementation template model (a ‘how
to’ approach) with a system focus rather than yet another guideline.
The project also needed to address dissemination and adoption of the model. It
was suggested that collaboration with the PSA team working on new DAA
guidelines was desirable. PSA guidelines would then feed into QCPP guidelines as
a way of increasing awareness and as a roll-out methodology. This relationship with
PSA was formalised.
Ideally, that other stakeholders should be expanded to include hospital pharmacists
(practitioners as well as the peak body), pharmacy information technology providers
(dispensing software, Palm etc), other IT providers (e.g. Medical Director) and
others supporting medication administration e.g. providers of medication chart
systems in RCFs.
Part of the best practice project should be to raise awareness among RCFs of what
is entailed in the provision of a DAA service, including the costs of that service.
The dearth of information about the stability of medications in packs should be
included (note: This was included in the Phase 3 proposal).
The panel noted that pharmacies from Western Australia were omitted from Phase 2.
Contacts have now been made with pharmacist DAA providers and staff from RCFs in
Western Australia. Later consultation was be sought from all states and territories
based on the panel comments.
The draft community model was considered by the panel (Figure 4.1). The panel felt
that identifying suitable recipients for DAAs should be included in the model. For
community DAA users, some baseline assessment of medication knowledge and errors
with medication management should be included as a means of monitoring the impact
of the service on these aspects.
A very early model was developed for RCFs focusing on the information flows as poor
communication was identified as a cause of many problems with systems (Figure 4.2).
The RCF systems are more complicated than community patient systems. This model
was presented to the expert panel and input invited. There was concern expressed that
it was assumed the medication chart was correct and the pack was wrong. The panel
Quality Medication Care Group, School of Medicine, University of Queensland16
indicated that this was not always the case and the model needed to reflect this. The
possibility of HIC accepting a medication chart as a prescription should also be noted.
New patient to start DAA(need already assessed)
Is it 6 months since last
formalcheck?
Yes
No
Pa
tien
t-h
eld
te
mp
late
kep
t u
p-t
o-d
ate
by
pa
tien
t, c
are
r &
GP
Inclu
din
g m
edic
ations c
hanges n
ot needin
g p
rescription
DAA packed by staff using template
Pharmacist checks DAA using template
DAA sent to patient
Pharmacist check/update template before next packing
Patient-held template/ medication record
Tripartisan Agreement*• Patient/carer• GP (+/- specialists)• Pharmacy
Formalise service to be delivered, expectations & obligations
*see break-out box for content
Full review/ renewal of template
• Patient, GP & pharmacy
Template for medication packing• Shows medication regimen & preferred
time of day for each dose
Reflect• Patient habits/preferences• GP preferences• Current medication & optimal schedule• Type of pack & packing interval• Check medications suitable for packing• Other constraints e.g vision impaired
Pre
scriptions
dis
pensed
Template approved by all
Template copy to GPNew patient to start DAA(need already assessed)
Is it 6 months since last
formalcheck?
Is it 6 months since last
formalcheck?
Yes
No
Pa
tien
t-h
eld
te
mp
late
kep
t u
p-t
o-d
ate
by
pa
tien
t, c
are
r &
GP
Inclu
din
g m
edic
ations c
hanges n
ot needin
g p
rescription
DAA packed by staff using template
Pharmacist checks DAA using template
DAA sent to patient
Pharmacist check/update template before next packing
Patient-held template/ medication record
Tripartisan Agreement*• Patient/carer• GP (+/- specialists)• Pharmacy
Formalise service to be delivered, expectations & obligations
*see break-out box for content
Full review/ renewal of template
• Patient, GP & pharmacy
Template for medication packing• Shows medication regimen & preferred
time of day for each dose
Reflect• Patient habits/preferences• GP preferences• Current medication & optimal schedule• Type of pack & packing interval• Check medications suitable for packing• Other constraints e.g vision impaired
Pre
scriptions
dis
pensed
Template approved by all
Template copy to GP
Break out box
Issues to address in agreement Obligations of parties - examples In simple language
Explanation of how service expected to work
Patient consent to service and necessary information
sharing between GP & pharmacy
Patient, GP and pharmacy agreement to service
obligations
Agreed cost of service including any GP costs as
negotiated
Billing/account aspects
Where prescriptions and original packs physically
stored
Is pack to be collected or delivered (address
timeliness & exception procedures)
How medication regimen changes to be handled
Duration of agreement & an understanding that the
situation to be reviewed every 6 months
Pharmacy to prepare packs at agreed interval
(frequency & period in advance of distribution)
Any additional support services e.g. education, Home
Medicines Review the pharmacy agrees to provide
GPs responsibilities for owing prescriptions &
prescription continuity (e.g. writing repeat
prescriptions without patient consultation) including
timeliness of prescription receipt
Pharmacy processes negotiated to fit in with GP
practice
Address GP expectations for payment & who will pay
if any cost
Patient & GP to give timely notification of pharmacy
of any medication changes including those that do
not generate a prescription e.g. ceasing a medication
Patient/carer & GP to maintain patient held template
Figure 4.1 Best practice model for operation of a DAA service for community patients
The panel raised other issues to be addressed in an RCF model:
Situations where 2 pharmacies are used, each providing DAAs or where one
pharmacy provides a DAA and another fills the occasional prescription and so that
item is not packed.
Patients getting medicines from either hospitals (e.g. outpatients for specialist
items) or mental health clinics – not from the pharmacy packing DAAs. The
pharmacy needs to maintain the profile but decisions on best practice of packing
these non-pharmacy sourced medications are needed.
What happens with non-packed medications (also an issue for RCFs)
Quality Medication Care Group, School of Medicine, University of Queensland17
GP
Pharmacy* supplying RCF
New resident
RCF
*assumes main supplying pharmacy but resident has choice to choose separate one
Current drug regimen
Order toadminister drug
Medication chart
Advise of medication
supply system,
obligations &
expectations
Order to supply
Medication chart
Authorising
payment for
supply
PBS prescription
Reminder on
continuity of
supply
RCF medication system
including DAArefill
schedule
System of supply – obligations & expectations
including responsibility for script management,
who notifies pharmacy of changes
Mutual obligations &
expectations
Figure 4.2 Best practice model for information flows necessary to operate a DAA
service for RCF residents
Subsequently, based on preliminary feedback, an RCF model showing the steps
involved in a DAA service in RCFs was devised to trigger discussion in focus groups
and structured interviews (Appendix C).
4.1.2 SUMMARY OF FOCUS GROUPS
Overall a total of 11 focus groups were conducted, with the following groups of health
care professionals: community pharmacists, hospital pharmacists, RCF nursing staff
and RCF management e.g. Directors of Nursing (DONs). Throughout the course of
these focus groups the challenges and rewards of DAA provision were examined, both
within the current context, and with regard to the development of a best practice model
for DAA supply. Through these focus groups it was possible to identify the limitations of
DAA supply, the benefits of DAA supply and the changes that can be made to ensure
that DAA provision results in optimal outcomes for all concerned. In this section, the
findings of the focus groups are summarised. The full focus group reports can be found
in Appendix C.
4.1.2.1 Participants
Of the 28 community pharmacists invited to participate, 19 (68%) were able to take part
in the telephone focus groups at the times specified. Reasons for non-participation
were: being too busy, being on holidays, not being available at the specified times and
not having very much involvement with DAAs.
Four focus groups were conducted there were 20 participants in total from 6 states,
including 2 technicians managing DAA services in two sites (one of whom was not
expected in the group but invited by her employer).
Of the 19 hospital pharmacies approached, staff from 11 (58%) were available and
participated at the specified time. Two focus groups were conducted with 11
participants in total. Participants represented 5 states, tertiary, regional and rural
hospitals, public, repatriation and other private hospitals.
Quality Medication Care Group, School of Medicine, University of Queensland18
Out of the 32 facilities contacted, 13 (41%) were able to participate in telephone focus
groups within the timeframe specified. Reasons for non-participation included:
Not being available at the dates and times chosen for the focus groups.
Being on leave.
Too busy (e.g. undergoing accreditation).
Staff shortages.
Staff changes so contact person no linger available.
In total five focus groups were conducted:
Three Directors of Nursing (DON) focus groups with a total of 14 participants from
13 facilities across 6 states representing metropolitan, regional and rural RCFs.
Two RCFs staff focus groups (Registered Nurses (RNs), Enrolled Nurses (ENs)
and Assistant Nurses), with a total of 16 participants from 2 facilities in one state.
The RCFs had high care and low care units, and some had self-medicating
residents.
Participation rates in preliminary model development are summarised in Table 4.1.
Table 4.1 Summary of participation in preliminary model development
Stakeholder group
No. invited to participate
No.participating
Participantsinvolved in Phase 2
Characteristics of respondents & comments
Expert panel 25 14 attended meeting; 11 sentmaterials only
10 had been members on Phase 2 Reference Committee
See Section 3.1.1. *3 community pharmacists on panel also participated in community pharmacist focus groups
RCFmanagement
32 13 RCFs of 8 participants involved with Phase 2
Nine interviews were conducted with non-pharmacist (and non-pharmacy student) staff
i.e. dispensary assistants/technicians (DAs) who were directly involved in the packing
and checking and other aspects of the operation of a DAA service. Five DAs worked in
pharmacies whose owner/manager had taken part in the pharmacist focus groups. The
interview questions were more specifically targeted to in-pharmacy procedures that
those asked of community pharmacists, and sought information on training and
knowledge of these key participants in DAA services.
The represented 5 states, metropolitan and regional centres, and provided services to
RCFs and/or community patients. Community patients included some intellectually
handicapped patients who lived in special accommodation, indigenous health services,
drug and alcohol rehabilitation patients and patients in a renal unit. Some respondents
prepared more than one pack type. The pack types provided were:
Webster multidose 7 day packs (6 respondents) including 1 ‘jumbo’ pack provider
Webster unit dose (systems), a service provided to RCFs (4 respondents)
Webster Flexipaks (1 respondent)
Douglas cold seal packs (1 respondent)
Douglas non-cold seal packs (1 respondent)
Medico packs linked to the FRED dispensing software (1 respondent)
Dosett boxes (2 respondents) – there were provided to some community patients at
patients’ request.
In reviewing the flow diagram sent to respondents in advance (see Figure 4.3), five felt
that steps were missing:
The return and correction of a pack if there were changes (a point raised by two
respondents). If the medication change did not involve a new drug (e.g. drug
ceased or a dose time change), then the steps of dispensing the prescription and
checking the dispensed medication would be skipped, i.e. the workflow would jump
from step 2 to step 6. Further, there may be a need to trigger Step 13, the return of
the old, now incorrect, packs to the pharmacy prior to effecting any changes in the
pack. Changes to packs was a “huge time factor” especially when GPs visited an
RCF during the week covered by the DAA (when there could be as many as 10
charts/day with changes). One DA indicated that some doctors annotated
prescriptions or orders “make changes when the next DAA is packed” – this was
felt to be helpful and reduced waste.
Quality Medication Care Group, School of Medicine, University of Queensland25
One respondent also felt that the diagram could indicate that non-packed and “prn”
medications could be prepared in a separate location and by separate staff to those
preparing DAA packs.
Explicit mention of inventory management processes (2 respondents) when
assembling medications to dispense and pack i.e. is there enough stock on hand to
fill packs at the next scheduled packing day? If not, the workflow would include
ordering and receiving this stock from wholesalers/distributors.
The explicit inclusion of a step that when dispensing the last repeat on a
prescription, a repeat reminder be printed and sent to the doctor (to reduce owing
prescriptions).
Steps to deal with telephone orders.
While some of these additional steps were felt to be specific cases of a more general
workflow, aspects such as stock control and notifying when a new prescription is due
were later added to the best practice models.
4.1.3.1 Roles and practices experienced by respondents
In examining the roles of the dispensary assistant/technicians (DAs) in providing a DAA
service, respondents were asked which steps in the work flow diagram (Figure 4.3) that
they were involved in.
1. Medication order• Prescription from GP• Medication chart from RCF
2. Develop or update pharmacymedication profile
0. GP orders & prescribes medication
2a. Support activities
• Prescription management*
• Accounts
3. Dispense prescription
7. Pack medication according to profile
8. Check packed medication
10. Deliver / collect medication
11. Medication receipt &/or storage
12. Administer medication
* If chart for administering and prescription for payment
Non-packed
medications
13. Medication returns to pharmacy
Different procedures needed for:• New resident• Medication change• Other situations eg. going to &
returning from hospital
External Internal to PharmacyKEY:
4. Check dispensed prescription
6. Prepare for packing (labels, assembling medications etc)
5. Store dispensed medications not yet packed
9. Store packed medication
Figure 4.3 Work flow diagram used as a prompt for dispensary assistant/technician
interviews
At various times, the respondents performed the steps internal to the pharmacy
excepting steps 4 and 8, checking activities to be done by a pharmacist, according to
PSA guidelines (Pharmaceutical Society of Australia 1999). Developing and reviewing
the medication profile might be expected to be a pharmacists professional
responsibility, however, two respondents mentioned that they were involved in entering
medications into the profile while another did this step in conjunction with the
Quality Medication Care Group, School of Medicine, University of Queensland26
pharmacist. One respondent explicitly mentioned that this activity was performed by the
pharmacist. The main focus of technicians/assistants was on the steps related to actual
packing of the DAA and the processes involved in prescription management (including
following up with doctors to get new prescriptions or warn that a new prescription would
soon be required). One technician noted that she liked the responsibility of looking after
the DAA service. This same technician visited new community-based DAA users in
their homes soon after DAA were started to check that the patient was managing the
new system.
In reviewing the work steps, most respondents described their processes. Some
practices deviated from those in the PSA guidelines (Pharmaceutical Society of
Australia 1999) or professional practice standards [Pharmaceutical Society of Australia,
2006 #321]. For example, in 3 cases, packs were prepared based on the printed
backing foils or header cards rather than any profile. The PSA guidelines require
reference to the original medication chart or prescription each time a DAA is filled while
the standards require reference to the current prescription or medication regimen (p18).
One reason for the practice was that the foils were the most up-to-date information and
the profile had to be updated to actually produce the foils.
In the interviews, practices related to several specific aspects of the service were also
explored:
How non-packed medications were handled
Procedures for medications returned in DAAs
Starting a new customer on a DAA
Templates, tools or materials used to facilitate DAA supply
4.1.3.1.1 Non-packed medications
The PSA standards [Pharmaceutical Society of Australia, 2006 #321](p18) include two
indicators related to non-packed medicines i.e. that the medication history lists those
medications not delivered in a DAA and that the patient or carer has a list on non-
packed medicines. Procedures to ensure that non-packed medications were supplied
sometimes depended on whether medications were supplied to an RCF or to
community patients:
Non-packed medications were recorded in the profile (4 responses) and were
added to the delivery list for community patients as required (patient notified
pharmacy). In once case, the pharmacy kept track of need based on expected use
and items were sent when “due”.
Non-packed medications were written on the header or top card of the DAA.
The community patient advised the pharmacy when supply was required (2
responses) but support was provided by the pharmacy e.g. if a patient was forgetful
or had not requested re-supply, the pharmacy would ring to check.
Supplies of non-packed medications were checked as part of an “active home visit”
approach to delivery
Non-packed medications were ordered by the facility as needed (3 responses)
(after routine, monthly supply of eye drops or RCF orders resulted in stockpiles in
the RCF in two cases). Stocks were monitored on an informal basis. One pharmacy
initiated a monthly order list for non-packed medications for some RCFs.
The DA checked non-packed medications e.g. eye drops, when in the RCF and
sent supplies as required.
4.1.3.1.2 Medicines returned in DAAs
Quality Medication Care Group, School of Medicine, University of Queensland27
Respondents were asked how returns were handled since all medications in a DAA
should have been taken by a patient so that the return of medications in a DAA can
potentially indicate a medication management problem. Further, the PSA guidelines
recommended that “unwanted medicines from the DAA should be returned to the
pharmacy for safe disposal and should not be reused. The pharmacist should request
information…. and note the reason(s) for any unused medicines” (Pharmaceutical
Society of Australia 1999) and follow up returns “known or suspected to be due to non-
compliance”. Actions to deal with the medications included:
Placing returned medicines into the Return Unused Medicines (RUM) bins (3
responses). In one case, this explicitly referred to medicines in original packs and in
another, to multidose pack contents.
Re-using drugs returned by a given patient for that same patient’s next pack
provided the integrity of the medication was felt to be maintained (2 responses).
The batch numbers recorded in one pharmacy were used to indicate older packs
whose integrity was more likely to be compromised. The reuse of certain
medications was used to minimise waste and cost to the patient.
Re-using if needed for the same patient if for on-going use otherwise disposal of
medications (2 responses).
Discarding all medications returned from indigenous communities.
Recording, monitoring and followed up varied, with the following procedures being
described by DAs:
Having follow-up procedures for medicines that are returned, where the situation
was recorded and reported up to the pharmacist (and possibly the doctor) for
further action (2 responses). In one pharmacy, incidents were reviewed weekly.
No formal recording of returns but informal monitoring and follow-up if further action
is required (4 responses).
Packs used were disposable so not usually returned but for some community
patients where there was a suspicion of problems, the delivery person was asked to
bring back packs to check whether the patient was coping. No records were kept of
returns but actions taken to address any concerns were formally documented.
For indigenous communities, a stock-take of packs in the health centre was done
each fortnight as compliance varied; there was no way of finding out reasons for the
unexpected non-use of packs. The pack type used was disposable so no returns.
Other pack types require the person to return a plastic frame, for example, and this
requirement did not match philosophy of the community.
4.1.3.1.3 Starting a new community DAA customer
The practices of pharmacies and roles of DAs in starting a new customer on a DAA
varied. Some (2 responses) had no involvement with new DAA users. Some DAs (4
responses) undertook much of the start-up activities including explaining the costs to
the patient or their family, what’s involved (including individualising delivery procedures
(some using a leaflet to explain)), what is needed by the pharmacy and any pharmacy
policies. Two technicians mentioned occasionally making an assessment of the
patient’s suitability for a specific type of DAA, even if informally. One technician was
also involved with liaison with or referral to the GP when new community patients
wished to start using a DAA. Another DA sometimes visited new DAA patients in their
homes to explain the system and collect medications for initial packing. The visit
include demonstration of a sample pack and how to use it (said to be “very helpful” by
Quality Medication Care Group, School of Medicine, University of Queensland28
the DA). Four pharmacies had information sheet to help explain the service (e.g. need
for regular visits to the doctor, the fees involved) to new DAA users.
4.1.3.1.4 Supporting templates, tools or materials
Awareness and use of tools to support DAA activities was limited. Only some DAs were
able to describe existing templates, tools or materials used to facilitate DAA supply,
and these 4 of these DAs described several each. In a number of cases, these were
tools provided by the manufacturer of the DAA type used. Tools used included:
Manufacturer’s forms (3 responses, e.g. entitlement, Medicare number checking.)
and sample packs (1 response).
Manufacturer’s procedures and IT systems (2 responses).
Other IT or dispensing systems that allowed the profile to be used as a batch sheet.
A checklist for new patients (to ensure all necessary actions had been taken).
A change notice form requesting old packs back so they could be changed.
An order form for non-packed medicines for use by a clinic and indigenous
communities.
A template in the computer to generate requests for scripts on the last repeat. The
pharmacy tried to align re-supply/re-order times so that several requests could be
sent at once.
An incident book in which medicines returned by DAA patients, errors in packing
and out-of-stock events were recorded. This book was reviewed weekly.
A DAA collection chart to record collection of DAAs by community patients.
A “request for medication chart” to send to a doctor for RCF residents and a
“medication change” form. This was also used to get information from hospitals.
A recorder sheet for the packer to note that the last prescription was used and a
new one would be required from the GP.
A form to advise the RCF of changes in individual packs e.g. a medicine ceased so
here is a new pack, send the old pack back, or, an additional medicine is in an extra
pack this week but will be in a pack with other medicines from next week.
Three DAs worked at sites with a quality improvement approach to practice, each
developing and using a number of tools. Two DAs mentioned using these tools to
monitor and solve problems. Two other DAs demonstrated little appreciation of quality
improvement, one saying there were no problems with existing tools and another
stressing the need to minimise paperwork. Some improvements to tools and materials
were suggested as being helpful such as:
Automated reordering of new prescriptions from doctors (writing up lists of doctors
and prescriptions required, faxing the list and following up was felt to be the most
time consuming activity).
Using the manufacturer’s software that integrated DAA profile and label production
with dispensing (the pharmacy prepared their own version of the packing form and
used free-form labels felt to be not very professional by the DA).
As checking logs and delivery logs were documentation tools included in the PSA
guidelines and in the QCPP resource kit, respectively, use of these tools was
investigated. Records of packing and checking varied:
Checking was recorded on a checking/packing sheet (4 responses)
The packer and checker each signed the foils.
No checking logs were kept but checked packs are put in a separate location (1)
Quality Medication Care Group, School of Medicine, University of Queensland29
Maintenance of a delivery audit trail (as required by the distance dispensing PSA
standard [Pharmaceutical Society of Australia, 2006 #321]) also varied.
No delivery receipt records were kept in one pharmacy.
Delivery and collection (from a patients home or RCF) logs were kept (2 responses)
while another site where patients collected DAAs from the pharmacy kept a
collection log.
One pharmacy supplying isolated indigenous communities used registered couriers
and logged every aspect of the delivery.
4.1.3.2 Problems associated with DAAs and their solutions experienced by
dispensary assistants/technicians
Problems experienced by DAs were generally in the areas of:
1. Obtaining and maintaining a current medication profile. Includes problems learning
of changes to medications and ensuring these changes are reflected in packs.
2. Obtaining valid and legal prescriptions in order to dispense medications in time for
the cycle of packing a DAA. This led to problems with “owing” prescriptions, with
continuing medications supplied before the GP sent a PBS prescription.
3. Patients, RCF staff and doctors having unrealistic service expectations or lack of
awareness of constraints faced by the pharmacy in providing a DAA service.
4. Continuity of medication information and supply issues when patients were
admitted or discharged from hospitals (a combination of problems determining the
current profile, getting prescriptions in a timely manner and unrealistic
expectations).
5. The risks of packing errors, including problems reading medication charts.
6. Control/supply of non-packed medications in RCFs and for community patients.
7. The risks of patient, carer or RCF misuse of the DAA.
8. Needing to record and type in the medication list in many places e.g. the
dispensing software, any profile and the DAA labelling software.
Details of problems or contributing factors and strategies used by pharmacies
employing the respondents are described in the following sections.
4.1.3.2.1 Current medication profiles
In order to pack a DAA that is correct at the time of use, the pharmacy needs to know
the full and current drug regimen, and this information must be reliable. For community
patients, some changes that involve changing a dose time or ceasing a drug would not
require the GP to write a new prescription. Relying on patients to advise of changes or
to provide a full drug regimen was felt to be inadequate. Obtaining a full and current
drug regimen from the GP at the time of starting a DAA was the preferred solution –
generally the regimen was to be in writing.
Of particular concern was detecting changes that do not result in a prescription being
written (e.g. to stop a drug). Pharmacies must have mechanisms in place to detect and
implement medication changes that result in a new prescription and those where no
prescription is written. Solutions and tools suggested were:
Only accept notification of changes in writing from the GP (3 responses). In one
pharmacy, with any change the GP had to provide a new, complete list of current
medications including non-packed medicines, not just writing the change.
The pharmacy reconfirms the profile with the GP every 3 months (in writing) when
the patient sees the doctor. This facilitates concordance and ensures that the
patient sees their doctor at least every 3 months.
Quality Medication Care Group, School of Medicine, University of Queensland30
Faxing the doctor a medication change or addition form to complete and fax back
as documentation of the change.
Have forms to fax to GP to confirm that the pharmacy profile is current.
Establish that the GP has the responsibility to notify the pharmacy of changes
especially if a drug is stopped.
In RCFs, problems related to an instruction for a medication change can be due to:
The RCF notifying the pharmacy of only some of the changes for a patient,
particularly where the change does not necessitate the supply of a drug (e.g.
stopping a drug or changing the dose time).
The doctor writing a change in the patient notes but not on the medication chart.
The doctor giving a verbal order (telephone order) to nurses so that the order is not
on the medication chart.
Where the prescription conflicts with the medication chart. The order can be written
in the medication chart but the doctor writes a slightly different order on a
prescription written on his or her return to the surgery.
The change not communicated in a timely manner to the pharmacy. Lack of
communication between GPS and the RCF was felt to be the main contributor.
The usual approach to this problem was to specify that the whole medication chart
must be faxed through to the pharmacy before a pack is change, not just the page with
the change.
One DA who was involved in providing DAAs to indigenous communities and hospital
clinics found that medication orders (the charts and prescriptions) from the doctors
were not always completed properly, with details about the drug regimen or details
required for PBS claims, etc, being omitted. Nurses and community health workers
were trained to check for these details at the time the order was written.
Lack of communication of medication regimen changes within the pharmacy was
another possible problem. Solutions were:
Stressing to all pharmacy staff the importance of communication of medication
change information as part of the supply of DAAs.
Having systems in place to record notification of medication changes.
4.1.3.2.2 Timely receipt of valid and legal prescriptions
Continuity of supply of ongoing medicines for DAA users requires continuity of legal
and valid prescriptions. In some situations, medications were supplied in advance of
the receipt of a prescription because of patient need and the time constraints of
packing a DAA. This can lead to the problem of “owing scripts”, a situation that had
been experienced by all respondents. The practice is contrary to regulations and, if a
prescription is not ultimately received, the pharmacy also bears the cost of the
medication. One DA said that “owing scripts” was more a problem for community
patients because the doctors wanted to see the patient before writing a new
prescription and it can be difficult for a patient to get an appointment in time.
A related problem was that of GPs giving the pharmacy a telephone prescription (which
was then dispensed in good faith) but failing to reduce that verbal order to a written
prescription in the timeframe required. In one area servicing a renal unit, phone orders
from the doctor were relayed by the nurses to the pharmacy, increasing chances for
translation/transfer errors. Procedures were put in place to address phone orders
including a requirement to fax the written order to the pharmacy. Another DA indicated
Quality Medication Care Group, School of Medicine, University of Queensland31
that if a doctor had given the RCF staff a phone order, the pharmacy had forms to fax
to the doctor (a medication change or addition form, also used for community patients)
to get documentation directly from the doctor.
The need to dispense prescriptions in advance of when the medications would be
taken by the patient and frequent medication changes (that increased the rate of
wastage) also generated problems when GPs complained that new prescriptions were
ordered “too early” (as flagged by prescribing software). This was particularly a
problem with PBS items requiring an authority. [Authority prescriptions are written for
certain, generally high cost, PBS items where the quality, re-supply dates and other
aspects of the prescription required prior approval from the PBS. If these conditions are
not met or details do not match when the pharmacy submits a claim for payment to the
PBS, the claim is rejected and the pharmacy bears the cost of the dispensed
medication.]
One DAA noted that any system to improve communication and prescription continuity
broke down if it was blocked by the doctor’s receptionist. This problem was
experienced especially in relation to Schedule 8 medication prescriptions.
How busy GPs were and their workloads were two explanations offered for the problem
of untimely prescriptions but a lack of appreciation/respect or understanding of
prescribing and dispensing regulations was felt to be a contributing factor (see
4.1.3.2.3). The efficiency of DAA packing was reduced if the prescribing and
dispensing of medications was not organised in advance of the day DAAs were
packed.
There were two main strategies used to overcome the “owing prescription” problem:
Pharmacies had a policy not to do “owing scripts”. Some had procedures to inform
the RCF, patient and GPs that no medications would be supplied without a current
prescription. Others advised the RCF or patient that a new prescription would be
needed and gave the RCF or patient the responsibility to organise a new
prescription with the GP (with some additional reminders from the pharmacy). The
latter strategy was not optimal because any delays reduce the efficiency of packing,
but one DA noted that “patients learn what to expect and are well trained” i.e. they
take ownership or responsibility for the continuity of prescriptions (patients were
charged for DAAs at $3.20 per week in this pharmacy). An alternative method was
to ask that the prescription be faxed to the pharmacy and the pharmacy would
collect the original at a later time.
Many pharmacies had reminder systems, some elaborate, to notify GPs when new
prescriptions would be required for a patient. If the last repeat was dispensed, a reminder was printed and sent to the doctor.
One DA felt that the dispensing system she used was inadequate in that it could not
generate a list of prescriptions due in the next two weeks. Unless this information was
recorded manually, GPs might be contacted on many occasions where a different
prescription for a different patient was requested each time.
One solution being developed in a region was to hold talks between the pharmacies to
work towards common strategies to be used by the pharmacies in interacting with the
GPs so, for example, if a doctor is asked to write a prescription, pharmacies use the
same request form and the same procedures (e.g. that the pharmacy will ask the
doctor for all items due in the next set period).
Quality Medication Care Group, School of Medicine, University of Queensland32
4.1.3.2.3 Unrealistic expectations
Many of the problems with communication of medication changes and appropriate and
timely prescriptions relate to a lack of understanding or appreciation by RCF staff and
doctors of the rules and regulations with which the pharmacy must comply to dispense
and supply medication – “staff at the RCF and doctors and patients aren’t aware of the
consequences to pharmacy of them not doing their jobs”. For example, RCF staff
expected the pharmacy to supply medications that were ordered by the doctor on the
medication chart. Staff did not understand that the chart was a legal order for
administration of the medicine but not a legal order to supply for the pharmacy, and
that a prescription was required. Another DA said that communication problems, not
following agreed procedures and problems with the rules and regulations were more
often a problem with new doctors visiting an RCF but also occurred with new RCF staff.
There was an attempt to implement an induction program about the pharmacy system
for the RCF in response to “more things starting to go wrong” but this failed since it was
up to the RCF to initiate the process.
In these situations, mutual awareness-raising about the situation from the perspectives
of all (RCFs, GPs and pharmacy) and education was suggested as solutions:
One pharmacy met with the Director of Nursing at the RCF they serviced once a
year to confirm the quality of the service (e.g. meeting expectations).
Tripartisan meetings to discuss issues such as owing prescriptions were held in a
small community. These were effective in raising awareness among fairly
supportive local doctors but problems still arose with locum doctors.
An induction for new doctors visiting an RCF (and new RCF staff) to cover the
RCF-Pharmacy systems and how to deal with the pharmacy supplying the RCF.
The “cracked record” approach, repeated explanations of regulations/procedures.
A written tool explaining the systems would be helpful.
4.1.3.2.4 Continuity of medication information and supply with hospital admission and
discharge
From the DA perspective, there were problems with admission and discharge to
hospitals. When a patient is admitted to hospital, unless the pharmacy is notified, new
DAAs may be packed that are not required. This medication is then wasted, particularly
where medications are different at discharge. Wastage also occurred when patients
did take their medications in the DAA to hospital. Some hospitals did not use these
medications and re-dispensed a new supply. One DA had the experience where the
medications of an RCF resident were sent in the ambulance with the resident except
for the patient’s dangerous drugs (Schedule 8 (S8) medications, that must be recorded
in a register) as the RCF would not send S8s with ambulance officers. The pharmacy
was then called to transfer the S8s to the hospital.
More issues were raised with discharge from hospital. In order to pack a DAA, a
pharmacy needs to have reliable information about a patient’s whole and current drug
regimen plus valid and legal prescriptions are required for any drug dispensed. This is
even more complicated when a patient is discharged back to an RCF where a valid
order from which an RN can administer a drug is required. Lists of drugs at discharge,
especially those prepared by non-medical staff, do not meet the requirements of the
various Poisons Regulations for an order to administer. The experience described by
one DA was not uncommon:
Quality Medication Care Group, School of Medicine, University of Queensland33
A patient arrived at the RCF with little warning, accompanied by a list of current medications
at discharge that had been prepared by the hospital pharmacy (a “MedProf”). The RCF
wanted medications on hand by the time the next medications were due and so expected
the community pharmacy to pack a DAA based on the MedProf. The pharmacy did not do
this as the MedProf was not a legal prescription or administration order. The doctor was
required to attend the RCF to write the medication chart and the prescription, which delayed
the supply of medications. If the doctor cannot see the patient on the day of discharge, the
doctor may call to give a verbal prescription to pack a DAA as per the discharge MedProfs.
The pharmacy is asked to pack the medications supplied by the hospital (so there is no
dispensing fee for the pharmacy to defray this cost) and then, when the doctor sees the
patient the next day, there are medication changes that generates rework in packing a DAA.
Patients discharged from hospital with PBS prescriptions (valid, legal prescriptions) still
generate significant work. For example, after receiving the prescriptions, the patient
profile is cross-checked and any differences clarified with the patient’s GP before the
DAA is packed. If there is not enough time to do check and confirm the complete drug
regimen, to avoid re-packaging if there are later changes, a small amount is supplied in
original packs until the profile is confirmed and a DAA can be prepared.
Failures to do with admission and discharge planning means that medication continuity
is not addressed. There needs to be a reliable and legal flow of medication regimen
information between settings. Some experiences or suggested strategies that address
some of these needs include:
Two pharmacies in a smaller community are advised when patients go to hospital. In one, packs are put on hold until discharge. The pharmacy packs from the discharge
note sent from the hospital (where there is always a doctor on duty). Problems
occurred, however, it a patient returned from the capital city where they had been
seeing a specialist or were in hospital.
In another, if a patient is admitted to the public hospital, the patient’s own medicines (in
DAA and original packs) are used. In the private hospital, DAA medicine are not used
but the remainder of previously dispensed medicines in the original pack is used when
the hospital “gets the pharmacy to send up all the medicines from the patient’ bucket [of
stored medicines in the pharmacy] with a copy of the pharmacy profile”. On discharge,
the pharmacy “collects medications from the hospital and the doctor’s medication chart
or discharge summary which is used to update the profile. [We] then pack up what’s left
of that week to get back in sync [packing cycle synchronisation]”.
A solution offered in another smaller community was for the GP to be more involved
in discharge at the hospital.
While not directly related to a hospital admission, some community pharmacies packed
unusual medicines supplied by hospitals (e.g. trial drugs or medicines supplied under
Section 100 or other provisions) into DAAs for community patients. In this case,
continuity of supply was an issue as the pharmacy had to rely on the patient to bring
medications from the hospital to the pharmacy to be repacked into a DAA. This was
often not timely with respect to the packing cycle especially for mental health patients.
4.1.3.2.5 Other problems
Errors in packing DAAs were explicitly mentioned as a problem by some DAs:
The need, in some systems, to re-key drug regimen firstly into the dispensing
program and then into the DAA program was felt to be a potential source of errors.
Double checking of the profile against the dispensing system was used.
Quality Medication Care Group, School of Medicine, University of Queensland34
Medication charts could be hard to read due to bad handwriting and so the wrong
drug dispensed, however, prescribing errors could occur with computerised charts,
so that confirmation with the prescriber was required. This stopped work flow.
Variable doses where the GP crossed out parts of the order and overwrote an order
with a new dose meant that it was not always clear what the prescriber intended.
Human error through decreased concentration and not checking properly could
lead to errors leaving the pharmacy. Noise and interruptions were felt to contribute
to errors and increased the checking time. Good packing and checking procedures
and the right environment (a dedicated area with enough space, quiet area away
from the dispensary or customers and without interruption) were needed (5
responses).
New staff unfamiliar with the medicines could make more errors. Efficient and
effective staff training was felt to minimise errors.
Few error logs were maintained. Two DAs mentioned recording errors. In one
pharmacy, these were reviewed weekly and in another, only errors that were detected
after a pack left the pharmacy were recorded but there was informal monitoring of
unrecorded errors detected and fixed before the DAA was delivered.
Users of the DAA also made errors. One factor contributing to these errors was a
different “generic” brand being packed in the DAA to that on the medication chart. In
one pharmacy, the DA made sure that the brand written on the chart was the brand
packed and so this required the GP to write the appropriate brand name (not always
the originator brand name) on the chart. This policy was developed by the pharmacy
over the years and was well known to the GPs.
Solutions to omission or oversupply of medications not packed in the DAA included:
Recording non-packed medications on the profile.
Auditing supplied on hand, either in the RCF or in the patient’s home.
Monitoring prescription or supply refill intervals.
Problems also occur when a community patient is unable to manage the DAA, for
example, because of confusion. Problems using the pack correctly were linked to poor
explanations to the patient about how to use the pack. One DA visited a new DAA
patient at home, soon after a DAA was started to check that the patient could manage
packed medications. Another DA explained that patients were trained/counselled to
leave a dose in the pack if they had forgotten to take it at the correct time.
One DA had observed problems due to a community nurse sometimes “popping” out
the wrong blister. She attributed this to the specific type of DAA saying they were “not
so straight forward for busy nurses or nurses not familiar with them”.
Poor integration of the various software programs involved contributed to inefficiency
(due to re-keying of medication details) and errors. In one instance, the drug regimen
was keyed three times (the profile form in a word processing document, the dispensing
program and then the packing program). One DA used a system where the dispensing
and DAA labelling functions were integrated.
One DA raised problems with the delivery of medications to community patients who
were not home at the time of the delivery. The pharmacy staff did not like to leave
medicines “on the front porch” as requested and did make some re-attempts to deliver,
but at times, medicines were left.
Quality Medication Care Group, School of Medicine, University of Queensland35
4.1.3.3 Training and instruction on DAA preparation received by dispensary
assistants/technicians
The training on DAA packing procedures technicians received was largely informal and
involved observing the processes then doing the processes, sometimes under
supervision and sometimes with the support of written procedures (Table 4.5). The
practical approach was felt to be good but in some instances, more structure or formal
training was felt to be an improvement.
Table 4.5 DAA training received by dispensary assistants/technicians
Type of training received Comments on quality and improvement needed
On the job supervision for several days when starting packing DAAs
Supervision was good so no need for improvement. Supervised practice was better than reading “essays” or lots of written procedures.
A short period of observing a peer then followed in-house written procedures. Also a brief mention of DAAs in the Guild Dispensary Assistant course completed later
In-house training was felt to be good because hands on learning was needed. The Guild course could have been improved – it emphas- -ised what a DA can or can’t do but not about the packs or procedures (the “how to do”).
On the job, practical training after being given a verbal overview of the whole system by pharmacist e.g. shown the profiles, described where stock came from and how to get all the necessary equipment
Overview was good but a chart (like the one provided to the interviewee) would be a good idea.
Trained by pharmacist. Given the documen- -tation, the packing chart was explained plus how to use it when packing. Packing then demonstrated then the DA started packing
No improvement felt to be needed.
Given detailed, step-by-step written procedures to read (but also had had prior experience) then intensive one-on-one supervision
No improvement needed. Training covered everything.
Limited initial training (8 years ago) – learned by observation. At current site, Webstercare manual followed and initial observation
Could be improved if manual supplemented by video/DVD on how to pack or an occasional short, face-to-face seminar to refresh ideas.
Followed the Webster manual and then applied common sense
Manual and update adequate but also take good ideas from other pharmacies
Observed packing then packed initially under supervision. Was given written procedures and booklet on FRED (dispensing/packing program)
Did Guild technician course. Happy with training but would have liked company training on FRED when new system was introduced.
Described as deep-end, hands on. Observed initially then supervised by an experienced DA (for about 3 weeks) as more details about packs, labelling, etc was slowly introduced
Hands on learning was felt to be good. Was told about the DAA steps but did not see the big picture i.e. awareness of doctor and RCF perspectives.
All respondents felt that they gained skills, knowledge and awareness as they
performed their DAA roles:
Problem recognition and development of strategies to prevent problems (4
responses). One respondent referred to the processes generally while another
mentioned problems related to prescribing rules and regulations (e.g. those for
dangerous drugs). Two respondents mentioned drug-related problems:
Quality Medication Care Group, School of Medicine, University of Queensland36
identifying possible problems while updating the patient profiles (e.g. an RCF drug chart
that had both Naproxen and Celebrex but no ceased stamp on the Celebrex that should
have been ceased). These were communicated to the pharmacist and so prevented a
drug interaction and possibly adverse outcome.
identifying possible areas to rationalise doses (e.g. 1x100mg instead of 2x50mg tablet).
Informal increase in drug knowledge (3 responses) – one technician also attended
pharmacy continuing education (Guild and PSA).
Organisation skills and time management (3 responses).
An appreciation or understanding of other parts of the health care system involved
in DAA provision (3 responses): The nursing home perspective e.g. how medication charts work, problems for nurses
when doctors order things but don’t write them on the medication chart or that the
Douglas packs were not so straightforward for busy nurses.
How the renal unit and indigenous health services work in relation to prescription
management and drug supply and the PBS.
How to deal with indigenous communities.
The importance of/need for constant contact between pharmacy and doctor, and
pharmacy and clients to make the system work.
Community patient needs and benefits from DAAs (2 responses) e.g. how DAAs
improved patients who are disorganised with their medication management, better
monitoring, see fewer different doctors and more holistic medication profile.
Improved attention to detail.
Communication skills (with doctors, nurses, pharmacist and other staff).
If the DAA packing area is near or in the main dispensary there are lots of mistakes.
It is important to be away from the dispensary to decreasing interruptions.
Procedures followed by most were detailed (8 responses) but were flexible and could
be changed to do things differently. Four respondents mentioned discussion and
changes as part of a continuous quality improvement (CQI) process. Examples of CQI
strategies given by two respondents were:
To decrease interruption (including a separate location to dispensing/front of shop).
Changing the person who packs from pharmacist to technician.
Streamlining checking.
Introducing a system to re-order prescriptions from doctors.
Developing procedures about how long to pack in advance in unusual situations
e.g. Christmas.
Two respondents commented on the importance of all people involved being aware of
and following the procedures “packing doesn’t work unless all people involved do the
same thing”. This team approach “ works better and has less stress”. Others overcome
this possible variation by having a single person doing the packing (3 responses) and
one had detailed procedures for pack changes and new packs needed outside the
usual packing cycle when the DA was not on duty. Others (3 responses) had flexibility
to a point (e.g. different procedures for deblistering) but a key sequence of steps or
guideline that had to be followed - “everyone does have a different way to pack but
there are core guidelines for the different pack types that are followed”. It was important
to “find a routine and stick to it”.
One respondent said that there were no written procedures when she first started in the
pharmacy. This caused hold ups as the DA was “forever chasing up my mentor” (an
experienced DA in the pharmacy). Written procedures have since been developed.
Quality Medication Care Group, School of Medicine, University of Queensland37
4.1.3.4 Legislative, standards and guideline framework related to DAAs
There was variation in the level of awareness DAs had of the legislative, standards and
guideline framework they had to work within – one DA did not recognise any
legal/guideline context except the PBS rules whereas others were aware of several but
not all of the other main legislation/guidelines (Table 4.6). Of the 5 main elements of
the legal/guideline framework, awareness of the PSA guidelines was lowest while
awareness of PBS rules was highest, perhaps because the PBS rules created the
greatest number of perceived barriers to an efficient service. From the DA perspective,
the barriers to efficient DAA provision were related to the PBS rules:
Waiting for prescriptions from the doctor (not timely response) to meet PBS rules is
an efficiency barrier (5 responses). Dealing with the rules led to the practice of
“owing” prescriptions as medication were required for packing before the
prescription was provided by the doctor. Owing prescriptions contributed to rework.
The 20 day rule (4 responses). One DA said that planning was needed to reduce
the effect of this rule on efficiency.
Processes related to authority prescriptions (2 responses). One DA encountered
problems requesting authority prescriptions but HIC said the prescription should not
be due (although the pharmacy needed the prescription ahead of time to pack a
DAA). There was no flexibility, the request was rejected and the pharmacy did not
get paid.
Doctors not reducing a verbal prescription into writing within three days.
Rules related to safety net entitlements.
Table 4.6 DA awareness of legislation/guideline framework for DAA provision
Number of respondents Regulation, standard or guideline Aware of
rules Aware with prompting Unaware or not explicitly
mentioned
PBS rules 5 4 aware but did not see PBS rules as “legislation”
Poisons regulations or similar 5 4
PSA DAA guidelines 2 1 6
QCPP standards 4 1 4
APAC aged care standards 1 1 – aware of document but not looked at
5
2= Not applicable
Regulations related to Schedule 8 or dangerous drugs were mentioned as an efficiency
barrier by one respondent. Three DAs did not mention barriers explicitly; they just
accepted working within the rules.
Respondents were asked to reflect on why pharmacies and RCFs may have trouble
complying with the legal/guideline framework and what might improve the situation.
The postulated reasons and possible solutions are summarised in Table 4.7.
Quality Medication Care Group, School of Medicine, University of Queensland38
Table 4.7 Reasons and possible solutions postulated for non-compliance with
legal/guideline framework
Reasons postulated for compliance problems Possible solutions A lack of understanding of the rules (PBS, Poisons etc) by others (doctors, nurses, carers and patients) (3 responses) so, for example, doctors don’t do their part and don’t recognise what the pharmacy needs to operate within the system - “Pharmacies sometimes overlook the guidelines to get medications for patients”.
Increasing doctor awareness of the PBS rules and what is required by the pharmacy (2 responses)
Holding a tripartisan meeting with RCF staff, doctors and pharmacy to discuss issues (1 response).
Problems getting repeat prescriptions from doctors leads to “owing” prescriptions (3 responses)
Institute a “no script, no pack” policy (2 responses). One pharmacy arranged a meeting with local doctors before instituting a “no script, no pack” policy; another attached notes to advise where the patient or RCF why the drug is missing so that patients and RCF pressure doctors
Time pressures (2 responses). To reduce the time taken and to increase efficiency, pharmacies may try and “cut down on legalities to make the process quicker”
That the PBS rules don’t match the practicalities of packing DAAs (2 responses).
Changes to PBS rules as they apply to medications in DAAs (2 responses). A different claim type for DAA-packed medications was suggested as a solution to make PBS rules more practical for DAA packing including exemptions to the 20 day rule and more flexible authority and owing prescription rules. A different type of PBS prescription for medications supplied in a DAA where a medication chart or full regimen written by a doctor is valid as a prescription that can be submitted as a valid PBS claim (1 response)
People not thinking about the situation and solutions (1 response).
Participative problem solving sessions with the people involved can lead to improved procedures as a result (1 response).
Not enough human resources (1 response), particularly in country areas where there are fewer GPs and pharmacists. It can be difficult for all involved to do all things because of the workload in servicing the community.
More human resources needed to follow all the rules and guidelines (1 response).
4.1.4 GP PERSPECTIVES
From the GP perspective the problems/issues arising from DAA provision and use
include:
Pharmacy request scripts too early (contrary to dispensing software alerts) or after
having just written another prescription (legalities) or requests are inaccurate i.e.
medication has been ceased, script is at RCF, patient is deceased.
Pharmacy nags doctors with multiple faxes about prescriptions.
Pharmacy packs too far in advance (5 weeks) consequently more than 1
prescription is required and the prescriptions need to be written on different days.
Doctors would prefer to generate scripts electronically to keep records so some
have a computer in the RCF linked to the GP surgery.
DAAs result in medication wastage.
Quality Medication Care Group, School of Medicine, University of Queensland39
Inability to monitor wastage or compliance based on script rates.
Lack of flexibility and responsiveness to changes (delays in starting and changing
medicines).
Alterations to patients medicines requires detailed communication.
Pharmacy need to check the charts more frequently.
Nurses/RCFs relinquished responsibility for medication management and use of
DAAs deskills nursing staff.
DAAs make more work for Doctors and RNs but makes life easier for pharmacist.
DAAs increase workloads for everyone due to documentation and bureaucracy.
GPs can’t charge a fee for writing prescriptions without a consultation with the
patient so are not reimbursed for their time.
Some of these solutions to issues from the GP perspective were incorporated into the
recommendations for an integrated best practice model for the RCF and community
settings.
4.1.5 OTHER CONSULTATION
As part of consultations with other key informants, problems associated with DAA
services were explored; methods and tools to facilitate best practice implementation
were also investigated.
Consultation with the staff at PSA led to the identification of key documents and key
informants. The approach to best practice was supported by PSA who indicated that
the work would inform the revision of the PSA DAA guidelines and professional
standards. At the suggestion of PSA:
More detailed questions exploring the dispensary assistant experience with training
and skill development were included into this phase of the data collection (see
4.1.3.3).
Additional questions about the existence of Australian standards for the
manufacture of DAAs and the possibility of a United Kingdom requirement for new
drug products to be stability tested in DAAs before registration were asked of TGA.
[At present, there was no Australian TGA standard specifically addressing the
design and manufacture of DAAs nor was TGA aware of any UK proposal for pre-
registration stability testing.]
Key informants and key documents were follow-up as described in 3.1.7.
A key strategy to the implementation of best practice models identified by the expert
panel was inclusion into the Quality Care Pharmacy Program (QCPP). At a meeting at
QCPP, plans for revision of the type and format of QCPP standard were discussed.
The new format was to be:
An expression of the standard.
A list of actions required to meet the standard.
Evidence required for assessment of the performance of actions and compliance
with the standard.
Resources to support the pharmacy in attaining the standard.
It was felt to be desirable that a best practice model for DAAs included information on
what a pharmacist and pharmacy had to do to implement the standard and, that
procedures and templates were included.
Quality Medication Care Group, School of Medicine, University of Queensland40
At the preliminary stage, the individual assessors approached did not offer advice on
problems associated with DAAs nor their solutions. It was felt that assessing elements
of a DAA service was beyond the scope of their assessment process. Further, one
respondent indicated that “provided a facility has a good pharmacy, they will have no
trouble meeting the medication management standards”. In the light of this response,
further input from the assessors’ organisation was deferred until responses on a draft
model would be sought.
Discussions about drug stability and the information held by the TGA were held with Dr
Kelly. He indicated that, in the absence of specific stability studies (not required by
TGA), a risk assessment approach based on the manufacturers’ pack type and storage
and shelf life information, as suggested by the researchers, would be feasible as a way
of minimising drug instability when products were repackaged into DAAs. TGA reviews
the stability of medications in original packs prior to registration but this information was
not released, nor was it all available in a searchable, electronic format, and was not
available for grandfathered drugs (those registered prior to 1989). Dr Kelly indicated
that TGA did not have standards for repackaging but drew attention to other
pharmacopoeial sources (British Pharmacopoeia (BP) and United States
Pharmacopoeia (USP)) of standards. On later review by the researchers, only the USP
had standards relevant to the issue of drug stability after repackaging into DAAs (see
4.3 ). Moisture transmission was identified as a key factor in drug stability so that,
where drugs were package in “tropicalised packs”, this would be a good indicator of
potential instability in a DAA. For example, amoxicillin and clavulanic acid products
were packed in tropicalised packs as the clavulanic acid was degraded quickly with
moisture. Dr Kelly suggested seeking information from the manufacturers of DAAs
about moisture and air transmissibility. As part of a risk minimisation approach to
packing DAAs, Dr Kelly also recommended the Code of Good Manufacturing Practice
(GMP) as model for standards and procedures.
At Dr Kelly’s suggestion, ADRAC was approached about adverse reactions associated
with drugs in a DAA. The Adverse Drug Reactions Unit responded that:
“We have no record of any reports where DAAs have led or contributed to an
adverse drug event.”
In Phase 2 and in the focus groups and surveys undertaken in Phase 3, poor
communication was commonly cited as a barrier to safe and efficient DAA services.
The Aged Care HMR Coordinator of the South East New South Wales (NSW) Division
of General Practice, Julia Clapin, was contacted about her experiences with a project
to communications and systems between GPs, RCFs and pharmacists. She described
a process of initial surveys to identify issues followed by successive meetings in the
various towns in the divisions where the focus was on systems – the difficulties and
how to address them. An external mediator was felt to be valuable at these meetings.
Medication charting and owing prescriptions were common problem areas to be
addressed through this co-operative process but Julia stressed that one solution did
not fit all situations; some needed basic systems to be put into place while others were
more advanced in their problem solving and co-operation. In one town, the pharmacy
had good systems in place that included:
Communication systems.
Colour coding of communications e.g. requests for prescriptions owning were on a
specific colour of paper.
Quality Medication Care Group, School of Medicine, University of Queensland41
Reminder systems, for example, for when new prescriptions would be required.
In another RCF where residents visited the GP in the GP’s rooms, the solution to poor
charting of medications was to send the RCF chart with the resident so that it was on
hand at the time of consultation.
A number of GP divisions were conducting aged care projects related to computerised
medication charts and dispensing from medication charts, rather than the present
systems where a GP had to write both a medication chart for administration and a PBS
prescription for dispensing. Strategies included:
GPs generating medication charts from their own prescribing software, for use in
RCFs, rather than hand writing charts provided by facilities.
To do this, GPs would keep the primary electronic record in the facility or, there
would be a remote link to the GPs surgery computer so that when the GP was in
the RCF, changes could be made to the medication chart and a new one printed in
the facility (or in the surgery and sent to the facility).
In all of these projects, there was considerable individualisation of any intervention or
strategy to that it suited the circumstances of a given RCF and its related practitioners.
Julia felt that the best practice solution proposed by the researchers of defining
obligations and expectations, and promoting mutual awareness, would be beneficial in
addressing some of the problems encountered in her Division. The approach of
defining issues but not prescribing solutions allowed flexibility in implementation.
The problems and solutions associated with continuity of care was the focus of
discussions with the Queensland Health Safe Medication Practice Unit. The issues
identified in the hospital pharmacy focus groups were reiterated in these discussions
(e.g. concerns over the integrity of medications in packs). The problems discussed
applied to any patients admitted to or discharged from hospital and occasionally
specifically address the added complications of medications packed into a DAA.
Variation in the practices of RCFs and pharmacies related to how the DAAs were used
and the nature of a given DAA service increased the complexity. Concordance
checking and the taking of drug histories on admission was complicated by
medications packed in DAAs including how the DAA was labelled, what was packed
and whether the pack was sent to the hospital on admission. These aspects were to be
addressed by a reconciliation process for all admissions. A discharge process was also
in development to ensure timely medication information transfer and “adequate”
medication supply. The issues raised by community pharmacists and dispensary
technicians that reflected the variations in practices e.g.:
what level of documentation/veracity of the discharge medication regimen was
acceptable to a given community pharmacy,
of fitting in to the DAA packing cycle,
just which medicines needed to be supplied for an individual patient, and,
what “generic” brands were provided,
had not been addressed to date.
The unit was taking an enterprise wide approach to developing solutions to problems
related to the continuity of medication information and supply through the development
of tools and protocols. Improving communications with RCFs, GPs and community
pharmacists was felt to be critical. One strategy identified was a set of templates that
allowed hospital pharmacies to generate communications lists. After reviewing an early
Quality Medication Care Group, School of Medicine, University of Queensland42
draft of the best practice models, the following ideas were offered to improve problems
associated with the hospital admission and discharge for people using DAAs:
The GP admission letter could be used to provide a minimum dataset to the
hospital when a patient was admitted.
Hospital pharmacies could ask local RCFs and their supplying pharmacies about
their preferences.
The staff of the Safe Medication Practice Unit felt that the flow diagrams for the whole
DAA service (including interactions with hospitals) and the approaches to improving
communication and mutual awareness raising would facilitate improved practice.
A director of the HDG Consulting Group was consulted over the issue of potential
overlap between the best practice models and associated tools with those developed
as part of a Victorian project to address implementation of APAC guidelines for
medication management in residential aged care facilities (Australian Pharmaceutical
Advisory Council 2002). As part of the project, facilities were surveyed about their
current practices and those areas of the guidelines where implementation was most
troublesome. DAAs were the subject of Recommendation 9. Problems related to what
type of DAA was used as this would affect who could administer medications and
administration procedures. Labelling of DAAs could also contribute to non-compliance
with the guidelines. The ordering and supply of all medications using the approved
generic name (rather than a “generic” or alternative brand) was suggested as a
solution. The draft resource kit developed in the project included:
Competency assessment tools for Personal Care Assistants (PCAs) assisting with
medications. This checklist included the basic level of skill and practical aspects
such as washing hands with soap before administering medications.
Resident self-medicating assessment tools.
Draft procedures for policies such as administration of medications when a resident
is out for the day. This had additional ramifications where medications, often
enough for one week, are in a DAA.
Sample letters to GPs about the Medication Advisory committee.
A handouts or information sheet for residents and family explaining the medication
management systems. This did not their role and responsibilities as suggested in
the best practice model in this study. This aspect was supported by Dr Saxon.
As the tool developed for DAAs was a yes-no checklist to say whether the aspects of
Recommendation 9 were addressed, there was felt to be little overlap with the best
practice models but that the two projects would be synergistic.
4.2 LEGAL PERSPECTIVE ON BEST PRACTICE
The report received on how best practice can impact on liability indicated that a number
of strategies proposed as part of the best practice models were important for error
prevention and that the documentation suggested in the model was important as
liability risk reduction. The full report is in Appendix D but the key points were as
follows:
The resource (cost and time) pressures placed on the pharmacist in providing a
DAA do not diminish in any way the professional obligations of the pharmacist in
relation to the dispensing of medication and that appropriate procedures and written
protocols for dispensing that meet professional standards must be observed.
That staff involved in packing DAAs are appropriately trained and supervised by the
pharmacist who is ultimately responsible for the service, and that there are policies,
Quality Medication Care Group, School of Medicine, University of Queensland43
procedures and records of staff DAA training and supervision and for other aspects
of dispensing medicines and packing and checking DAAs (e.g. how a pharmacist
would check a pack). The dissemination and regular communication of these
policies and procedures should also be recorded (e.g. at induction or through
minutes of staff meetings).
To ensure the integrity of packing, there must be appropriate systems and
procedures in place including the checking of medication charts, the reviewing of
original packaging of medications and the keeping of appropriate records that
shows these procedures were followed.
Providing medicines in a DAA does not remove the responsibility to provide
counselling to patients who live in the community and to residents within the
residential care facility and their carers to ensure that medication regimes are fully
understood so as to be safe and effective. Counselling of patients discharged from
an RCF (where medicines had been packed) to go to their home with medicines in
original packs was particularly emphasised.
Recording those medicines not packed in a DAA and a record of counselling
provided for those medicines was noted.
The role of a clear agreement between the RCF and pharmacy that set out
responsibilities and obligations was seen as a valuable way of minimising errors,
such as those arising out of poor communication of medication changes. A well set
out agreement would not to create additional burdens on parties but to reinforce
separate obligations of the RCF and pharmacy related to DAAs.
Pharmacists also have an obligation to RCF staff. An RCF has a separate
responsibility to ensure that there are appropriate systems in place to ensure
appropriate medication administration but pharmacists providing DAAs in this
setting have a separate obligation to ensure that there is effective communication
with appropriate RCF staff. This obligation includes facility access to accurate and
up-to-date information about medication. The recording of the provision of such
information was emphasised as was the retention of these records. Pharmacy
procedures related to, and the maintenance of records of all communications with
an RCF were seen as a useful risk management tool. Similar record keeping by
RCFs was also a risk management tool.
Errors arising from poor communication related to medication changes for
community dwelling DAA users could be addressed by an agreement between
involved parties (including, as needed, carers/close relatives) provided that the
agreement was clearly understood by the patient who did not feel pressured to
agree. It was felt to be important to include how medication changes were to be
communicated in any written DAA service agreement between a doctor and a
pharmacist.
Clear and unequivocal directions to the pharmacist about medication changes were
felt to minimise error but the pharmacist must still give careful regard to changes
and verify with the prescriber as needed. Pharmacy systems (procedures and
records) to deal with ceased or changed medication disposal or separate storage
were discussed. For community patients, pharmacy procedures related to, and the
maintenance of records of communication of medication changes (e.g. by
annotation of dispensing records) was seen as a risk management tool. A central
storage system of all data in relation to a patient was encouraged.
The re-use of previously dispensed medicines (dispensed for one patient then
subsequently re-dispensed for another) was rejected and discussed at length.
Quality Medication Care Group, School of Medicine, University of Queensland44
Documented pharmacy procedures for the disposal of returned medicines were
seen as critical.
A pharmacy system to regularly ensure the pharmacy-held profile is accurate would
minimise the possibility of errors related to medication changes.
A number of risk management strategies were discussed related to packing errors.
These include appropriately trained and supervised staff if DAAs were not packed
by a pharmacist, and documented policies and procedures that include checking
and supervision (including evidence that the practices were followed e.g. signed
record that a pack was checked by a pharmacist). This was added to policies,
procedures and records of checking and supervision in the dispensing step. The
packing of a medicine for daily administration that was intended for weekly
administration was seen as a not infrequent error that needed to be addressed by
pharmacy systems, as the consequence was potentially fatal. RCFs needed
appropriate systems in place to prevent any packing error being compounded by a
Registered Nurse.
Three less typical examples of legal issues arising from the supply of medications
packed in DAAs were described, in particular, one case concerning recycling of
returned medicines.
4.3 DRUG STABILITY IN DAAS AND GOOD PACKING
PRACTICE
This report was initially prepared by Dr Alison Haywood (School of Pharmacy, Griffith
University) based on conceptualisation and contributions from Dr Julie Stokes (School
of Medicine, University of Queensland). Professor Michael Roberts (School of
Medicine, University of Queensland) contributed to the initial conceptualisation of the
utilisation of available stability information. Professor Beverly Glass (School of
Pharmacy and Molecular Sciences, James Cook University) and Associate Professor
Andrew McLachlan (Faculty of Pharmacy, University of Sydney) provided advice and
reviewed the various drafts of the report. This section was revised by Julie Stokes to
add a practice slant to theory. Given the limited information about the stability of
medicines in DAAs, the approach taken was to examine the theoretical factors affecting
drugs stability and identify methods to apply the theoretical concepts to current DAA
practices. Avenues to obtain more evidence on drug stability in DAAs are also
addressed.
4.3.1 INTRODUCTION
Drug stability is a rather broad term that encompasses chemical1, physical2,
microbiological3, therapeutic4 and toxicological5 stability not only of the drug substance,
but when taking account of the excipients, also the drug product (Allen et al. 2005; USP
<1191> 2005). Stability has been defined as the extent to which a product retains,
1 Each active ingredient retains its chemical integrity and labelled potency, within specified
limits.2 The original physical properties, including appearance, palatability, uniformity, dissolution and
suspendability, are retained. 3 Sterility or resistance to microbial growth is retained according to the specified requirements.
Antimicrobial agents that are present retain effectiveness within the specified limits. 4 The therapeutic effect remains unchanged.
5 No significant increase in toxicity occurs.
Quality Medication Care Group, School of Medicine, University of Queensland45
within specified limits and throughout its period of storage and use (i.e. its shelf-life),
the same properties and characteristics that it possessed at the time of its manufacture
(USP <1191> 2005). A similar requirement applies to prescription medicines registered
with the Australian Therapeutic Goods Administration (Therapeutic Goods
Administration 2004) (see http://www.tga.gov.au/pmeds /argpmap14.pdf). Stability of
manufactured dosage forms is routinely confirmed by the manufacturer, where stability
studies on active substances and packaged dosage forms are conducted by means of
“real-time”, long-term tests and accelerated stability tests at specific temperatures and
relative humidity that represent storage conditions experienced in the distribution chain
of the climatic zone(s) of the country or region of the world concerned (Aulton 2002;
Therapeutic Goods Administration 2004; USP <1150> 2005). Stability testing is the
primary tool used to assess expiration dating and storage conditions for pharmaceutical
products.
Although stability of a dosage form is often seen to be the responsibility of the
manufacturer, this does not include removal from the original packaging and
repackaging, which is often discouraged by the manufacturer (Church et al. 2006;
Walker 1992). According to the terms of each product licence held by a manufacturing
company, transfer to any type of compliance device/Dose Administration Aid (DAA) can
not be recommended without an assessment of the stability of the product in the device
in question being carried out, and the product licence adjusted accordingly (Church &
Smith 2006; Walker 1992). In electing to repackage a drug product into a DAA,
pharmacists must therefore consider the implications of the transfer to a non-
manufacturer’s pack on drug stability.
Despite the widespread use of Dose Administration Aids (DAAs) (medication
compliance devices/unit-dose containers), there is little available data regarding the
stability of the drug products during packaging or storage in these devices or even
whether a stability problem does or does not exist (Walker 1992). Pharmacists have
limited references on the stability of drug products during repackaging or storage in
DAAs. Additionally, there is no information on the stability of medicinal products when
placed in close physical contact with other medications. Limited studies have been
performed to provide specific evidence on the stability of drug products in DAAs, and
these studies do not reflect the complexities relating to the interaction of active
ingredient(s), excipients, specific dosage form/ drug delivery technologies (including
techniques used in the manufacture, such as coating), packaging materials and pack
types, or the environmental conditions at the time of repackaging or on storage.
In short, there is no simple way of determining whether a drug product is suitable to
pack into a given DAA. Pharmacists, therefore, must make an informed judgement as
to the suitability of a given drug product for inclusion in a DAA for provision to a patient.
This report provides a framework for pharmacists to consider the stability implications
of repackaging drug products into DAAs by:
Identifying stability ‘risk management’ issues and providing recommendations for
gathering further evidence to improve the quality of decisions pharmacists make
about repackaging of drug products into DAAs.
Providing a ‘risk assessment’ approach for using the currently available stability
information for a specific drug substance /product.
Quality Medication Care Group, School of Medicine, University of Queensland46
4.3.2 PHYSICOCHEMICAL STABILITY IMPLICATIONS RELATING TO SOLID DOSAGE FORMS
Drugs degrade to different extents on separate exposure to heat, moisture, air
(including oxidation) and light. Combinations of these stresses can cause complex
behaviour (Florence et al. 1998; Glass et al. 2004). In solid state formulations, the
presence of excipients further complicates the matter because the excipients may
increase, have no effect on, or decrease the inherent stabilities of the drug. Any
degradation will usually adversely affect the therapeutic activity of the drug substance
(Glass et al. 2004).
Many drugs are susceptible to some form of chemical decomposition when formulated
in either liquid or solid dosage forms. Such degradation not only leads to a loss of
potency of the drug substance but may, in some cases, cause changes in physical
appearance of the dosage forms, such as discolouration following, for example, the
photochemical decomposition of the drug substance (Aulton 2002; Florence & Attwood
1998; USP <1191> 2005).
In addition to the potential chemical decomposition (by possible hydrolysis, oxidation,
isomerisation, polymerisation, photochemical degradation) of the drug(s) and/or
excipient(s), physical changes to the solid dosage form, for example, changes in tablet
hardness, friability, disintegration and/or the dissolution rate may lead to both altered
physical appearance and/or bioavailability of the drug substance.
Manufacturers’ packaging is designed to protect products from environmental factors
encountered during storage, including light, air (oxygen, carbon dioxide and other
gases), and moisture whilst ensuring limited interactions between the product and the
packaging material. The dosage form itself may have design features to enhance the
physicochemical stability of the product, for example, titanium dioxide may be added to
the coating of tablets or the gelatin shell of soft gels to protect light-sensitive active
ingredient(s) and/or excipient(s).
Finally, inappropriate storage, transport and in-use conditions (especially exposure of
the product to changes in temperature) may enhance physicochemical instabilities and
degradation rates of the active ingredient(s) and/or excipient(s) or result in an increase
sublingual tablets and wafers (Corlett 1996; Pharmaceutical Society of Australia 2004;
Walker 1992).
Certain medications are not suitable for dispensing in some DAAs for practical reasons.
Large solid dosage forms (e.g. Gaviscon® tablets), or multiple tablets to be taken at a
single time, may not fit into the individual compartments of the DAA. “As required/ when
necessary” (PRN) medication, if placed in a DAA with other medicines, may be taken
unnecessarily on a regular basis (Corlett 1996). Additionally, drug products that look
alike should not be dispensed in the same DAA (Corlett 1996).
Splitting tablets to obtain the required dose further exposes the solid dosage form to
environmental factors (particularly when the tablet has a protective coating), which may
lead to stability problems, especially if the remaining tablet halves are stored for an
extended period in unsealed blisters/ zip-lock bags/ plastic labelled bottles/ bulk
canisters or cassettes in automated systems.
4.3.5.3 Moisture-sensitive solid dosage forms
The stability of solid dosage forms may be affected by the moisture content of the
atmosphere in the container in which they are stored (Allen et al. 2005; Chen et al.
2003; Florence & Attwood 1998). Moisture is considered to be an important factor that
must be controlled in order to minimise decomposition during storage (Allen et al. 2005;
Aulton 2002; Badawy et al. 2001; Florence & Attwood 1998). In addition to the dosage
forms specifically designed to rapidly disintegrate when exposed to moisture, other
drug products can be adversely affected by moisture.
The original manufacturer’s package design is particularly critical for a moisture
sensitive product, since it must ensure that the product is adequately protected from
moisture during its shelf-life (Badawy et al. 2001). Desiccants (e.g. small packets of
silica gel) are frequently included in the packaging configuration of these products in
order to maintain low relative humidity inside the package and hence protect the
product from moisture (Allen et al. 2005; Badawy et al. 2001).
Drug products packed with a desiccant or packed in ‘tropicalised’ packs (double-sided
foil packs that keep light, air and moisture out) are a warning of potential instability due
Quality Medication Care Group, School of Medicine, University of Queensland53
to moisture. Careful consideration of DAA type, packing practices (including any
intermediate steps) and handling by the RCF or patient should be given in making a
decision to pack such products in a DAA. A recent example of moisture related
instability relates to Adalat Oros ® where the product information states “The drug
release mechanism of Adalat Oros is triggered by moisture. Contact of the tablets with
moisture may not be apparent but loss of contents may have already occurred. To
prevent this, the tablet must be kept in its original blister foil packaging until
immediately before use” (TGA approved product information amended 20 July 2004).
The extent to which the moisture within a DAA triggers nifedipine (active ingredient of
Adalat Oros ®) release (a light sensitive drug, see 4.3.5.5) is unclear (see 10.2.2.1 for
other examples). One strategy suggested to minimise this instability risk was packing
the medicine still in its individual foil wrapper into the DAA blister and “instructing the
carers to remove the foil before administration” (per Auspharmlist.net.au posted
17/10/2005 and a practice described by a different pharmacist on 28/10/2005). The
trade off is the increased complication of medication administration for the carer or
patient and the risk that the patient may swallow the medication still in the foil.
Swallowing a medicine still in its blister or foil can lead to intestinal perforation and has
been fatal (Blister-strip warning. 1996).
Temperature cycling can also lead to increased condensation in DAAs with suboptimal
seals/closures, resulting in increased physicochemical stability issues and also an
increased potential for microbial contamination. In use, DAAs may be subjected to a
reasonable degree of handling, during which accidental rupture or opening of nearby
compartments, blister seals or sachets may occur (Saville 2001), allowing further
exposure to humidity.
In addition to causing chemical degradation in susceptible active ingredients or
excipients, moisture can cause tablets to harden (or soften) with a subsequent effect
on disintegration and dissolution behaviour (Al-Zein et al. 1999).
It is essential that the packaging material and closure of the DAA offers sufficient
moisture protection for moisture-sensitive solid dosage forms. The pharmacist should,
at the very least, ensure the integrity of the DAA before it leaves the pharmacy and
counsel the patient not only on storage conditions, but also to be vigilant and monitor
that the integrity of the DAA is maintained throughout the dosage period. One option is
to reduce the interval between placing a moisture sensitive medicine in a DAA and the
time when the last dose is used in that DAA.
Quality Medication Care Group, School of Medicine, University of Queensland54
Package Insert/ CMI
OH&S issue?
Protect from light –
is DAA light-proof?
Protect from moisture –
is DAA moisture-
proof?
Transfer to DAA NOT
recommended
Problems not
expected
Unsuitabledosage
form?
* E.g. Cytotoxics
* E.g. Sublingual/
buccal/ dispersible/
effervescent tablets
YES
Are there known
interactions with
packaging/ other drugs?
YES
YES
YES
Protect from heat –
Heat sealing process?
Protect from air –
is DAA airtight?NO
NO
NO
NONONO
YESYES
NO
NONONO
NONONO
YESYES
YESYES
Figure 4.4 Decision-tree based on currently available information for repackaging medications in DAAs
Quality Medication Care Group, School of Medicine, University of Queensland55
4.3.5.4 Solid dosage forms sensitive to air (oxidation)
It is essential that medications required to be protected from air (especially oxygen) are
stored in an airtight container (Allen et al. 2005; Florence & Attwood 1998; USP
<1191> 2005) and that exposure to air during packing be minimised. A suboptimal
seal/closure on the formed container (DAA) will decrease the protective properties of
the container – insufficient temperature, time or pressure during a heat or cold seal
operation may enable the passage of moisture or oxygen through the seal area over
time (USP <1146> 2005) in addition to any air or moisture that permeates the
packaging materials.
Again, the pharmacist should, at the very least, ensure the integrity of the DAA prior to
leaving the pharmacy and also counsel the patient on both the timely removal of drug
products from the DAA and the importance of monitoring the integrity of the DAA.
4.3.5.5 Light-sensitive solid dosage forms
Although many drugs are found to decompose on exposure to light, the practical
consequences may not necessarily be the same for all these compounds (Tønnesen
2001). Some drugs will decompose by only a small percentage after several weeks of
exposure, while other substances, such as derivatives of the drug nifedipine, have a
photochemical half-life of only a few minutes (Tønnesen 2001).
Light-sensitive drugs can be affected either by sunlight (especially the UV component)
or artificial light sources (e.g. fluorescent light especially the UV component). This may
not only lead to photodegradation of the drug substance (with a subsequent loss of
potency or therapeutic activity), but also to a change in the physicochemical properties
of the product (e.g. discolouration/ changes in appearance, and dissolution rate)
(Tønnesen 2001).
The method most commonly used to protect photosensitive drugs is to place the
product in suitably protective packaging (e.g. amber or coloured non-transparent
containers). Some tablets have been photostabilised by the addition of light-absorbing
compounds (including scavengers and quenchers) and/or film coating with opaque
films (e.g. yellow iron oxide and titanium dioxide) (Glass et al. 2004; Tønnesen 2001).
As oxidation may be catalysed by exposure to light, some solid dosage forms with
susceptible ingredients may be stabilised (to a limited extent) by the addition of
antioxidants (i.e. agents that are more readily oxidised than the drug/excipient) (Aulton
2002).
It is essential that the packaging material and seal/closure of the DAA offer sufficient
light protection for light-sensitive solid dosage forms. The USP (USP <1146> 2005)
advises that if light protection is required for a drug product, the repackager (see
definitions in Appendix E) should follow the requirements for light transmission (USP
<661> 2005) and that this testing should be conducted on the formed container (i.e.
DAA), because the light protective properties of the film are compromised once the film
is thinned during the forming process (USP <1146> 2005). While this USP standard
does not apply to pharmacists repackaging medicines for individual patients, it gives an
indication as to the rigours required to optimise the stability of a drug product.
The pharmacist should ensure that DAAs containing light sensitive drug products are
prepared in a way to minimise the exposure to light and that they are stored in a dark
Quality Medication Care Group, School of Medicine, University of Queensland56
environment. This includes storage in the pharmacy, any RCF or the patient’s home.
An additional strategy is to consider an ‘amber’ DAA or if using a light-protective ‘cover’
or a ‘sleeve’, counsel the patient not only to avoid excessive exposure to light of the
DAA on storage, but also to maintain storage in this ‘cover’ or ‘sleeve’.
4.3.5.6 Solid dosage forms sensitive to heat
The stability parameters of a drug product are known to be influenced by environmental
conditions of storage including exposure to adverse temperatures (Allen et al. 2005;
Aulton 2002; USP <1150> 2005; USP <1191> 2005). In general, the rate of a chemical
reaction increases exponentially for each increase in temperature (Aulton 2002; USP
<1191> 2005). Exposure to increased temperatures may therefore enhance
physicochemical instabilities and degradation rates of the active ingredient(s) or
excipient(s), or result in an increase in potentially toxic degradation products (Aulton
2002; Florence & Attwood 1998; USP <1191> 2005). Thus the likely storage and
transport conditions of medicines in a DAA may be factors in deciding whether or not to
repackage a drug product into a DAA.
Heat is also a consideration where the DAA has a heat sealing process. Critical sealing
parameters include pressure and temperature control, since any undesirable variation
in these parameters may yield inadequate seals (USP <1146> 2005) thus contributing
to instability associated with moisture or air ingress into the DAA.
The pharmacist needs to take note of the above stability considerations regarding
exposure of medications to heat during DAA preparation, storage (in the pharmacy and
in RCFs) and during delivery. Patients should be advised on storage of DAAs in their
homes in order to avoid excessive exposure to high temperatures. An appropriate DAA
pack type/configuration (i.e. cold adhesive seal as opposed to heat seal) should be
selected for drug products sensitive to heat, in the absence of other stability
considerations.
4.3.5.7 United Kingdom manufacturers’ position on repackaging
In a UK survey conducted in two rounds (late 2002 and September 2004),
manufacturers were asked about the suitability of their products for repackaging into
DAAs (Church & Smith 2006). Fifty medical information departments responded about
392 products (note that there are very many more products available in the UK).
Responses were assigned to 6 stability codes. The majority of products were assigned
the code to indicate that there was no stability data so repackaging into a DAA could
not be recommended (see Table 4.9). This approach is cautious and may be
unnecessarily restrictive. No product was listed where the manufacturer held stability
data which indicate suitability for a DAA.
In using the list of products published by Church and Smith several issues need to be
considered by Australian pharmacists:
The likely storage conditions that manufacturers needed to consider were ambient
temperatures and humidity in the UK.
A number of products are not used in Australia. Even if the active ingredient and
the drug company are the same in the UK and Australia, the product name can be
different so that pharmacists can only assume stability code applies to the same
product formulation as available in Australia.
Quality Medication Care Group, School of Medicine, University of Queensland57
A number of products assigned to Stability Category 1 are those that would not be
packed for OH&S reasons or because the dose form is unsuitable (e.g. dispersible)
On review of products with codes other than 2, several were excluded from packing
because of occupational health and safety (OH&S) issues for the packers and some
because the dose form was unsuitable. A number of others were drugs or dose forms
that are currently not available in Australia or would not usually be packed into a DAA.
Thus additional stability information that might be of assistance to Australian
pharmacists was available for 84 items (Table 4.9). These items are listed in Table
4.10. For Code 3, a risk assessment approach could be taken by pharmacists in
choosing to pack these items.
Table 4.9 Frequency of stability codes in from a UK survey of manufacturers (Church
& Smith 2006)
Number of products Stability code
Overall Exclude unhelpful items*
1. Do not put into a DAA 62 26
2. Not recommended as no stability data is available. 27 -
3. Not recommended as no stability data available but
company indicated reasons for specific concerns.
3 11
4. No stability data but it is “probably” suitable to pack.66 36
5. Stability data available in a container other than
original packaging, but not necessarily in a DAA.
4 11
6. Stability data in a DAA shows product suitable for
repackaging. *Items excluded for the following reasons: OH&S, inappropriate dose forms, items definitely not available
in Australia, item an antiinfective or prn drugs (e.g. vardenifil for erectile dysfunction) and not likely to be in
a DAA.
Quality Medication Care Group, School of Medicine, University of Queensland58
Table 4.10 Products where UK survey produced stability information potentially useful to Australian pharmacists
Code Drug and Brand Manufacturer Explanation
4 Acarbose Glucobay Bayer Should be stable for 7 days
5 Alendronate 5mg Fosamax MSD Stable for 3 months at 40C and 75% relative humidity
4 Anastrazole Arimidex AstraZeneca Probably ok for short term storage in a compliance aid
1 Aripiprazole Abilify BSM Hygroscopic
3 Atorvastatin Lipitor Pfizer Disintegrates in bright sunshine
1 Baclofen Lioresal CephalonUK Protect from moisture
4 Bisoprolol - IVAX Probably stable for 7 days
4 Cabergoline Cabaser Pfizer Probably stable for 7 days
1 Cabergoline Dostinex Pfizer Should be stored with desiccant
1 Carbamazepine Tegretol CephalonUK Susceptible to moisture
1 Carbamazepine Tegretol Retard CephalonUK Susceptible to moisture
3 Carvedilol Eucardic Roche Protect from light
3 Chlorpromazine Largactil Hawgreen Can cause contact dermatitis when handled (wear gloves); must be protected from light
1 Ciclosporin Neoral Novartis Ethanol vaporises out of the capsule when out of the original packaging
4 Citalopram Cipramil Lundbeck Probably stable for 6 months
4 Clobazam Frisium SanofiAventis Probably stable for 4 weeks
3 Clonazepam Rivotril Roche Store in dark
4 Co-beneldopa Madopar Roche Probably stable for 14 days
1 Co-careldopa Sinemet BMS Powder is hygroscopic
1 Co-careldopa Sinemet CR BMS Powder is hygroscopic
1 Co-careldopa Sinemet 110/250 BMS Blue dye can fade on light exposure; moisture will cause levodopa to turn black on prolonged exposure
4 Desloratidine Neoclarityn ScheringPlough Unlikely to be hygroscopic
4 Dexamethasone - Organon Unlikely to be any issues
1 Diclofenac +misoprostol Arthrotec Pfizer Misoprostol is extremely moisture sensitive and may degrade
1 Didronel PMO - P&G See Calcium carbonate and etidronate
4 Digoxin Lanoxin GSK Probably stable for 14 days
5 Dipyridamole MR Persantin Retard BI Moisture sensitive; stable for 30 days out of the original container, eg, plastic dispensingbottle
Quality Medication Care Group, School of Medicine, University of Queensland59
Code Drug and Brand Manufacturer Explanation
5 Dipyridamole MR/ Asasantin Retard BI Stable for 30 days out of the original container, eg, plastic aspirin dispensing bottle
3 Dipyridamole Persantin BI Protect from light
4 Donepezil Aricept Pfizer Probably stable for 14 days
4 Dosulepin (dothiepin) tablets Generics(UK) Probably stable for 7 days
4 Dosulepin (dothiepin) capsules Generics(UK) Probably stable for 7 days
1 Enalapril Innovace MSD May hydrolyse at high temperature and in the presence of moisture; 10% loss of potency occurred when exposed to 40C and 75% humidity for 13 weeks
4 Escitalopram Cipralex Lundbeck Probably stable for 6 months
5 Esomeprazole Nexium AstraZeneca Stable for 6 months at 25C and relative humidity of 60%
4 Ferrous sulphate - Alpharma Probably stable for maximum of 14 days (could taint other tablets)
4 Fluoxetine Prozac Lilly Probably stable for up to 4 weeks
4 Gliclazide Diamicron Servier Probably stable for 8 days
4 Gliclazide Diamicron MR Servier Probably stable for 8 days
4 Haloperidol - IVAX Probably stable for 7 days
4 Indapamide Natrilix Servier Probably stable for 8 days
4 Indapamide Natrilix SR Servier Probably stable for 8 days
4 Isosorbide mononitrate - Alpharma Probably stable for up to 14 days
4 Isosorbide mononitrate - IVAX Probably stable for 7 days
5 Isosorbide mononitrate Imdur AstraZeneca Stable for one month
4 Isosorbide mononitrate Monomax SR Trinity Probably stable for 4 weeks
3 Ketoprofen Orudis Hawgreen Must be protected from light
3 Ketoprofen MR Oruvail Hawgreen Must be protected from light
4 Lansoprazole Zoton Wyeth Probably stable for up to 4 weeks in a compliance aid which offers a barrier against moisture*
1 Loperamide Imodium Janssen-Cilag Moisture sensitive and could change colour
1 Misoprostol Cytotec Pfizer Moisture sensitive and may degrade
5 Nicorandil Ikorel SanofiAventis Stable for at least 7 days in a dry environment
1 Nifedipine Adalat LA Bayer Very light sensitive and will significantly degrade very quickly; moisture can affect release mechanism
Quality Medication Care Group, School of Medicine, University of Queensland60
Code Drug and Brand Manufacturer Explanation
4 Nizatidine Axid Lilly Probably stable for up to 4 weeks
4 Olanzapine Zyprexa Lilly Probably stable for up to 4 weeks; sensitive to light; wear gloves if breaking or dividing the tablets due to potential contact dermatitis
1 Omeprazole capsules - Generics(UK) Hygroscopic
5 Omeprazole capsules Losec AstraZeneca Stable for 14 days at room temperature 25–30C) and relative humidity up to 75%
1 Oxybutynin XL Lyrinel XL Janssen-Cilag Hygroscopic; packaged in high density polyethylene bottles with a dessicant
4 Paracetamol - Alpharma Probably stable for up to 28 days
1 Pergolide Celance Lilly Unstable — light sensitive and extremely hygroscopic (notably the 50 g strength)
4 Perindopril Coversyl Servier Probably stable for 8 days
4 Perindopril/indapamide Coversyl Plus Servier Probably stable for 8 days
4 Potassium chloride Slow K Alliance May absorb water; probably stable in an airtight compliance aid for 14 days
1 Pyridostigmine Mestinon Valeant Store in a brown bottle with a desiccant
4 Quetiapine Seroquel AstraZeneca Probably stable for up to 7 days
4 Quinine sulphate - Alpharma Probably stable for up to 7 days
1 Rabeprazole Pariet Janssen-Cilag Hygroscopic
1 Ranitidine - Generics(UK) Hygroscopic and may turn brown
1 Ranitidine Zantac GSK Hygroscopic and degrades in the presence of water
5 Ropinirole Requip GSK Stable for up to 28 days below 25C and at 60% relative humidity
1 Selegiline HCl Zelapar ZeneusPharma Hygroscopic
3 Spironolactone - IVAX Protect from light
3 Tamoxifen Nolvadex AstraZeneca Known to be light labile
3 Tamoxifen - Generics(UK) Sensitive to light
1 Topiramate Topamax Janssen-Cilag Hygroscopic; hydroxylates and becomes unstable with water
5 Zafirlukast Accolate AstraZeneca Stable for 30 days at 60-80% relative humidity
3 Zopiclone - IVAX Protect from light
4 Zuclopenthixol Clopixol Lundbeck 2mg stable for 14 days; other strengths stable for 4 weeks Manufacturer key: BI Boehringer Ingelheim; BSM Bristol Myers Squibb; GSK GlaxoSmithKline; MSD Merck Sharpe & Dohme; P&G Proctor & Gamble; SS Sanofi Synthelabo
Quality Medication Care Group, School of Medicine, University of Queensland61
4.4 SYNTHESIS OF PROBLEMS AND POSSIBLE SOLUTIONS
From the preliminary investigations, a number of problems and possible solutions
emerged, reflecting the perspectives of the various stakeholders. For some problems,
having only one perspective limited solutions arising from the various stakeholders, so
that other possible macro or “big picture” solutions were devised by the researchers to
address some problems. In this section, problems and possible solutions that formed
the basis of best practice models are summarized (see Table 4.11 to Table 4.15).
Table 4.11 Problems from the RCF (management and staff) perspective and possible
solutions (soln)
RCF (management and staff) perspective Issue Accuracy of packs (packing from previous version of chart due to packing in advance or
poor information flow, discrepancies in brand of drug on chart versus brand packed, omission of medications, use of multi-dose packs and tablets that a difficult to identify, prn medication directions, changes )
Soln Improve systems for information flow to pharmacy through standard templates including minimum data sets and IT solutions
Utilise information and communications technology solutions wherever possible, that are not reliant on significant personnel or maintenance to support
RCF/GP awareness of packing cycles Quality assurance/ monitoring procedures introduced in pharmacy RCF responsibility to check packs if administering done by ENs or PCAs and for self-
medicators Review of charting systems to eliminate ambiguous orders and ensure that charts are
up to date. Issue Doctors having to write-up prescriptions when they have already written the medication
order on the patient’s chart. Soln Developing a system and procedures where the medication chart acts as prescription
for pharmacist payment via HIC/Medicare Australia. i.e. modify the supply claim mechanism
Issue Labelling of DAAs Soln DAA should have patient’s photo on it
Packs labelled with both generic and prescribed name of medicine Information on suitability for crushing required on pack Patient or drug information should not be obscured in anyway
Issue Pharmacy responsiveness to change (what do RCF staff do when doctors orders do not match available pack )
Soln Disseminate between stakeholders information affecting the packing cycle e.g when is packing done, when staff are available in pharmacy and RCF to deal with changes, usual doctors rounds when more likely to make changes.
Procedures i.e. details of how to identify tablet to be removed and when to expect changes to occur
Issue Staff issues – use of ENs & PCAs and agency (problem recognition & legality actions) Soln RCF responsibility to check packs if administering done by ENs or PCAs and for self-
medicators. Additional checking step where RN checks pack to ensure accuracy before other staff administer from pack. Facilities need to take responsibility and record all pack errors to be fed back to pharmacy – awareness of problems should help overcome them
Agency guidelines/training (currently done but need details of what to do when things go wrong ie dropped tablets)
Standardise charts to reduce variation for new staff/agency staff Issue Hospitalisation of residents and transfer of information about medication changes and
administration instructions. Soln Model/guidelines for transition between RCF and hospitals
Quality Medication Care Group, School of Medicine, University of Queensland62
Table 4.11 continued RCF (management and staff) perspective Issue Need for written orders to administer medication (need to be accurate, timely and
responsive to changes) Soln Hospital discharge summary until new profile available – stop gap measures.
Procedures for dealing with changes to regimen after hours or when no RN available Promotion/ communication between RCF staff and doctors regarding staffing/legal &
other practical issues that may affect the facilities ability to deal with medication changes or particular doses
Issue Need/have existing good procedures/systems for monitoring and addressing problems and communication flows to keep staff informed
Soln Quality assurance/ monitoring procedures introduced in pharmacy Internal communication models – what needs to be communicated Procedures in the RCF e.g.:
how prn medicines handled, how practices need to differ if PCA or an RN administer medications, practices for packed versus non-packed medicines including ensuring non-packed
medicines are ordered & use monitored, how GPs order prn and non-packed orders and how administration recorded,
after hours medication changes & emergency supplies for ceasing supply of a particular packed medication or supply for a patient
Issue Integrity and reliability of medication charts (transcription errors to charts, how do you know it is current and complete).
Soln Standardisation? Electronic records? Use of pharmacy profile with regular validation?
Issue Patients using other pharmacies and doctors – variation in procedures and conflict between patient choice/rights and quality of care.
Soln Needs to be part of DAA contract with RCF and agreement with the patient i.e. explicit agreement about roles, obligations and responsibilities
Procedures for exceptions (available for agency staff) Issue Monitoring compliance for self-medicating patients Soln Agreement with residents
Table 4.12 Problems from the GP perspective and possible solutions
GP perspective Issue Pharmacy request scripts too early (contrary to dispensing software alerts) or after having
just written another prescription (legalities) or requests are inaccurate i.e. medication has been ceased or prescription is at RCF.
Soln Tripartisan agreement stipulating obligations and procedures Promotion and awareness raising of issues from perspectives of all stakeholders Template for reminders (what should it say) – Reminder should have patient’s name,
date of birth & Medicare number Keep resident/family, RCF and GP informed of what is needed when
Issue Doctors would prefer to generate prescriptions electronically to keep records in their medical software
Soln Provision of prescription computer in RCF or access to surgery computer from RCF, reminders prior to RCF visits
Issue Inability to monitor wastage or compliance based on prescription rates Soln Template for requests contain this information is fed back to GPs to allow monitoring Issue Lack of flexibility and responsiveness to changes Soln Raise awareness of RCF and pharmacy situations to modify GP perception or resolve
any real problems Issue Nurses/RCFs relinquished responsibility for medication management Soln Awareness raising to illustrate RCF procedures to GPs and address real problems Issue DAAs make more work for GP but easier for pharmacist Soln Raise GP awareness of all steps involved in safe, effective and efficient DAA service
(promotion to address perceptions and deal with any real problems) Raise pharmacy awareness of GP concerns and address any real prblems
Quality Medication Care Group, School of Medicine, University of Queensland63
Table 4.13 Problems from a community pharmacy perspective and possible solutions
Community Pharmacy perspective Issue Reliable and timely information on what to pack (includes charts-prescription mismatch) Soln Pharmacy require or accept written document/GP/doctor endorsed statement only
Review of medication charting procedures to eliminate ambiguous orders and ensure records are up-to-date
Whole new profile required from doctor including full drug regimen – view whole detailed drug chart – what has changed, what is the same
New packing template developed when changes occur RCFs use pharmacy-provided charts based on medication profiles & endorsed by GP For hospitals using PBS – doctors use medication chart as discharge prescription
(electronic document) signed by doctor faxed to pharmacy (new medications, changed or ceased medications – full regimen)
Issue Paperwork trail – owing prescriptions and PBS rules, and laws. Legalities of dispensing owing script if something happens to patient
Soln Raise awareness of other stakeholders of rules and laws pharmacy must operate within and any related pharmacy policy/procedure
Reminders (when done well) – template for managing prescriptions including means of monitoring wastage for GPs
Potential of medication chart to be accepted as the primary legal document for a residents prescribed medication to be accepted as PBS items.
Change the system of supply/payment system for patients on chronic medications packed into a DAA
Issue Unrealistic expectations from RCFs, doctors Issue No payment for high cost service Issue Resistance of other players to pharmacy attempts to improve situations Soln Address perceptions through promotion and mutual awareness raising
Promotional materials need to include what is required from all parties, what pharmacy has to do, why pharmacies need prescriptions in advance
Develop fee structures for DAA services e.g. an annual fee or provider-type fees paid for services such as prescription management, liaison with nursing staff, GPs and non-medical carers.
Also more explicit contracts and service agreements – verbally explain expectations and procedures and why do the pharmacy charges as it does!
Issue Pharmacy held responsible for issues such as charting and provision of sign-off sheets Soln Address service expectations in an agreement and payment for these services Issue Difficulty getting information on what products are suitable for packing or where to go for
this information. How to source information when nothing available. How to factor weather conditions especially in tropics i.e wet season in NT
Soln Use a risk assessment approach to existing information (decision tree) Collect and disseminate more information of drug instability Development of list of commonly packed products Document judgement calls made on suitability for packing
Issue DAA patients unique in their care needs and sometimes need to bend the rules to ensure continuity of supply
Soln Develop models of guidelines and legislative changes to facilitate caring fro DAA patients’ but ensure payment and monitoring of medication use by GP
Issue Errors in packing and checking DAAs Soln Appropriate packing procedures, environment (quiet, no interruption) and staff training
Pharmacy needs quality assurance/monitoring systems. If error occurs, determine how/why error occurred with goal of problem identification & resolution); packer & checker counselled in a non-accusatory way. Education/training offered if needed
Issue Need to rush packing and checking when patient discharged from hospital Soln Earliest possible notification from hospital pharmacy
Use of discharge template – full and accurate information Procedures to deal with hospital or clinic-supplied medications where there is an
expectation that a community pharmacy will repackage these into a DAA Issue Different points of view for hospital doctors and GPs; or GPs left out of the loop Soln Communication protocols for DAA patients entering & exiting hospital
Quality Medication Care Group, School of Medicine, University of Queensland64
Table 4.14 Problems from a hospital pharmacy perspective and possible solutions
Hospital pharmacy perspective Issue Lack of notification re: patient discharge Soln Need something to disseminate to medical staff to increase awareness that the
pharmacy needs a day or two to organize new DAA packs so need earlier than usual notification of discharge
Issue Difficult to identify which patients use DAAs, what type and who the community pharmacy who packs is
Soln Development of a patient, community pharmacy, RCF database about DAA use RCF charts/patient medication list/DAAs to be brought to hospital and have details of patient’s type of DAA and contact details for community pharmacy and GP.
Issue Difficulty communicating with community pharmacy & GPs Soln Electronic communication pathways for medication profiles between community and
hospital pharmacies Automated faxing or encrypted emailing of medication information sheet to GP on discharge
Issue Wastage – community pharmacy/ RCFs disposing of medication dispensed by hospital pharmacy rather than packing it
Soln Data base of RCFs and community pharmacies that contains details of their preferences with regard to using hospital medicines versus receiving discharge prescriptions only
Issue Privacy issues – need patient consent to share information with other Health PractitionersSoln Tripartisan agreement to initiate/continue DAA supply to include consent
Consent may be implied but explicit consent is preferred to prevent future problems
Table 4.15 Problems from a community patient perspective and possible solutions
Community patient perspective Issue Difficulties handling/using DAAs Soln Assessment of suitability of pack (including type) that involves an observation of a
patients ability to manipulate a pack Promotion of some negative aspects (problems) and potential solutions so that new DAA patients have a realistic expectation and strategies to deal with problems Ability to use a DAA and medication compliance may change with time so need monitoring systems in place e.g. repeat patient assessment (including knowledge and willingness to pay for pack) and formal review of returned medicines
Issue Complacency and the potential for error in packing Soln Responsibility of patient for hand-held template and facilitating information flow
Procedures to ensure patient medication knowledge is retained after starting DAA Issue Cost Soln Raising awareness of what is involved in a DAA service (promotion to address
perception) Be explicit about patient/carer obligations and responsibilities in a service agreement
Using a DAA complicated the continuity of care between the community or RCF and
hospital. The problems addressed by APAC guidelines on the continuum of quality use
of medicines between hospital and community (Australian Pharmaceutical Advisory
Council 1998) exist but there are specific DAA-linked problems including:
Hospital pharmacists concern about integrity and identification of drugs in DAAs
brought in on admission
Need for timely charting in RCFs and prescription for medications supplied under
the Pharmaceutical Benefits Scheme (PBS)
Preparing DAAs is resource intensive for community pharmacies. Requests for
DAAs outside scheduled packing times can be difficult to manage
Community pharmacy concern about packing medications dispensed by others
Quality Medication Care Group, School of Medicine, University of Queensland65
Needs of stakeholders to facilitate the smooth transition from hospital to the community
and RCFs for patients who use pharmacy-packed DAAs include:
Community pharmacies need to know when a DAA patient (community or RCF-
based) is admitted to hospital. Hospitals could advise community pharmacy of DAA
patient admission.
Hospital pharmacists need to know a patient uses a DAA and with whom to liaise
(GP, community pharmacy) as discharge approaches. When a DAA patient is
admitted, the hospital could trigger an alert/warning associated with patient ID
and/or address (for RCFs) on dispensing system to notify staff to activate DAA
organization protocol.
Failure to consider the policies and procedures of others involved in patient care
can lead to wastage and interruptions to the continuity of medication supply.
By compiling information on local community pharmacies who provide DAAs,
hospitals can address patient needs. The database would details such as who to
contact for a given patient and pharmacy details such as the type and cost of
packs, whether delivery available, requirements regarding prescriptions and stance
on using hospital dispensed medications. This information can be utilised when
advising patients about possibility of using DAAs if a new patient or organising
discharge supply for current DAA users.
Hospital staff, medical, nursing and pharmacy, need to recognise that discharging a
patient who uses a DAA requires more advanced notice as part of discharge
planning. At least a days notice is needed and staff need to be aware of potential
issues with continuity of care if discharge occurs outside business hours. Leading
up to discharge, both the GP and community pharmacy could be notified. The
hospital can confirm when the when new pack can be provided and whether
discharge medication is required.
When a patient returns to an RCF, staff need a legally valid and reliable order to
administer drugs. There can be delays in getting this order from the GP due to poor
information flow and the timing of discharge. Patients could return to an RCF with a
valid medication chart.
GPs need a timely discharge summary and need to know about changes for their
own records.
To pack the DAA, the community pharmacy needs timely, reliable and complete
information on the patient’s current drug regimen and a valid prescription for
payment. The hospital could fax or email (where security is addressed) a copy of
hospital prescriptions and discharge medication list to community pharmacy.
Community patients need to be understand what is required of them to facilitate the
transfer of medication regimen information and continuity of supply. Actions of
community patients also contribute to ‘continuity of information and supply’ failures.
A patient held template for community patients or a copy of the RCF medication
chart, contact details of a patient’s GP and pharmacy, and consent to share
information as part of a service agreement would aid medication information
transfer.
There are may variations in the systems of DAA service provision and in hospital
medication supply systems. Appropriate strategies that recognise these variations in
practice when care setting transitions occur reduce the chance of failures in continuity
of information and supply for DAA users.
Quality Medication Care Group, School of Medicine, University of Queensland66
4.5 BEST PRACTICE MODEL FOR THE PROVISION AND USE
OF DOSE ADMINISTRATION AIDS IN THE COMMUNITY
4.5.1 INTRODUCTION
Dose Administration Aids are widely used in Australia as a tool to assist with
medication management in a variety of settings. Approximately 80% of community
pharmacies in Australia pack and supply DAAs to meet the needs of consumers living
in the community. DAAs are used in the community by consumers who need and/or
would like additional support in managing their medications due to multiple solid
medication use, poor health, and disability.
The provision and use of DAAs is a complex process involving a number of steps and
the input and collaboration of different parties (the pharmacy, doctors, patients, carers
and sometimes community support services such as community nurses). Figure 4.5
shows the steps involved in DAA provision and use once the decision to commence
DAA use is made. From the pharmacy perspective, dispensing, packing and checking
medications for supply in DAAs, and likely support services such as managing the
patients’ prescriptions, account keeping, and delivery, take significantly more time than
providing the same medications in original packs.
Best practice is defined as a technique or methodology that, through experience and
research, has been proven to reliably lead to a desired result. The QMC group (funded
by the Australian Government via the 3rd Community Pharmacy Agreement Research &
Development Grants managed by the Pharmacy Guild of Australia) has conducted obs-
-ervational research and sought extensive feedback from a range of key stakeholders
from the medical, nursing and pharmacy professions, as well as industry, consumer
and government sectors to identify the factors that are contributing to unsafe practices,
reduced effectiveness and inefficiencies in the provision and use of DAAs. Key barriers
to safe, efficient and effective provision and use of DAAs in the community include:
Poor communication and the breakdown of information flows between parties.
Poor awareness of responsibilities and obligations which cause systems to fail
and require rework (i.e. follow-up communication, re-packing).
Community patient loss of ownership with respect to medication management,
as evidenced by lower levels of medication knowledge and high levels of
dependency on others for help with medication management.
Lack of monitoring and accountability for the quality of the DAA services.
The high cost of providing DAA services to community patients borne by
community pharmacy.
4.5.2 PREAMBLE
4.5.2.1 Purpose and scope of this best practice model
There are currently a range of standards and guidelines relating to the supply of medic-
-ines in the community and to the provision of dose administration aids by pharmacy.
These include: Professional Practice Standards (Pharmaceutical Society of Australia
2002), Dose Administration Guidelines (Pharmaceutical Society of Australia 1999),
Quality Care Pharmacy Program (QCPP) professional practice guidelines, DAA
specific manufacturers/suppliers guidelines (i.e. Webstercare, Douglas, Persocare,
Quality Medication Care Group, School of Medicine, University of Queensland67
AHPS and MPS). Various nursing guidelines and policies or procedures of nursing
organisations make reference to the roles of nurses. There is no existing resource,
however, that accounts for the multi-disciplinary involvement in the provision and use of
DAAs and that includes specific processes or procedures for optimising the provision
and use of dose administration aids for community patients.
1a. GP writes medication order• Records full current drug regimen for community patients
1b. GP writes prescription as order to dispense
3. Develop or update pharmacy-held medication profile• Review regimen, prepare dosing schedule (what to take when)• Develop packing plan (identify what is packed & not packed)
2. Transmit drug regimen information to pharmacy
3a. Support activities• Manage prescriptions
(e.g notify when new script due)
• Accounts, adequate stock on hand
8. Pack medication according to profile
7. Prepare for packing (labels, assembling medications etc)
6. Store dispensed medications not yet packed
10. Filled packs stored in pharmacy
12. Medication receipt • Counsel on medications; provide written information (CMI, MIC)
15. Return unused medication to pharmacy• Pharmacy records & monitors missed doses, taking
action as needed; deals with returned medications
No
n-p
ac
ke
d m
ed
icati
on
s
Potential patient role:
• Monitor need for new
prescription;
• Obtain new prescription
as needed;
• Deliver new prescription
to pharmacy
Doctor Doctor & patient
Pharmacy support staff e.g. Dispensary technician
Pharmacist Patient
Patient & pharmacy
Key
Potential patient role:
• Monitor need for new
prescription;
• Obtain new prescription
as needed;
• Deliver new prescription
to pharmacy
Doctor Doctor & patient
Pharmacy support staff e.g. Dispensary technician
Pharmacist Patient
Patient & pharmacy
Key
9. Check packed medication
5. Check dispensed prescription
11. Deliver / collect medication
4. Dispense prescription
Figure 4.5 Steps and input required in the provision and use of DAAs, once DAA
service initiated
Quality Medication Care Group, School of Medicine, University of Queensland68
While profession-specific guidelines are needed, there is also a need for an over-
arching integrated model. This proposed model takes into account existing professional
standards and relevant legislation, and makes recommendations to address the
limitations in the provision and use DAAs, to optimise the effectiveness of this
intervention and to facilitate interaction between health professionals and health
consumers. Despite this teamwork approach, many of these recommendations may
appear pharmacy centric. This is because much of the resource and cost burden
arising from the provision of DAAs is borne by community pharmacy. However, optimal
provision and use of DAAs cannot occur without multi-disciplinary involvement and the
recommendations reflect the roles and responsibilities of the different stakeholders
(see Figure 4.5, where they key indicates who generally performs the various steps).
The scope of these recommendations is restricted to issues relating specifically to DAA
use where it is different from medication management using original packs. The key
issues in providing medications in a DAA rather than original packs are:
(1) To supply the correct medicine, the pharmacist needs to know the patient’s
complete and current drug regimen (as all items must be packed) rather than
only a single new or changed medicine to be supplied in its original pack.
(2) The pharmacy, instead of the patient, tends to take responsibility for managing
prescriptions and ensuring the continuity of medication supply in an effort to
streamline processes.
(3) The extra steps involved in producing DAAs and the interdependence of the
stakeholders increases the potential for errors at any point in the process and
by anyone involved.
The purpose of this document is not to dictate or set standards in medication
management and the use of dose administration aids. Instead, the goal is to provide a
range of strategies and tools to overcome the barriers to safe, effective and efficient
DAA provision and use. Professional bodies may choose to incorporate these issues
into future guidelines or standards. The recommendations outlined in this document are
inter-related and together constitute a possible implementation model as outlined in
Figure 4.6.
This best practice model is based on the current situation whereby no subsidies are
available for the provision of DAAs. Were funding to become available in the future for
DAA provision, the elements of the model (specifically recommendation 1.3.1 and
recommendation 1.3.2) could be incorporated as part of the system of checks and
balances that would be necessary to ensure accountability with respect to the quality
and quantity of DAA services offered to community patients. An additional benefit is
that these documents could also be used to form part of a program evaluation of any
future DAA implementation. The model concentrates on the nature of information
transfers and processes rather than a specific mode (e.g. fax, encrypted email, etc).
Future IT developments (e.g. communication channels, software, encryption and public
key technologies, etc) may well be enablers to further increase safety and efficiency.
Quality Medication Care Group, School of Medicine, University of Queensland69
Potential new DAA user• Does patient meet DAA criteria? Have ability to use DAA?• Assess baseline medication management ability, knowledge• Communication between patient/carer, GP, pharmacy +/- community nurses involved
• Any other related services e.g. HMR?
Is it 6 months since last
formalcheck?
Is it 6 months since last
formalcheck?
Yes
No
Patie
nt-
held
tem
pla
te k
ept
up-t
o-d
ate
by
pa
tien
t, c
are
r &
GP
Inclu
din
g m
edic
atio
ns c
hang
es n
ot ne
edin
g p
rescription
DAA packed by staff using template
Pharmacist checks DAA using template
DAA sent to patient
Pharmacist check/update template before next packing
Patient-held template/ medication record
Tripartisan AgreementFormalise service to be delivered, expectations & obligations
Patient/carer GP (+/- specialists)
Pharmacy +/- community nurses
Full review/ renewal of template• Patient, GP & pharmacy
Template for medication packing• Shows medication regimen & preferred time of day
for each dose
Reflect• Patient habits/preferences; GP preferences• Current medication & optimal schedule• Type of pack & packing interval• Check medications suitable for packing• Other constraints e.g vision impaired
Pre
scrip
tio
ns d
ispen
se
d
Template approved by all
Template copy to GP
Figure 4.6 Community patient implementation model
4.5.3 RECOMMENDATIONS
4.5.3.1 Assessing community patients’ need for DAA services
Prior to agreeing to initiate a DAA service, the needs and ability to manage a
DAA should be formally assessed including a structured assessment of pre-DAA
medication management ability and medication knowledge.
Users of DAAs generally have higher care needs and pharmacies appear to be well
able to identify people in need of a DAA. The existing approach to targeting DAA
services (Pharmaceutical Society of Australia 1999) emphasises individual assessment
while suggesting the following criteria:
Taking five or more medications
A history of poor compliance
On a complex medication regimen
Showing signs of cognitive or physical impairments
In Phase 2 ((Ientile et al. 2004), Table 4.33), the most frequently cited characteristics of
patients who benefited from using a DAA were similar to the PSA guidelines:
The elderly
Those with multiple medications
Those who were confused or with cognitive impairment
Those unable to coordinate their medications correctly
Those with physical barriers to medication taking
This assessment is apparently, informal and undocumented (4.1.2.2 and 4.1.3).
Quality Medication Care Group, School of Medicine, University of Queensland70
Analysis of HIC data (sections 7.2 and 5) supports the view that DAA users have
greater care needs as many of the covariates of higher health costs were proxies for
greater disability and/or poorer health. Since the people for whom DAA use might be
suggested generally already have impaired medication management abilities, formal
assessment of DAA needs and ability to use a DAA will help pharmacies provide an
appropriate service and to identify any potential risks for a given patient in using a DAA
by:
Appropriately defining specific needs (e.g. to deal with poor vision or reduced
mobility) that will need to be addressed by the DAA service provided.
Identifying potential risks from using a DAA (e.g. reduced medication knowledge)
so that risk reduction strategies can be put into place.
Ensuring that a specific DAA is optimal for an individual patient. Some patients
have more difficulty using certain devices than others.
Identifying other potential medication use problems (e.g. storing medicines in
inappropriate places).
Keeping a record of this assessment would provide a means of:
Accountability that DAAs are provided to people in need of them.
Quality assurance and patient risk management. Pre-DAA assessment of abilities
and knowledge will allow future monitoring of medication management ability and
knowledge when periodic re-assessments are performed and compared against
baseline (see recommendation in section 4.5.3.9).
As part of the assessment phase, health professionals involved in the medication-
related care of the patient need to be consulted in the decision to initiate a DAA
service, as each may be affected by the decision. Further, alternative medication
management strategies may be deemed to be more suitable for a specific patient or
additional services, such as a Home Medicines Review (HMR) may be requested.
4.5.3.2 Tripartisan agreement specifying obligations and promoting mutual
awareness
Community pharmacy should negotiate an agreement between themselves, the
community patient and/or carer and the patients’ doctor/s prior to beginning to
provide a DAA service.
The agreement should address:
That it is important for the pharmacy to know the correct, current medication
regimen so that a DAA containing the correct medications can be provided on time,
i.e. the importance of pharmacy profile maintenance.
How the service will work:
What type of pack will be provided (brand, period, how many packs needed,
how many weeks’ worth will be provided at once).
What medication will be packed (including whether some or all non-prescription
items such as complementary or vitamin supplements are to be packed) and
what is the optimal schedule (see 4.5.3.3).
How far in advance of distribution will the DAA be packed (and in some cases,
which day of the week) (i.e. what is the packing cycle?).
Where original pack medication and prescriptions will be stored.
Whether the pack will be delivered by the pharmacy or collected by the patient,
the delivery procedure and schedule.
Quality Medication Care Group, School of Medicine, University of Queensland71
What will happen if a medicine in a pack needs to be changed before the
current pack is finished.
What is the cost of the service, who is the payer and how the pharmacy will
charge for the service.
Procedures for start-up and preparing the first pack e.g. will medicines already
dispensed and in the person’s home be packed or will the person be asked to bring
all their medicines from home to rationalise?
Privacy: Patient consents to sharing of medication-related information between
doctor/s and pharmacy (and hospital and community nurse if required).
Medication changes:
Whose responsibility will it be to inform the pharmacy of medication changes
(patient/carer or doctor)?
What format is acceptable, (i.e. written instructions only- amended to full
regimen indicating ceased/changed and new drugs)?
What constitutes timely notification of changes so that rework is minimised.
The circumstances when a medication change can be delayed until the next
pack or when a change needs to be made more quickly.
Prescription management:
The understanding that a valid prescription is required before medicines can be
supplied in a pack, and that, depending on the packing cycle, the date a
prescription is needed may be earlier than if original packs were provided.
Will the pharmacy remind the patient to visit the doctor to obtain a new
prescription?
Will the pharmacy generate prescription reminder notices for the doctor/s?
GP practices related to prescriptions and repeat reminders e.g. under what
circumstances is the GP willing to write a prescription without seeing the
patient, any GP costs.
Who is responsible for following up that prescriptions are at the pharmacy in
time for packing (patient/carer or pharmacy).
What will happen if patient has no new prescription (i.e. medication not packed
in the DAA?).
Procedure for use of the DAA and what to do in exceptional situations e.g.:
When things go wrong (i.e. what to do if drop medication or open wrong blister).
What to do if have to have a prescription filled at another pharmacy (not the one
that provides DAAs) e.g. hospital pharmacy.
When patient away from home e.g. on holidays or in hospital, where there might
be a need for more packs than usual or where there is an interruption in the use
of packed medications (e.g. hospital admission or discharge).
How to ensure optimal use of the DAA:
How the pharmacy will provide counselling and medication information (e.g.
CMIs) to the patient (or their carer)
Patient agrees to return unused medication in the DAA to the pharmacy.
Pharmacist reviews returned DAAs to assess compliance or problems with use.
Pharmacist counsels patient/carer if problems occur.
Consent to re-assess how patients are managing with the DAA after a specified
time (such as 6-monthly) (see 4.5.3.1). This may involve a home visit.
The process of completing an agreement of this nature ensures that the patient and/or
carer receives adequate information about what is involved in receiving a DAA service
and training in how to use the DAA and what do if problems occur, and can address
Quality Medication Care Group, School of Medicine, University of Queensland72
patient preferences. In addition, this agreement allows the pharmacy, doctor and
patient to negotiate who will be responsible for undertaking the necessary activities to
ensure the accuracy of the DAA, the continuity of medication supply and the optimal
outcome for the patient. The agreement also allows the GP’s expectations and practice
patterns, policies and procedures to be recognised.
This agreement may also provide the pharmacy with some form of risk-reduction in
relation to potential future claims arising from situations where a patient/relative makes
a claim against the pharmacy.
4.5.3.3 Patient held medication template to empower patient and facilitate
communication about medication regimen and changes
The pharmacy, in collaboration with the patient and the patient’s doctor, should
establish a template for medication packing for each community patient
receiving a DAA service.
The patient should carry a current copy of this medication template with them as
an accurate record of their current medication regimen.
The patient’s doctor should record medication changes on the patient’s template
(in addition to other processes used by the pharmacy, patient and doctor to
maintain the currency of the profile, such as routine 6-monthly concordance
checks).
The patient medication template addresses a variety of issues. From the pharmacy
perspective it is essential that the pharmacy have a full and accurate record of the
patient’s current medication regimen from which to pack the DAA. This record should
reflect the patient’s current, optimal schedule and include detailed directions for
medication use including the dose, the frequency, the preferred time of day for each
dose (taking into account doctor and patient preferences) and which medications are
suitable for packing (see 4.5.3.6). The process of constructing the patient template
provides the pharmacist with the opportunity to review the patient’s medication regimen
to minimise the risk of medication misadventure, to ensure unsuitable medications are
not packed and to maximise the therapeutic benefits.
This template should be approved by the patient and the patient’s doctor (who, as the
primary medical care provider, has the ultimate responsibility for prescribing the
patient’s current drug regimen) and all parties should maintain a current copy. When a
change is made to the template, the pharmacy should provide a new copy of the
template to the patient and retain the out-dated template (to minimise latter confusion if
two different template are in circulation). This process ensures that the pharmacy has
an accurate record of the patients medication regimen and directions for use
(according to the intentions of the prescribing doctor) and that the doctor/s is aware of
what medications the patient is taking (both prescription and OTC), how the patient is
taking them and also provides the opportunity for the doctor to review the regimen if
necessary. This approach should be supported by a quality assurance activity, a
routine, 6-monthly check of concordance between the GPs records, current
medications as reported by the patient, and the pharmacy profile.
Quality Medication Care Group, School of Medicine, University of Queensland73
The strategy of having the patient/carer maintain a copy of the template has a three-
fold effect. First this strategy encourages the patient to take ownership of their
medication management by increasing their responsibility in that process and
improving their knowledge of their medication. Secondly, this medication template
facilitates communication about medication changes, especially those not requiring a
prescription (i.e. ceased medication or changes to medication dosages). In these
instances, the doctor can annotate the patient-held record and the patient and/or doctor
can provide written communication of medication changes to the pharmacy. Thirdly, the
medication template constitutes a legitimate/authorised, reliable, current and complete
record of the patient’s regimen which could be helpful if the patient consults with a new
health professional or is admitted to hospital.
4.5.3.4 Communicating medication changes
The pharmacy should maintain a written record of all communication regarding
medication changes for patients receiving DAA services.
The accuracy of the DAA pack is largely dependent on the currency of the pharmacy
profile. Maintaining a current medication profile requires full and detailed
communication about the medication regimen and any changes. In the community
setting, where the responsibility for informing the pharmacy of medication changes
generally falls to the patient, it is quite likely that the pharmacist may not be informed of
medication changes, particularly when the changes are not documented on a
prescription (i.e. if patients have a change in the dosage of their medications or cease
using a medication). The patient presenting the pharmacy with a new prescription does
not constitute adequate communication as the pharmacist may be uncertain of whether
this new medication is in addition to the current regimen or to replace another
medication.
The pharmacy should request that all medication changes be communicated in writing
(whether by fax, email, letter, etc). The communication should reflect changes in the
context of the patient’s full regimen and should be unambiguous and detailed enough
to allow the pharmacist to alter the medication profile and the DAA without further
consultation. The patient-held medication template should fulfil this requirement (see
4.5.3.3). An updated medication template should be provided to the doctor/s and
patient/carer to acknowledge the pharmacist’s receipt of the communication and to
ensure that all parties have a current record. Written records and documentation of
medication changes should be maintained by the pharmacy to ensure the accuracy
and accountability of DAA services. Note that if the pharmacy retained the out-dated,
changed copy of the patient-held template after annotating that the change had been
made to the profile, this should be sufficient documentation that a change was
communicated and enacted.
4.5.3.5 Continuity of care between hospital and community for patient with
DAAs
Each community pharmacy, hospital, doctor and patient should be aware of their
role and develop procedures to meet their responsibilities to ensure continuity of
care between hospital and the community, for patients using DAAs.
Quality Medication Care Group, School of Medicine, University of Queensland74
When a community patient uses a DAA, the issue of ensuring continuity of care on
discharge from the hospital to the community is more complicated and requires greater
collaboration between the hospital and community health care services (general
practitioners, community pharmacy, community nurses and carers). Stakeholders all
have roles and responsibilities in ensuring continuity of care for DAA users. To improve
practice, the needs and perspectives of all stakeholders should be considered. A failure
to consider the policies and preferences of others involved in patient care can lead to
wastage and interruptions to the continuity of medication supply.
Figure 4.7 shows the steps and inputs required to facilitate the supply of medicines
packed in a DAA after discharge from hospital. When a patient is admitted to hospital,
irrespective of whether they have their medicines in original packs or DAAs, the
hospital needs reliable information on the patient’s medication regimen, and so this
aspect is beyond the scope of this best practice model. Because the use of a DAA
complicates discharge, on admission, a hospital pharmacy (or the ward staff where
hospital pharmacists are not available) need to know firstly that a patient uses a DAA
and who to liaise with (which GP, community pharmacy and, if applicable, community
nursing service). In addition, the hospital pharmacy needs sufficient notice from the
hospital medical staff to allow for the necessary liaison. To pack the DAA, the
community pharmacy needs timely, reliable and complete information on the patient’s
current drug regimen (the discharge medication regimen) and a valid prescription for
payment and/or a sufficient supply of medication. To maximise safety and to minimise
wastage while ensuring essential medications are available, the hospital pharmacy
should liaise with the community pharmacy about what medications to supply on
discharge. For example:
The patient may be confused if the hospital supplies a different brand of an item
than is usually packed so that no discharge supply is a safer option.
The patient may be prescribed a drug that is not routinely available in the
community (e.g. a trial drug) so that discharge supply of this item is necessary for
continuity of supply.
The continuity of medication supply for DAA users is adversely affected by weekend or
after-hours discharge where it is unlikely that a hospital pharmacist will be available to
coordinate the information flow between hospital and community and the community
pharmacy is unlikely to be able to provide a new DAA in a short-time frame.
Where a hospital pharmacist is not available to perform this liaison task (e.g. if outside
hospital pharmacy hours or there is no hospital pharmacy), the hospital needs an
alternative procedure to ensure the continuity of information and supply for DAA users
who are to be discharged from hospital.
When a patient is admitted to hospital, DAAs may facilitate information flow to the
hospital from the community provided that they are prepared according to the
recommendations contained in this best practice document. Specifically, DAAs need to
be labelled or accompanied by a current document with the following information:
patient’s name and address, pharmacy name and contact details, doctor’s name and
contact details, date packed, list of all packed medications and their administration
schedule and list of non-packed medications also taken by patient.
Quality Medication Care Group, School of Medicine, University of Queensland75
Community patient
using DAA
Patient/relative brings DAA and medication profile to hospital
Hospital compile medication history/profile:
• Requires reliable & current information
• Sourced from patient/ carer, GP & community pharmacy
• Includes ADRs, allergies, DAA use
Patient receives treatment including changes to current medication regimen
Remind discharge doctor that sufficient notice of discharge is required to organise DAA
Contact community pharmacy to inform them that patient has been admitted, to withhold packs pro tem & new pack will be required. Ascertain pharmacy preferences for discharge procedures
If community pharmacy is a new contact add details to hospital pharmacy DAA database including preferences for discharge medications or prescriptions, packing schedule & fax number Medical staff inform
hospital pharmacy that patient will be discharged
If DAA pack not available on disch-arge, dispense enough of needed medications until DAA is ready
Community pharmacy to pack DAA from the hospital discharge list
General Practitioner contacts community pharmacy to give the OK to dispense from the hospital discharge medication listCommunity pharmacy
delivers DAA pack (or patient/carer collects)
Patient receives DAA in time for next doses OR uses hospital supply until DAA arrives
Adm
issio
nD
urin
g h
ospita
l sta
y 24hrs
†prio
r to d
ischarg
eD
ischarg
eP
ost d
ischarg
e
Counsel patient at discharge, explaining what has been arranged regarding DAA provision
Fax/email* a copy of hospital prescription and discharge summary to community pharmacy
Check pharmacy database. Contact pharmacy to confirm what pres-criptions &/or medicine is required & when pack will be available
† earlier if weekend discharge* If security addressed
Figure 4.7 Steps and input required to facilitate continuity of medication supply for
patients using DAAs.
The following tools and documents could be utilised to facilitate information flows
between the community and hospital:
Quality Medication Care Group, School of Medicine, University of Queensland76
Patient held template (see recommendation 4.5.3.3 and Appendix F).
Hospital pharmacy database on community pharmacy and DAA patients (see
Appendix F for details of the type of information to collect/record).
Patient agreement to consent for information sharing (see recommendation
4.5.3.2).
4.5.3.6 Quality control, quality assurance/ monitoring procedures for packing/
checking and communications flows
Community pharmacies need to introduce procedures for quality control and
quality assurance and monitoring in DAA provision. These activities should be
recorded. Good packing procedures should more closely resemble those of
good manufacturing practice that incorporate quality processes. Support for the
development of a new Pharmacy standard or guideline “Current Good Packing
Practices” would significantly contribute to improving current practice.
Patients should be educated on the appropriate storage of medications in a DAA.
In the absence of better evidence on the stability of medications repackage into
DAAs, pharmacists should use existing guidelines on specific dose forms
considered as unsuitable for packing.
More evidence on the suitability of medications for repackaging into DAAs is
required. Strategies include:
Collecting empirical evidence by reporting observed changes in medications
in packs to a central body.
Perform short term, in use stability studies on the most commonly packed
solid dosage forms.
Using regulatory/manufacturer approaches to make more stability
information available to pharmacies.
Written procedures should include quality control measures (to prevent errors by
inspection) and quality assurance procedures (to reduce errors by having quality
systems and allowing for problem identification and resolution). Appropriate
documentation should be kept to support these procedures. Record keeping and
monitoring may be time consuming but it can help improve the safety and efficiency of
DAA provision and also constitutes a form of legal risk-reduction. Currently, RCFs have
procedures for reporting errors and dealing with administration problems such as chart
audits, and staff training. Community Pharmacy needs to implement similar procedures
i.e. packing and checking audits, communication audits, staff training and procedures
for dealing with packing errors – i.e. checker informs the staff member who packed of
error, the situation/system is reviewed and remedial action is taken e.g. the staff
member instructed/ trained/coached on ways to prevent making this mistake again.
Patient education on the preferred storage and use of DAAs is especially important and
patients should be encouraged to report any visual or other problems which may be
related to the medication’s stability to their pharmacy.
There is a need for more information about the stability of repackaged medications to
support pharmacists’ decision making. There are several approaches:
The current guidelines of drug dose forms unsuitable for DAA packing (effervescent
tablets, dispersible tablets, buccal tablets, sublingual tablets and wafers) are
Quality Medication Care Group, School of Medicine, University of Queensland77
appropriate because these medications are administered in a different way to the
majority of medications in multi-dose packs (taken at one time). Further, these dose
forms are more fragile and may be damaged if packed in a multi-dose DAA
The PSA (as a producer of professional guidelines and a body representing
pharmacy overall) should act as a central body to receive and collate reports of
physicochemical stability reported by health professionals. The reporting
mechanism could be in a form similar to the Adverse Drug Reaction Advisory
Committee (ADRAC) reports (See section 10.2.2.1 for sample).
Stability studies could be performed. To standardise the studies, an easily
accessible (widely disseminated) guideline outlining the methodology for
conducting these short term, in-use stability studies should be developed in
consultation with the various regulatory authorities, manufacturers of DAAs and
experts in the field. Drug targets for studies should be prioritised based on the most
commonly packed medications and those where there is a higher theoretical risk of
instability.
The Therapeutics Goods Administration (TGA) should support the collection of
short-term stability data derived under defined conditions and without the protection
of the original packaging for all medicinal products. Alternatively, manufacturers
should be required to indicate those products for which there is an absolute
contradiction to short term storage in the absence of original packaging. Such
information would permit the assessment of the true significance of stability
problems should a medicinal product need to be removed from its original pack
several days prior to administration.
4.5.3.7 Efficiency in the pharmacies procedures
The cost of providing DAAs to community patients can be reduced by increasing the
efficiency of supply. There is a great deal of variation in the efficiency of DAA supply to
community patients and between pharmacies supplying RCFs and pharmacies
supplying community patients only. Inefficiencies can arise when rework is required
due to poor systems to manage medication change (as well as wastage of dispensed
medications) (see 4.5.3.4), and in the internal practices in the pharmacy. There are a
number of strategies and procedures utilised by some community pharmacies that
minimise the time required to manage the supply of medications in DAA without
compromising the quality of the service. These strategies relate to prescription
management, packing procedures and appropriate use of human resources.
4.5.3.7.1 Prescription management
During the packing session, the pharmacy should record the quantity of
dispensed medication and the availability of repeat prescriptions to ensure
efficiency in managing the continuity of supply of medication for patients using
DAAs.
The pharmacy should have a system to provide written prescription requests
and reminders to doctors and/or patients to ensure that prescriptions are
available for all PBS medication to be packed in a DAA.
Prescription requests and reminders can improve the relationship between
stakeholders and increase the efficiency of communication between doctors and
community pharmacy, IF they are done well. Doctors have commented that reminders
need to be timely and accurate, that it should be easy to identify to which patient/s the
Quality Medication Care Group, School of Medicine, University of Queensland78
communication relates (i.e. include patients names, address and Medicare number).
This communication should also allow for doctors to monitor compliance/medication
use and presents an opportunity for information sharing and accuracy checking. A
template of the type of information that could be useful in a prescription reminder is
included in Appendix F.
Doctors and patients need to recognise that this is a service provided by pharmacy that
is labour intensive and costly. Pharmacist need to inform other stakeholders that a
current prescription is a legal requirement associated with the supply of prescription-
only medicines and to receive subsidised medicines under the PBS.
In devising their systems, pharmacies need to recognise doctors’ practice patterns e.g.:
Requests for new prescriptions for a patient should be sent in batches even if some
prescriptions are needed some time in the future rather than one at a time. This
saves the GP having to view the chart for each separate request.
Dispensing software may alert the GP that a prescription is “too early”, so that the
written prescription request should include a “date prescription needed by” to allow
the GP to check medication usage. The GP needs to be aware of the timing of the
packing cycle followed by the pharmacy as an explanation of why a request is
apparently “early” (see 4.5.3.2).
For pharmacy, there are opportunities to provide this service more efficiently by
optimising communication channels (i.e. fax or email) and utilising existing
dispensing/packing software reports.
4.5.3.7.2 Packing procedures
Due to variation in the types of DAAs used, the patients’ medication regimen and even
the number of packs that the pharmacy produces, a variety of packing methods and
procedures are utilised by community pharmacies. Some of the strategies that reduced
the amount of time spent packing a DAA included:
Packing then checking the DAA as separate procedures performed by different
staff members,
Having a dedicated packing space (where interruptions can be minimised) and
experienced staff, where possible.
Conducting staff training – so that all staff packing DAAs know what to do, what
order to do it in, and that this procedure is followed consistently by staff preparing
DAAs (an area to address in any quality assurance program, see 4.5.3.6).
4.5.3.7.3 Roles for pharmacists and non-pharmacist staff
Efficiency in DAA provision can be enhanced by good human resource management.
Specific tasks in the provision of DAA services should be performed by a pharmacist.
These tasks are those associated with the pharmacist’s professional responsibility i.e.
prepare packing profile, checking the dispensing of original packs, changing and
reviewing the packing profile, and checking the accuracy of the pack after packing.
Dispensary assistants and other non-pharmacist staff were found to be more efficient
at packing (take less time to pack and are a less expensive resource), possibly
because they can perform this task with fewer distractions. Provided that non-
pharmacist staff receive adequate training and supervision, the quality and accuracy of
DAA services is not compromised (Ientile et al. 2004).
Quality Medication Care Group, School of Medicine, University of Queensland79
4.5.3.8 Negotiating a mutually acceptable payment for DAA services
Community pharmacy needs to negotiate a mutually acceptable price for
providing DAA services and to provide this service only to patients who will
benefit from it.
The provision of DAAs is labour intensive and costly to pharmacy however the benefits
of this service have been recognised by doctors, patients, relatives and carers. The
majority of patients using DAAs have indicated that they are willing to pay for this
service and the amount that they are willing to pay is greater than the current amount
they pay. In addition, patients who paid for DAAs were more likely to appreciate this
service. However, some consideration should be given to the patients who are unable
to afford this service but depend on it for their medication management. Pharmacies
need to negotiate a realistic price for this service and to ensure that a patient has a real
need (see 4.5.3.1) to ensure that this service is worthwhile. Providing DAA services to
patients who are unwilling to pay or have no real need for this service constitutes a
waste of pharmacy staff time and efforts which could have a greater impact on
outcomes for the community and be better rewarded if re-directed to other services.
4.5.3.9 Ensuring adequate monitoring and care of DAA patients.
Community patients using DAAs should be monitored by their general
practitioner and community pharmacy to ensure that:
Any problems patients might have when starting to use a DAA are detected
and addressed.
Patients continue to manage their medications when some or all are in a
DAA.
Community patients who start using a DAA may have difficulty using the device
correctly. While educating patients about how to use the device (4.5.3.2) will minimise
this risk, close monitoring by the patient’s GP and community pharmacy in the initial
phase of DAA use should be carried out to identify and solve ‘teething’ problems.
As DAA use continues, systems are required to monitor that a patient continues to be
able to manage their medications. Evidence (Ientile et al. 2004) indicates that a small
proportion of DAA patients are not seeing their GPs regularly as the pharmacy is
managing their prescriptions and the GP is writing prescriptions without the patient
attending a consultation. While these patients may have more interaction with the
community pharmacy than other DAA users, this interaction is less likely to involve
communication about medication and counselling. DAA users in the community are a
high needs group that require monitoring to ensure they are coping with the medication
management. This monitoring should involve:
Seeing the GP for prescription renewal and consultation at least once per 3
months.
Seeing the pharmacist for medication information as needed (when medication
changes or problems are observed) or at least once per 3 months.
Returning DAA packs to prevent hoarding of unused medications and to allow for
compliance monitoring.
Regular re-assessment each six months to assess medication management ability,
medication knowledge and concordance in medication regimen records (see
4.5.3.1 and 4.5.3.2).
Quality Medication Care Group, School of Medicine, University of Queensland80
4.6 BEST PRACTICE MODEL FOR THE PROVISION AND USE
OF DOSE ADMINISTRATION AIDS IN RESIDENTIAL CARE
FACILITIES
4.6.1 INTRODUCTION
Dose Administration Aids are widely used in Australia as a tool to assist with
medication management in a variety of settings. Approximately 40% of community
pharmacies in Australia pack and supply DAAs to meet the needs of residential care
facilities (RCFs). DAAs are used in the majority of RCFs in Australia to assist facility
staff to provide safe and efficient administration of residents’ medication.
The provision and use of DAAs to RCFs is a complex process involving a number of
steps and the input and collaboration of different parties (the pharmacy, the RCF,
doctors, and the patients). Figure 4.8 shows the steps involved in DAA provision and
use. From the pharmacy perspective, dispensing, packing and checking medications
for supply in DAAs, and likely support services such as managing the patients’
prescriptions, account keeping, and delivery, take significantly more time than providing
the same medications in original packs.
Best practice is defined as a technique or methodology that, through experience and
research, has been proven to reliably lead to a desired result. The QMC group (funded
by the Australian Government via the 3rd Community Pharmacy Agreement Research
and Development Grants managed by the Pharmacy Guild of Australia) has conducted
observational research and sought extensive feedback from a range of key
stakeholders from the medical, nursing and pharmacy professions, as well as industry,
consumer and government sectors to identify the factors that are contributing to unsafe
practices, reduced effectiveness and inefficiencies in the provision and use of DAAs.
Key barriers to safe, efficient and effective provision and use of DAAs in RCFs include:
Poor communication and the breakdown of information flows between parties
A lack of monitoring and accountability for the quality of the DAA services,
A lack of mutual awareness among facilities, pharmacies and doctors about the,
existing profession-specific regulations, experiences and perceptions, practices
and costs involved in the provision and use of DAAs
A shift in the workload, cost and professional responsibility associated with
medication management from the facility to the community pharmacy.
4.6.2 PREAMBLE
4.6.2.1 Purpose and scope of this best practice model
There are currently a range of standards and guidelines relating to the supply of
medicines to residents in RCFs and to the provision of dose administration aids by
pharmacy. These include: Australian Pharmaceutical Advisory Council (APAC, 2002)
medication management guidelines), various Nursing Guidelines, Professional Practice
Standards (Pharmaceutical Society of Australia, 2002), Dose Administration Guidelines
(Pharmaceutical Society of Australia, 1999), Quality Care Pharmacy Program (QCPP)
professional practice guidelines and DAA specific manufacturers/ suppliers guidelines
(i.e. Webstercare, Douglas, Persocare, AHPS, MPS). However there is no existing
resource that accounts for the multi-disciplinary involvement in the provision and use of
Quality Medication Care Group, School of Medicine, University of Queensland81
DAAs and includes specific processes or procedures for optimising the provision and
use of dose administration aids for RCFs.
12. Medication receipt
13a. RCF stores medications (packed & non-packed)
13b. Self-medicator residents store medications
13c. Counsel self-medicators on medications (including CMI, MIC)
14. Administer medication (packed & non-packed) 15a. Self-medicators take medications (checked by RCF)
15. Record administered medications on RCF chart
16. Return unused medication to pharmacy• Pharmacy records & monitors missed doses,
taking action as needed; deals with medications
1a. GP writes medication order• On medication chart in RCF as order to administer
1b. GP writes prescription as order to dispense
3. Develop or update pharmacy-held medication profile• Review regimen, prepare dosing schedule (what to take when)• Develop packing plan (identify what is packed & not packed)
2. Transmit drug regimen information to pharmacy
3a. Support activities• Manage prescriptions
(e.g notify when new script due)
• Accounts, adequate stock on hand
4. Dispense prescription
8. Pack medication according to profile
9. Check packed medication
5. Check dispensed prescription
7. Prepare for packing (labels, assembling medications etc)
6. Store dispensed medications not yet packed
10. Filled packs stored in pharmacy
11. Deliver / collect medication
No
n-p
ac
ke
d m
ed
icati
on
s
Potential RCF role
Doctor Doctor & RCF Pharmacy support staff e.g. Dispensary technicianPharmacist RCF
RCF & pharmacyKey
Patient Potential RCF role
Doctor Doctor & RCF Pharmacy support staff e.g. Dispensary technicianPharmacist RCF
RCF & pharmacyKey
Patient
Figure 4.8 Steps and input required in the provision of DAAs by community pharmacy
to RCFs
Quality Medication Care Group, School of Medicine, University of Queensland82
While profession-specific guidelines are needed, there is also a need for an over-
arching integrated model. This proposed model takes into account existing professional
standards and relevant legislation, and makes recommendations to address the
limitations in the provision and use DAAs, to optimise the effectiveness of this
intervention and to facilitate interaction between health professionals. Despite this
teamwork approach, many of these recommendations may appear pharmacy centric.
This is because much of the resource and cost burden arising from the provision of
DAAs is borne by community pharmacy. However, optimal provision and use of DAAs
cannot occur without multi-disciplinary involvement and the recommendations
emphasise the roles and responsibilities of the different stakeholders. See Figure 4.8,
where the key indicates who generally performs the various steps.
The scope of these recommendations is restricted to issues relating specifically to DAA
use where it is different from medication management using original packs. The key
issues in providing medications in a DAA rather than original packs are:
(1) The pharmacist needs to know the resident’s complete and current drug regimen
rather than providing only a single new or changed medicine.
(2) The pharmacy instead of the facility takes responsibility for managing prescriptions
and ensuring the continuity of medication supply.
(3) The extra steps involved in producing DAAs and the interdependence of the
stakeholders increases the potential for errors at any point in the process and by
anyone involved.
The purpose of this document is not to dictate or set standards in medication
management and the use of dose administration aids. Instead, the goal is to provide a
range of strategies and tools to overcome the barriers to safe, effective and efficient
DAA provision. The following recommendations are interrelated and reflect the need for
further emphasis to be given to the process of tendering and contracting DAA services
as well as monitoring the quality of DAA services. Figure 4.9, shows the issues for
RCFs and pharmacies to consider in tendering/implementing for a DAA service and a
potential means of introducing quality assurance through existing accreditation
programs.
This best practice model is based on the current situation whereby no subsidies are
available for the provision of DAAs to RCFs. Were public funding to become available
in the future for DAA provision to RCFs, the elements of the model (specifically 1.3.2
and recommendation 1.3.5) could be incorporated as part of a system of checks and
balances that would be necessary to ensure accountability with respect to the quality
and quantity of DAA services offered to RCF residents. The model concentrates on the
nature of information transfers and processes rather than a specific mode (e.g. fax,
encrypted email, etc). Future IT developments (e.g. communication channels, software,
encryption and public key technologies) may well be enablers to further increase safety
and efficiency.
Quality Medication Care Group, School of Medicine, University of Queensland83
RCF• Assess pharmacy as provider of a
safe, effective & efficient DAA service*
• Address other relationships (e.g. GPs) to support DAA service
• What internal RCF processes are needed to integrate with process
• Any special needs• Have realistic expectation of fee for
service
Community Pharmacy• Consider how resources, & costs
of providing service affect ability to provide DAA service
• Define core DAA service components
• What RCF responsibilities & internal RCF processes needed
• Negotiate extras including support services, interactions with other (e.g.GPs, residents)
RCF-Pharmacy DAA Service Agreement should include:• Responsibilities for maintaining current pharmacy profile• How the service will work including: pack type, packing cycle,
responsiveness, exception procedures• Procedures for medication changes & prescription management• Other services the pharmacy will provide• Costs and account issues• Agree on key performance indicators (KPI) for pharmacy & RCF,
& their measurement
Tendering process
Contracting
Monitoring, QA of ongoing operation
External assessment via Aged Care Assessors check of key processes for links with pharmacy
RCF audit program of:• Internal processes associated
External assessment via QCPP assessor check of key processes for links with RCF
External check of service quality
*What to look for in a DAA service:
• Has pharmacy has been assessed externally for quality (i.e. QCPP)?
• Have own internal QC & QA programs?
• Are RCF needs (i.e.pack type, hours of service, problem resolution) met?
• Does service integrate RCF processes & internal procedures or special needs (e.g.use of non-RNs in medication rounds or support with chart production etc)
Figure 4.9 Implementation model for the provision of a DAA service to an RCF by a
community pharmacy
4.6.3 RECOMMENDATIONS
4.6.3.1 Residential care facilities tendering for pharmacy services
Residential care facilities need to evaluate a DAA supplier in terms of ability to
deliver a safe, effective and efficient DAA service.
Community pharmacy needs to negotiate a mutually acceptable price for
providing DAA services to RCFs.
Facilities and residents/carers and families need to address the issue of who
should pay for DAA services.
The provision of DAAs to RCFs is cost-effective in that there is a reduction in total
costs to facilities when DAAs are used compared to original packs (Ientile et al. 2004).
Quality Medication Care Group, School of Medicine, University of Queensland84
The proportion of cost borne by the pharmacy, however, is much greater as the
provision of DAAs is labour intensive and costly to pharmacy. Community pharmacy
services and DAAs are highly relied upon in RCFs, facilitating in the minimisation of
medication administration costs, staff costs and meeting accreditation standards.
However, the provision of DAA services is undervalued by facilities with just over half of
the pharmacies providing this service to RCFs being paid. The low rate of payment is in
spite of the fact that RCFs tend to report a high level of satisfaction with DAA services
and community pharmacy in general. This problem is compounded by pharmacies
discounting their services in order to win and/or retain RCF contracts. Providing DAA
services to facilities that are unwilling to pay for this service constitutes a waste of
pharmacy staff time and efforts which could have a greater impact on outcomes for the
community and be better rewarded if re-directed to other services.
The quality of DAA services impacts on savings to facility in terms of increased
efficiency in medication management, and the satisfaction of staff and doctors. Not all
DAA services are the same. Facilities need to consider whether the pharmacy is able
to provide a quality service. This involves considering whether the pharmacy has been
assessed externally for quality (i.e. QCPP), whether the pharmacy has their own
internal QC & QA programs, whether the pharmacy can offer a service that will meet
the needs of the RCF (i.e. type of pack, hours of service, problem resolution) and be
able to integrate with RCF processes and internal procedures (i.e. consider any special
needs such as use of staff other than RNs for administration, and additional support
required with charting etc). RCFs also need to have a realistic expectation of fees for
services provided.
Community Pharmacy need to consider what resources are required to provide this
service, what it will cost to provide this service and the impact it will have on business
profitability. For example, providing DAA services to 120 residents at two facilities
required approximately 22 hours of staff time per week (Ientile et al. 2004).
Facilities and residents/carers and families need to address the issue of who should
pay for DAA services. Residents have a right to choose which pharmacy to use
however facilities tend to discourage this practice by stipulating that the resident must
have medication packed into a certain type of DAA pack. Given this stipulation and that
the use of DAAs in an RCF is primarily to reduce administration costs for the facility
rather than to benefit the resident (i.e. administration from original packs by a RN would
not negatively impact on a resident) it would be reasonable to expect facilities to fund
this service. Potentially some of this cost could be off-set through resident contribution
and/or government funding.
4.6.3.2 Contracts specifying obligations and promoting mutual awareness
Community pharmacy should negotiate a written agreement between
themselves, the residential care facility and/or the resident and the residents’
doctor/s prior to beginning to provide a DAA service.
The provision of DAAs constitutes the primary interaction between community
pharmacy and the facility in terms of resources (pharmacy time and cost), however the
available templates for contracts between pharmacy and RCFs tends to be limited to
only the cost of the DAA service and additional pharmacy obligations, and not the
obligations of other parties. A detailed agreement between the pharmacy and facility
Quality Medication Care Group, School of Medicine, University of Queensland85
and the facility and the residents’ doctor/s provides a means of specifying obligations
and promoting mutual awareness.
The agreement should address:
That it is important for the pharmacy to know the correct, current medication
regimen so that a DAA containing the correct medications can be provided on time,
i.e. the importance of pharmacy profile maintenance.
How the service will work:
What type of pack will be provided (brand, period, how many packs needed,
how many weeks’ worth will be provided at once).
What medication will be packed (including whether some or all non-prescription
items such as complementary or vitamin supplements are to be packed) and
what is the optimal schedule (see 1.3.3),
How far in advance of distribution will the DAA be packed (and in some cases,
which day of the week) (i.e. what is the packing cycle?)
Where will original pack medication and prescriptions be stored?
What will the delivery procedures and schedule be?
What is the cost of the service, who is the payer and how the pharmacy will
charge for the service?
What will happen if a medicine in a pack needs to be changed before the
current pack is finished.
What is the cost of the service, who is the payer and how the pharmacy will
charge for the service.
Procedures for start-up and preparing the first pack e.g. will medicines already
dispensed and brought in by the resident be packed?
Privacy: Resident consents to information sharing between the facility staff, doctor/s
and pharmacy (and hospital if required)
Medication changes:
Whose responsibility will it be to inform the pharmacy of medication changes
(facility or doctor)?
What format will be acceptable, (i.e. written instructions only- amended to full
regimen indicating ceased/changed and new drugs, copy of full chart)?
How long will ceased medication (already dispensed to resident) be stored at
the pharmacy?
What constitutes timely notification of changes so that rework is minimised.
The circumstances when a medication change can be delayed until the next
pack or when a change needs to be made more quickly.
Prescription management:
The understanding that a valid prescription is required before medicines can be
supplied in a pack, and that, depending on the packing cycle, the date a
prescription is needed may be earlier than anticipated with original packs.
Will the pharmacy remind the facility to request the doctor visit the resident to
obtain a new prescription?
Will the pharmacy generate prescription reminder notices for the doctor/s?
GP practices related to prescriptions and repeat reminders e.g. under what
circumstances is the GP willing to write a prescription without seeing a resident.
Who is responsible for following up that the prescriptions are at the pharmacy in
time for packing (facility or pharmacy or doctor or other).
What will happen if resident has no new prescription? (e.g. medication not
packed in the DAA).
Quality Medication Care Group, School of Medicine, University of Queensland86
Procedure for use and what to do in exceptional situations e.g:
When things go wrong (i.e. what to do if facility staff/residents drop medication
or open the wrong blister).
What to do if a resident has a prescription filled at another pharmacy (not the
one that provides DAAs) e.g. hospital pharmacy
When a patient is away from the facility e.g. on holidays with family or in
hospital, where there might be a need for more packs than usual or where there
is an interruption in the use of packed medications (e.g. hospital admission or
discharge).
How to ensure optimal use of the DAA:
How the pharmacy will provide counselling and medication information (e.g.
CMIs) to the residents and facility staff.
Facility agrees to return unused medication in the DAA to the pharmacy.
Pharmacist reviews returned DAAs to assess compliance and problems with
use.
Pharmacist counsels staff/resident if problems occur.
Other services the pharmacy will provide (i.e. participation on medication
management committees, assistance with accreditation, medication reviews).
The process of completing an agreement of this nature ensures that the facility and/or
resident receives adequate information about what is involved in receiving a DAA
service and training in how to use the DAA and what do if problems occur. In addition,
this agreement allows the pharmacy, doctor and facility to negotiate who will be
responsible for undertaking the necessary activities to ensure the accuracy of the DAA,
the continuity of medication supply and the optimal outcome for the resident. The
agreement also allows the GP’s expectations and practice patterns, policies and
procedures to be recognised.
This agreement may also provide the pharmacy with some form of risk-reduction in
relation to potential future claims arising from situations where a patient/relative makes
a claim against the pharmacy.
4.6.3.3 Communicating medication changes
The pharmacy and the facility should maintain a written record of all
communication regarding medication changes for residents receiving DAA
services.
The accuracy of the DAA pack is largely dependent on the currency of the pharmacy
profile. Maintaining a current medication profile requires full and detailed
communication about the medication regimen and any changes. In the RCF setting, the
responsibility for informing the pharmacy of medication changes generally falls to the
RCF staff. The facility or the doctor presenting the pharmacy with a new prescription
does not constitute adequate communication as the pharmacist may be uncertain of
whether this new medication is in addition to the current regimen or to replace another
medication.
The pharmacy should request that all medication changes be communicated in writing
(whether by fax, email, letter etc). The communication should reflect changes in the
context of the patients’ full regimen and should be unambiguous and detailed enough
to allow the pharmacist to alter the medication profile and the DAA profile without
Quality Medication Care Group, School of Medicine, University of Queensland87
further consultation. An updated copy of the residents chart should fulfil this
requirement. Written records and documentation of medication changes should be
maintained by the pharmacy and the facility to ensure the accuracy and accountability
of DAA services.
4.6.3.4 Continuity of care between hospital and RCFs for patient with DAAs
Each community pharmacy, RCF, hospital and doctor treating RCF residents
should be aware of their role and develop procedures to meet their
responsibilities to ensure continuity of care between hospital and RCFs.
When a resident of and RCF uses a DAA, the issue of ensuring continuity of care on
discharge from the hospital back to the RCF is more complicated and requires greater
collaboration between the hospital and community health care services (RCF, general
practitioners and community pharmacy). Stakeholders all have roles and
responsibilities in ensuring continuity of care for DAA users (see Figure 4.10). To
improve practice, the needs and perspectives of all stakeholders should be considered.
A failure to consider the policies and preferences of others involved in patient care can
lead to wastage and interruptions to the continuity of medication supply.
Figure 4.10 shows the steps and inputs required to facilitate the supply of medicines
packed in a DAA after discharge from hospital. When a patient is admitted to hospital
from a RCF, irrespective of whether they have their medicines in original packs or
DAAs, the hospital needs reliable information on the patient’s medication regimen, and
this is achieved through the RCF faxing a complete (including prn medication) and
current copy of the residents chart. Because the use of a DAA complicates discharge,
on admission, a hospital pharmacy (or the ward staff where hospital pharmacists are
not available) need to know firstly that a patient is a resident of a facility that uses
DAAs and who to liaise with (which facility, GP and community pharmacy). In addition,
the hospital pharmacy needs sufficient notice from the hospital medical staff to allow for
the necessary liaison. To pack the DAA, the community pharmacy needs timely,
reliable and complete information on the patient’s current drug regimen (the discharge
medication regimen) and a valid prescription for payment and/or a sufficient supply of
medication. To maximise safety and to minimise wastage while ensuring essential
medications are available, the hospital pharmacy should liaise with the community
pharmacy about what medications to supply on discharge. For example:
The facility staff/resident may be confused if the hospital supplies a different brand
of an item than is usually packed so that no discharge supply may be safer option.
The patient may be prescribed a drug that is not routinely available in the
community (e.g. a trial drug) so that discharge supply of this item is necessary for
continuity of supply.
On discharge the RCF staff need a legally valid and reliable order to administer new
medications. There can be delays in getting this order because of the timing of
discharge and poor information flow to GPs.
Quality Medication Care Group, School of Medicine, University of Queensland88
RCFpatientusing DAA
RCF to fax patient chart & send DAA +other medication to hospital
Hospital compile medication history/profile:
• Requires reliable & current information
• Sourced from RCF, GP & community pharmacy
• Includes ADRs, allergies, DAA use
Patient receives treatment including changes to current medication regimen
Remind discharge doctor that sufficient notice of discharge is required to organise DAA
Contact community pharmacy to inform them that patient has been admitted, to withhold packs pro tem & new pack will be required. Ascertain pharmacy preferences for discharge procedures
If community pharmacy is a new contact add details to hospital pharmacy DAA database including preferences for discharge medications or prescriptions, packing schedule & fax number Medical staff inform
hospital pharmacy that patient will be discharged
Community pharmacy to pack DAA from the hospital discharge list
Community pharmacy delivers DAA pack
RCF receives DAA in time for next doses OR uses hospital supply until DAA arrives
Adm
ission
Durin
g h
osp
ital s
tay 2
4hrs†
prio
r to d
ischarg
eD
ischarg
eP
ost d
ischarg
e
Fax/email* a copy of hospital prescription and discharge summary to community pharmacy
Check pharmacy database. Contact pharmacy to confirm what pres-criptions &/or medicine is required & when pack will be available
Quality Medication Care Group, School of Medicine, University of Queensland106
The importance of correct, up-to-date medication regimen information.
Prompt information flow about medication changes from the GP to the pharmacy.
Procedure for exceptional circumstances as detailed. This could be expanded to
address the potentially prolonged turn around times for adjusting/changing a DAA
when a patient’s medications have been changed while in hospital. In some regional
areas, this can be up to 3 days.
Ensuring optimal use of DAA.
There were some reservations:
The costs of sorting out the legal issues in an agreement; “if you had to discuss the
agreement with a solicitor, how much would that person charge for their time to
cover all the clauses?”
Arranging for an agreement to be signed off may take time, and result in a delay of
service following hospital discharge.
Suggested changes to the agreement were:
“An agreed communication strategy to ensure community pharmacist and GP is
informed of the patient’s discharge medication (including any changes and reasons
why) in a timely manner”.
“As we are all partners in this patient’s healthcare, consider sending an updated
copy (when the 6th monthly concordance check is done) of the patient’s current
medication profile to the hospital for their records.”
To explicitly address practices related to narcotics/opioids or other schedule 8 or
recordable drugs.
Details about how far in advance DAAs are prepared and the day of the week and
where packs are stored was felt to be a little too involved.
Who had responsibility for keeping all the ‘paperwork’ was not defined in the model.
One hospital pharmacist said that “leaving out items in the agreement leaves the
parties open to disagreement/dispute”. Another hospital pharmacist indicated that the
issue of medications dispensed by other pharmacies not being repacked into a DAA
(and so potentially wasting the medicine, money and time) needed to be addressed,
but rather than including this into the tripartisan agreement, the pharmacist felt a
definitive statement by Medicare Australia and state health departments, etc, was
needed.
The respondents to the expert panel questionnaire were generally supportive of the
intention behind the tripartisan agreement recommendation i.e. ensuring all parties are
aware of what is involved in and preferences related to a DAA service before beginning
the service. When asked which aspects of the agreement were considered
important/useful, nine of the 15 respondents felt that all aspects were useful. Aspects
specifically mentioned were:
That it formalized the process and increased GP awareness/understanding about
how the service generally worked (4 responses); “By increasing the involvement of
the GP’s in a formal process, pharmacy is forcing them to accept responsibility and
educate them on the processes involved”. Increased understanding by the
patient/carer and expected goals/outcomes would also be addressed by such an
agreement (3 responses) and so prevents any misunderstanding.
“The most frequent source of error is poor communication and setting up the
correct structure from the start is essential. This understanding would set out in
advance the responsibilities of all parties based on a common sense approach”
Quality Medication Care Group, School of Medicine, University of Queensland107
How the service will work (6 responses) including what the packing cycle was (2
responses), what medications are to be packed (plus whether non-prescription
medicines will be packed) and costs.
Prescription management procedures (6 responses) including: GP preferences for
writing repeats or writing owing prescriptions (3 responses, including when a patient
needs to see the doctor); pharmacy policies/procedures on ‘script owing’ issues
(e.g. what happens when there is no current prescription at the time of packing).
Procedures for medication changes (2 responses).
How optimal use of the DAA was to be ensured including information/counseling to
be provided (2 responses).
Importance of pharmacy knowledge of a patient’s current drug regimen (2
responses).
Policies of pharmacy and GP related to S8 and authority prescriptions.
Start-up procedures.
Procedures for what to do in exceptional circumstances (felt to be especially useful
for carers).
Consent of patient.
Several respondents seemed to interpret the agreement as a more generic document
that might include with ‘tick boxes’ rather than a service agreement for a specific
patient. For example, when asked what elements of the agreement should be left out,
one respondent (a pharmacist) indicated that delivery details could be left out as “these
may alter from patient to patient”. Three respondents suggested standard templates or
tick box forms as means of operationalising the agreement. Having different documents
for the different 2-way interactions was suggested by two respondents:
“While it is important for patient/carers to understand what is involved in the
agreement – the sort of understanding required is at a broad level and some of the
more specific intricacies related to service provision (e.g. issues between
pharmacist and doctor such as circumstances under which GP is willing to write a
repeat prescription without seeing the patient) would only serve to confuse them
and would be better left out. Mostly, they need to know in plan English what they
have agreed to do”
Other changes to the elements of the agreement suggested were:
Procedures for exceptional circumstances do not need to be written (2 responses)
but advise patient to contact the pharmacy if something goes wrong.
Include the role/responsibility of parties in relation to the patient held template.
Consider how agreements would work when patients change doctors.
Change the phrase “consent to re-assess” to “consent to formal review”. This could
be done via a practice standard that requires direct discussion with patient.
Strengthen the statement about referring for an HMR “these are the patients at risk
of medication misadventure and therefore most likely to benefit from an HMR”.
“Throughout the model ‘education’ needs to be better described and articulated. I
can imagine that not all pharmacists would necessarily have a good understanding
of how to do this well and consistently and this does not provide them with the
means to do so”.
Include community nurse where relevant as a party to the mutual understanding.
Recognising the limitations under which a carer operates e.g.:
“Privacy issues related to the carer accessing information about the patient that
might need to be sorted through so that the carer is not ham-strung in this
process where they have responsibilities but few rights”
Quality Medication Care Group, School of Medicine, University of Queensland108
“Making carers responsible for informing the pharmacy of medication changes
would assume that carers themselves are informed and I can assure you that
this is not always the case. You can’t necessarily expect a carer to perform a
role that they may not, through no fault of their own, be able to deliver on”.
The articulation of responsibility for continuity of prescription and prescription
management also raised comment because pharmacists often undertook these types
of tasks in an attempt to improve efficiency and ensure that legal requirements are met
when the responsibility and ultimately, prescription writing was outside their control.
The agreement was felt to highlight this responsibility to doctors but alternatives were
suggested:
“Responsibility for notifying GP for new prescriptions is OK but the pharmacist
cannot be responsible if the doctor is not forthcoming. This can have serious impact
on health care and is beyond the pharmacist’s ability to control. This has to be made
clear so the pharmacist is not held to blame. Therefore a simplified means of
ensuring continuity of treatment e.g. the medication profile signed by the doctor
could serve as the primary documentation for a prescription and payment for supply
or emailed PDF prescription”.
There were four main areas of contention or concern: (1) the extent to which a GP
needs to be involved with the agreement; (2) the extent to which GPs would participate
in such an agreement; (3) the use of the word “agreement” and any legal/contractual
implications associated with the term; and (4) costs.
Two respondents interpreted the agreement as one that required GPs to act as a
gatekeeper for service access. This was not felt to be feasible and would unnecessarily
complicate the process. A gatekeeper role for GPs was not the intention for the
agreement, rather the intention was to open communication, increase awareness and
to be explicit about roles and responsibilities of stakeholders in the service. Other
respondents were concerned about the willingness of GPs to be party to an agreement,
particularly without a funding model for this activity.
“While a multidisciplinary approach is optimal, it cannot be contingent on all parties
agreeing. As long as one health professional determines the need (whether
pharmacist/carer/domiciliary nurse), and that need is documented, it should proceed
with the patient’s permission. Many GPs will not be bothered filling in another form”.
One respondent predicted that the only way such agreements would exist “in black and
white and not just verbally is for it to be tied together by reward and punishment”.
Use of the term “agreement” was felt to have contractual implications. One respondent
wrote that doctors “worry about the medico legal aspects of being bound by third
parties”. An alternative title without the legal implications was suggested as a “Tripartite
Communication/Obligation Framework”, and the phrase “The agreement should
address” would be “Areas/issues to promote awareness/understanding and
communication”.
Funding was felt to be a critical element for success or adoption. Costs were
associated with GP involvement and at the pharmacy level; “It takes time to establish
these agreement criteria so some funding should be allocated to cover the time and
expertise required to arrive at agreement”. One respondent estimated that each
agreement would 20 minutes if initiated at pharmacy level, therefore there was a need
Quality Medication Care Group, School of Medicine, University of Queensland109
to introduce an establishment fee. Other pharmacy costs would be related to staff
training on initiating discussion, completing form and faxing to GPs, etc.
5.2.2.3 Patient held medication template
A recommendation for the establishment of a template for medication packing
(including packed and non-packed medications) was included in the best practice
model. This template was to be carried by the patient and used by the GP to record
medication changes. Processes to ensure that the template was current were also
included. A draft template was prepared to support this recommendation.
Of the 11 community patients providing comments, 10 said that their pharmacy
currently provided them with a full list of their medications. In addition to drug regimen
information (drug name, directions and a reminder of when to take the medicine +/- the
indication), often included in products like “MedProfs”, the draft best practice template
included details about the pack type, the packing schedule, contact details for the
pharmacy and GP plus adverse events and recent medication changes. It also acted
as a communication tool between the GP and the pharmacy, and the community and
hospital. Eight of the community patients said that they would find the best practice
medication template helpful with respect to improving their memory and keeping family
members/carers informed of their medication status. Nine patients were willing to take
the medication template with them to their doctor, pharmacy and hospital.
All 7 community pharmacists felt the patient held medication template would be useful
to them in their practice while 6 felt the medication template would also be useful to the
patient (1 did not answer saying that the template would only be useful to some
patients but that much of the information is already on the pack so many use this
information). For pharmacists, the template would be useful:
To improve continuity of care.
For better communication (2 responses) (removes guess work or problems with
faulty communication; accurate communication done in a simple manner).
As the pharmacy had an accurate record of what is supposed to be in the pack.
As it would help reinforce to GPs that it was their responsibility to inform pharmacy of
changes to the template to ensure it is current.
Only if the pharmacy also held a copy.
For patients, the template would be useful to improve continuity of care (3 responses)
(the copy could be used at hospital, checking with any GP or an after-hours service; a
link between all health care professionals) and so the patient has an accurate record of
what is supposed to be in the pack.
Pharmacists were asked about advantages and disadvantages of the doctor approving
the template (or pharmacy packing profile) prior to the pharmacy packing the DAA
(Table 5.3). The advantages in clarity of communication were balanced against
problems with workflow were the GP to not approve the template in a timely manner.
Some minimal changes would be required to implement this recommendation (in the
situation where the pharmacy would produce a template) in some of the respondent’s
pharmacies:
No change in one pharmacy.
Impact would be small. Two pharmacies already had a template. The change would
be to provide a copy to the patient and doctor.
Quality Medication Care Group, School of Medicine, University of Queensland110
Organise a template for the patient to keep with them and 6-monthly updates. Cost
would be minimal.
Others would need to make more changes:
One pharmacy felt there would be limited impact unless the pharmacy had to provide
the template.
One pharmacy did not have templates and felt that templates would be more
paperwork and take time “chasing up GPs to provide templates”.
Table 5.3 Advantages and disadvantages of GP approval of a packing template prior
to packing
Advantages Disadvantages
Accuracy and ease of packing Delays if doctor is away Should be mandatory, we already require this The medications should be correct and puts legal responsibility on to the doctor
Time for the doctor to do the template may hold up process
Medications that have been ceased by the doctor but not by the patient would be obvious Save time in the communication between parties involved Confirmation of drug regimen Problems if doctors did not update the template
Three community pharmacists felt that implementing the best practice medication
template would be feasible. One of these pharmacists cited patient reluctance to have
the extra paperwork or responsibility as a barrier; another felt that the staff time was a
barrier but one which could be overcome with government remuneration while the third
wondered whether doctors would participate. Three other pharmacists felt that the
recommendation was feasible on condition that the GP co-operated. Another
pharmacist felt that a recommendation that involved GPs being prepared to do extra
work was not very feasible; “I see pharmacists trying to chase templates as well as
scripts”.
Five pharmacists explicitly mentioned benefits from this recommendation in terms of
the safety of the DAA system and Quality Use of Medicines (QUM). The pharmacist
who felt the recommendation was not very feasible indicated that it would beneficial if it
worked. The only improvement to the recommendation suggested was to combine the
doctor’s agreement to use the template with any agreement to write prescriptions for
pack patients (as per the tripartisan agreement).
The GP providing comments on the community model felt that a template would be
useful to the patient only if it was shared between all carers (including pharmacy and
GP) with the patient’s knowledge and consent. From the GP’s own perspective, the
template would be a useful aid in patient education. The GP felt there were advantages
in being able to review medications before packing and to coordinate this with a patient
visit to review medications and educate the patient. The GP liked the draft template
indicating that “it could be stored … and easily populated by the computer extracting
data from the script list. It could be sent with all script requests from the pharmacy to
the GP and the GP would then have a reliable update for the patient’s file and could set
up reminders for authority script due dates”. Disadvantages of the template from the
GP perspective also related to the GP not always being available, but there was also a
possible cost to the GP of patient non-attendance and the paperwork involved.
Quality Medication Care Group, School of Medicine, University of Queensland111
All but two of the hospital pharmacists commenting on the community best practice
model felt that a patient held template would be useful to the hospital pharmacist to
communicate medication changes to the community pharmacy. One pharmacist
indicated that it filled the information void between the hospital and community in the
absence of a system of electronic information transfer, but warned that the likelihood of
the template being completed was an inverse relationship with the amount of
paperwork required. Several pharmacists indicated that the template would be useful
when patients were admitted (6 responses), provided the template was sent with the
patient and the information was current. It saved ringing the GP or pharmacy to
establish current medications, or at least was a useful start. However, one pharmacist
felt that the medication history reviews performed in hospital were easier and
potentially more accurate/current. Another pharmacist wondered how the process
would be managed if the patient had more than one healthcare provider (e.g. GP and
specialist).
One hospital pharmacist indicated the templates were not feasible because there was
always a problem with them being up-to-date. Another pharmacist from a smaller
regional hospital felt the template not to be feasible and that communication of “full and
detailed” medication regimen information was a “luxury” at discharge as discharges
were often done on an ad hoc basis, for example, when the emergency ward was full,
so “who can we discharge”, often patients with complex regimens.
Suggested changes to the recommendation to make it more useful to hospital
pharmacists were:
Supplying a copy of the medication chart for each DAA refill.
Asking the community pharmacist to also send an updated copy to the hospital (with
patient consent).
Despite the potential usefulness of the template, there were reservations about its
feasibility from a hospital pharmacist perspective:
Engagement, integration and participation of stakeholders was needed for
successful uptake (2 responses). It would only be feasible if all stakeholders (patient,
healthcare practitioners (hospital, community and specialists)) participated in the
work practices changes and collaborated. Political drivers would be needed to
incorporate an idea like the template as a communication tool into hospital practice.
The requirement for the patient to have a responsibility to remember to take the
template to hospital was a possible barrier (2 responses). Community pharmacy-held
and provided records were more feasible as there was patient-to-patient variability in
medication knowledge and motivation to provide records.
There were fundamental barriers in current practice like the doctor or community
pharmacist not updating the template or the patient losing it.
One pharmacist from a smaller regional hospital indicated the best practice model
recommendation was not feasible because of lack of staffing and a “medical culture
that is overworked and plays down the significance of explaining changes to
medication”.
The new software needed and additional costs were barriers.
As with other stakeholders, there were mixed views about the patient held template
among the 14 respondents to the expert panel questionnaire. A patient held medicine
template was felt to be useful or helpful to:
Community DAA users by 10 respondents; 3 respondents agreed conditionally.
Quality Medication Care Group, School of Medicine, University of Queensland112
Pharmacies providing DAAs by 11 respondents; 2 respondents agreed conditionally.
GPs caring for patients who use DAA by 10 respondents; 2 respondents agreed
conditionally and one disagreed.
There was support for the concept of the patient held template, with 7 respondents
providing strong support. Reservations about the feasibility of a patient held template
related to the ability of the patient to carry out their role in ensuring the template was an
effective communication tool and the willingness/co-operation of the GP to complete or
use the template.
To make medication regimen communication patient dependent “is to invite
disaster. Communication of medication changes must be between health
professionals”.
“Patients are notorious for losing records and whilst for some this may this would be
very effective, the reality is that one of the types of patient who usually benefit from
a DAA are the cognitively impaired and you could not rely on them to carry this
information with them.”
There is a risk that GPs would not be willing to do the paperwork “Just look at
HMRs and the way doctors boycott the process because they are not willing to do
the paperwork required”.
In relation to the community patient implementation model (Figure 4.6) “the
willingness of GPs to check a DAA template in sufficient detail to be useful [to the
pharmacy], would be limited especially as they are not being paid for this activity”.
Two other concerns were that:
“With this written template, doctors/pharmacists may be LESS inclined to TALK to
people about the medication list. Patients/carers need to be able to actively engage
and clarify their understanding, discuss their concerns/issues”.
Difficulties might arise “when the DAAs are provided from a source external to the
patient’s community pharmacy e.g. on a contract arrangement” and the community
pharmacy is required to provide the patient held template.
One respondent said that the concept of the patient held template would be beneficial
particularly were a standard approach be used and would only be feasible if there were
mechanisms to ensure that this standard approach was carried out (e.g. QCPP
assessment).
Opinions on the advantages and disadvantages of doctors approving a patient’s
medication template/pharmacy packing profile prior to the pharmacy packing the
patients medicines into a DAA are shown in Table 5.4. Prior GP approval of the
template was not included in the preliminary best practice model but opinions as to
whether this should have been included were sought. Two respondents completely
disagreed with this suggestion e.g. “process cannot be contingent of GPs” and others
identified significant barriers to initiating a DAA. The main advantage of GP prior
approval of a profile before a DAA is packed was that the drug regimen was likely to be
more accurate and reliable (so avoiding re-work associated with incorrect medication
regimen information).
Quality Medication Care Group, School of Medicine, University of Queensland113
Table 5.4 Expert panel views on the advantages and disadvantages of GP approval of
a packing template prior to packing
Advantages Disadvantages
Would not happen without software/technology enabling linking GP and pharmacy medication records
Assumes doctor has accurate records (many don’t find time to get own records in order) Getting GP to do this unpaid task in a timely manner
Requires commitment from GP (2 responses). “they are unlikely to uniformly comply regardless of patient benefit. Will only be perceived as impost by pharmacy, especially if pharmacists are funded for DAA packing and they are not”
Includes GP in DAA decision-making process Alerts GP to the number of OTC preparations taken by the patient Educate GP on process of preparing DAAsImprove GP-pharmacy communication Accuracy of profile Opportunity for pharmacy to secure prescriptions needed to pack
Time taken; the process must be quick and simple
Need GP input for accurate profile Useful for direct communication between GP and pharmacy
Repeated contact with GP for confirmation would interrupt GP practice Risk if no source data to ensure accurate profile
Legal protection for pharmacy Up-to-data medication regimen information
Reduces errors from assumptions and out-of-date information
Additional task for already busy doctors Doctors would want to be paid for this task
GP awareness that patient has a DAA improves appropriate communication with the pharmacy when medication is changed
GP difficulties in finding time to complete paperwork
Have a common reference point for a persons medication regimen
Time taken and delays in getting access to the GP – “What are the risks to the patient with the need having been established but waiting to be actioned” Good document handling would be required
Three respondents indicated that there were likely to be additional costs involved in
using patient held templates (including staff time contacting doctors for confirmation,
faxing of templates to doctors, tracking paperwork and entering data, and costs for GPs
and other parties involved).
“It takes time of course to keep an update profile, however as this is possibly the
only complete profile it is worth the effort as it is also likely to identify non-
prescription medication being taken as well”.
The development and use of information and communication technologies were seen
as a way of minimising the time and costs.
Quality Medication Care Group, School of Medicine, University of Queensland114
Respondents to the expert panel questionnaire suggested many improvements to the
overall recommendation or ways of overcoming barriers to the adoption of the patient
template:
Change the phrase “this template should be approved by the patient and the
patient’s doctor” in the recommendation text to “endorsed by the patient and the
patient’s doctor” so that it is clear that it is not prescriptive that the GP approves a
template before packing can commence. This would also involved a change of
wording (e.g. to “template endorsed by all”) in Figure 4.6.
One respondent suggested that the recommendation needed to include strategies
that can be used when a patient cannot participate in the information transfer.
Others preferred making the process not dependent on reliable information transfer
by the patient or carer but provide the template to the patient e.g. “If doctors were to
fax/email dose and medication changes directly to the pharmacy then the pharmacy
could fax back the completed document for the doctors records”. A direct link
between the doctor and the pharmacy would be better as:
“Anyone can forget to bring their template with them when coming to the
pharmacy so some additional method would help such as email/post/deliver
copy to pharmacy. Cannot be relied upon as the only method of communication
and source of information. Failure to bring it to the pharmacy could mean a lot
of time spent by pharmacy seeking correct information. “.
Use information communication technology to facilitate effective and efficient
information transfer. “Hard copies would work in the interim with the pharmacist
maintaining the records [but] may be more cost effective if communicated
electronically”. However, there was a need to overcome barriers associated with
the “lack of technology that ‘talks to each other’”. One respondent suggested that a
future “Smartcard, if introduced, would offer the opportunity for the doctor to update,
patient to carry and pharmacist to check and record changes”.
One respondent overcame difficulties in getting GP time to complete the initial
template by providing “a kit to the patient/carer and they make a long appointment
for the Dr to complete the med profile plus all other information e.g. allergies etc.
the Dr is then involved in providing updates for their patient by whatever means
they find easiest”.
Processes need to recognise carers and other health professionals involved. The
community nurse and indeed, practice nurse, have a potential liaison role between
the patient, doctor and pharmacist that could “save pharmacists and doctors much
time and help them focus on their particular areas of expertise”. Carers need to be
kept informed of medication changes that might take place at a trip to the doctor or
pharmacy when the carer is not present. Also, if carer administering medications,
their preferences also need to be considered in determining the administration
schedule.
The model could include strategies for the GP’s practice to streamline information
sharing using the patient held template e.g.:
The GP practice could assign a liaison staff member “to reliably handle this
information”. “Often the [doctor] will provide the regimen but not specify dosage
times”.
“Finding ways of getting that info back into the patients record at the surgery”.
Need to add a process to check concordance between the GP records and the
template as HMR evidence is that GP records do not always match what the
patient does. [Note: the model includes a 6-monthly concordance check].
Quality Medication Care Group, School of Medicine, University of Queensland115
Three respondents made comments specifically related to improving the draft template
itself. The template may not be the optimal layout and could benefit from professional
design and a style guide on how to record information e.g.:
Half a tablet should be written as “half” not “1/2” as the latter could be mistaken as
one to two tablets. “These simple things can minimise confusion and prevent
medication misadventure.
Using plain English explanations for reasons for use (as the respondent anticipated
that this may not be as well done as the examples on the draft template). The
respondent also had concerns about the use of the phrase “urticarial rash” to
describe an adverse drug reaction (ADR). From a health professional perspective,
this term has specific meaning that would be more informative than “a lumpy, itchy
rash” to a prescriber considering prescribing a penicillin-related medication in the
future. Further developments of the template will need to consider the rationale for
including ADR information and the target audience for such information in any style
guide.
The recording of recent medication changes was of concern to two respondents e.g.
“We would have to be careful of how we have worded the ongoing documentation of
pack changes that appears underneath the current chart “. Completion of this section
by the doctor was suggested.
Additions to the template suggested were:
A column for brand name prescribed (to facilitate concordance checking with GP
records) and to change the column name “Brand Name(s)” to “Brand name packed
or dispensed”.
Add a “Notes to pharmacist – Dr only” column for the doctor to indicate changes to
the regimen such as “increased” or “changed dose” for continuing medications.
There would also need to be blank lines for new medications to be added and these
could be identified by the doctor using the “note to pharmacist column”. The
“Recent Medication Changes” section should be reserved for notification of ceased
medications i.e. called “Recently ceased medications”. This annotated form would
then go to the pharmacist who would amend the profile and provide a new template
to the patients.
Medicare number.
Specific space for carer’s contacts “so that this would act as an additional cue for
doctors/pharmacists to involve carers”.
One respondent also suggested including information about the storage of the drugs at
home, things that should be avoided (e.g. alcohol), possible side effects and to clearly
specify a person to contact, for example: in the event of missing a dose, side effects,
etc. Much of this information, however, would duplicate that of patient information
sources such as CMIs and unnecessarily complicate form whose purpose is to
communicate information to allow information on a patient’s current drug regimen.
Apart from models for funding that included GPs, and technology to assist the
communication, suggestions to facilitate implementation were:
Include the provision and use of the profiles as a standard under the Quality Care
Pharmacy Program (QCPP).
Communicate to “all parties the benefits of maintaining such a template in the
interests of safety, legal protection and value in an emergency situation”.
Quality Medication Care Group, School of Medicine, University of Queensland116
Incorporation of the production of the patient template into the dispensing process
so that the pharmacist could generate this quickly when processing prescriptions. A
“tick box to generate & notification of the last date it was generated to avoid printing
every time a script is filled” could be included. Another respondent suggested that
an “industry standard for inclusion in DAA software” was needed.
The establishment of a working party to improve the template and associated
communication.
5.2.2.4 Communication of medication changes
Poor communication of medication changes was a problem in phase 2 and was
identified in focus groups, particularly where the responsibility to communicate changes
fell to the patient. A recommendation to record communication about medication
change in writing was included in the best practice model. This would also act as an
audit trail.
The majority of patients (8 of 11 respondents) had their GP communicate medication
changes to their pharmacy. Only 3 of 10 respondents indicated that their GP
communicated ceased medications to them in writing. All patients were satisfied with
their current arrangement for communicating medication changes.
All of the 6 responding community pharmacists felt that this recommendation would be
feasible, two on the condition that an electronic record or the medication chart
produced by the computer were sufficient. Another felt it depended on the doctors
doing their part. Three pharmacists felt recording medication change information would
be beneficial for a safe, effective and efficient DAA service and 4 felt that medication
change records would reduce legal risk although one indicated that this would only be
the case if the communication was in writing and faxed/posted (i.e. hardcopy).
Pharmacists indicated that for most, few changes would be required:
Nil (2 responses); “we already do this and it works well”.
Minimal cost if the medication chart is printed when change occur.
Setting up systems for fax/email would help as all changes are done in consultation
with the doctor but sometimes just verbally.
Make sure all communication from doctors is in writing.
One pharmacist (who preferred electronic records) indicated that implementation was
likely to be difficult if it involved duplicating systems already in place. Another said that
government funding and education would assist with the implementation of this
recommendation. Overall, to meet this best practice recommendation, pharmacists
generally needed only to introduce the medication template and insist that medication
changes are communicated in writing. Only one improvement to the recommendation
was suggested: to make it mandatory for doctors to write and fax or post changes.
The GP who reviewed the community model indicated that she did not communicate
medication changes in writing. This would only be done if the pharmacy had consulted
the GP beforehand and the patient consented and an agreement had been worked out
beforehand.
The views of hospital pharmacists about communicating medication changes have
been incorporated into section 5.2.2.3.
Quality Medication Care Group, School of Medicine, University of Queensland117
Nine of the respondents to the expert panel questionnaire supported the
recommendation for communication of medication changes in writing as feasible,
promoting safe, effective and efficient DAA services and acting as a legal risk
reduction. In particular, the written communication of changes was felt to assist carers
who would not always be present at a doctor’s visit and yet still be expected to act as a
go between for doctors and pharmacists;
“Carers are often responsible for administering medication at home but are
excluded from the process around decisions related to medications (often on the
basis of ‘Privacy’). A real concern is when medications are cancelled or doses
changed and this is not communicated to the carer”.
Four respondents had some reservations about the recommendation:
“What is being suggested certainly is more complex but some form of a formal
process is certainly required to ensure medication safety and protect all parties”.
As long as there was co-operation from the doctor (2 responses) “Fine if the doctor
accepts that role of providing this information direct to the pharmacist in addition to
a copy going to the patient”.
There were concerns over making the patient responsible to report changes (2
responses) “I would feel very concerned if I relied totally on the patient to inform me
of any changes”.
Several respondents indicated that these practices were already in place with one
writing - “I cannot imagine a pharmacist changing a pack on the direction of the patient
alone so they would already be doing what is required in communication with the GP to
ensure accuracy and to keep risk exposure to a minimum”.
The impacts of this recommendation were felt to be small by 3 respondents who
indicated that these actions were already being done. Others suggested the following
impacts:
Costs associated with the operation of the system and record keeping (4
responses). One respondent indicated that these could be minimised by having a
standard reporting template for use by all pharmacies and another felt that an IT
solution would make the process more effective and efficient.
The pharmacy would need to have good communication with doctors.
Systems would be needed to maintain a good record of changes; “It is important
that whatever system is use can be followed in a chronological order”.
The improvement to the recommendation suggested was to allow an electronic record
or electronic communication. The wording could be changed from “written record” to
“non-verbal record”. Suggestions to assist implementation were to:
Have a standard template.
Include the process in an industry standard for DAA software and procedure
manuals.
Integrating dispensing software with profile maintenance so that a comment
(recording the date, the prescribing doctor and the medication details) is added to
the profile when a new medication is dispensed.
Add this process to the Quality Assurance (QA) cycle.
Include these packing support tasks in the service costing in any payment model.
Quality Medication Care Group, School of Medicine, University of Queensland118
5.2.2.5 Continuity of patient care between hospital and community
Recommendations in the best practice model also included strategies to improve the
timely flow of medication regimen information and supplies of medications when
community patients were admitted to hospital. The recommendation covered both
information to be provided to the hospital and practices in hospital to facilitate patient
discharge and transfer of care to the community.
As mentioned earlier, the majority of patients were willing to carry a medication
template to hospital with them to assist in the transfer of medication regimen
information.
Communication of discharge information for community patients to the community
pharmacies occurred infrequently; 3 pharmacists reported ‘sometimes’ receiving a
patient discharge summary from hospitals and another 2 had never received discharge
information; only 1 pharmacist reported ‘always’ receiving a discharge summary. In
terms of information flow to hospital, one pharmacist commented “we are always
available to help but often hospital staff don’t contact us”.
All 6 responding pharmacists felt the recommendations were feasible (although one
with a reservation about who would fund it) and three indicated it would be beneficial to
a safe, effective and efficient DAA service. Of the 4 responding pharmacists, 3
indicated that little or no change would be required for their pharmacy to implement this
recommendation. The other pharmacists indicated that the only change would be to
make sure that the medication template used by the pharmacy had enough information
on it for the hospital staff.
The model recommendations were felt to address community pharmacist reservations
about packing a DAA based on discharge information (provided it worked). The only
improvement suggested for the recommendation was that the community pharmacy
receive the information immediately at discharge (if not before). Pharmacy opening
hours were identified as a possible delay or complication for the discharge process
For community patients, the GP commenting on the community model indicated that
she sometimes received a discharge summary but that this was not always timely. The
GP did not communicate new medicines to the pharmacy packing a DAA for a
community patient because the information was received from the hospital too late
(improvements in this area were suggested). The GP did not believe the step of GP-
Community pharmacy communication was necessary as the patient and carer were
responsible for dispensed medicines.
Hospital pharmacists indicated that a discharge summary was always or mostly sent
when a community patient using a DAA was discharged from hospital (in one case,
provided the hospital staff send the patient’s medication to the hospital pharmacy for
finalising). The format of the summary varied:
It was part of the nursing discharge form but often poorly legible.
A discharge prescription that included all current medications (not a discharge
summary).
Discharge summaries were sent to the GP but not to the pharmacist.
Other hospital pharmacist views on patient assessment and the patient-held template
as a communication tool are summarised in sections 5.2.2.1 and 5.2.2.3.
Quality Medication Care Group, School of Medicine, University of Queensland119
Generally, hospital pharmacists were supportive of the model in Figure 4.7, related to
the continuity of care for DAA users, and felt it would be beneficial in promoting safe,
effective and efficient DAA services. There were concerns or issues about the
recommendations for community patients and residents of RCFs (both included in this
sections as comments from hospital pharmacists refer to both the community and the
RCF best practice models):
Step requiring the GP to contact the community pharmacy to confirm that it was OK
to pack from the discharge list may not be feasible (2 responses).
A lot of the processes for hospital pharmacists were extensions of the normal
process expected in discharging a patient from hospital, however the model was
complex and would need to be refined to a simple stepwise process for the hospital
pharmacist to follow.
The model may suffer from various hospitals’ polices e.g. the need to
unambiguously identify all drugs in a DAA brought to hospital.
There may be problems with staffing to carry out these activities.
One pharmacist questioned the need for the hospital pharmacy to notify the
community pharmacy that an RCF patient had been admitted (see Figure 4.10) as
this was not always feasible and the aged care facilities should have this
information and be able to inform the community pharmacy.
Another pharmacy followed the model on the whole expect that (a) the discharge
summary and medications were faxed by the ward to the GP, (b) the GP was not
asked to give the OK to fill a DAA from the discharge list and (c) if a DAA was in the
ward, the hospital pharmacy filled it and informed the community pharmacy (and
faxed a copy of the prescriptions).
Another pharmacist in a large tertiary hospital that the need to maintain a database
of community pharmacies was beyond their facilities and resources.
The model did not include patient consent to send the medication discharge
summary to the community pharmacy. [Note: consent for this information sharing
would have been included in the tripartisan agreement].
There was a concern as to whether RCF staff could administer medications from a
GP-approved discharge summary [Note: the previous step in the model allows the
GP to give a telephone order to administer medicines]. A global policy/standard
may be needed to address this issue.
Would an alternative model involving nursing and medical staff be needed for sites
without a hospital pharmacist? One pharmacist felt that the model would probably
not work with weekend discharges.
Implementing the draft best practice models would require some work practice changes
and impact on workflow:
The hospital pharmacy would have to send the discharge medication list to the GP
as soon a possible so the list could be approved by the GP before the patient
returned to an RCF.
The level of staffing might need to change.
The models require a change away from focusing on the items needed at discharge
and to improving the quality of care and so may impact on staff mix (need more
qualified staff not just support staff), and staff are not available.
The continuity of care model was also intended to address hospital pharmacists’
concerns about the reliability of medication information (from RCFs, the DAA pack or
the patient template) at patient admission. These concerns were addressed to some
Quality Medication Care Group, School of Medicine, University of Queensland120
extent (for example, by including the contact details for the GP and community
pharmacist on the template) but more was needed:
Medication regimen information and the patient template needed to be dated of
have another form of version control.
The DAAs themselves needed to be dated and include a description of all
medications (so that individual medications in a multidose pack can be identified). A
way of confirming that a patient had brought all current DAAs to the hospital (and
did not leave one at home, for example) was also needed.
An electronic record would address some of the problems about currency and
responsibility for record maintenance but the template was a good interim measure.
Other areas for improvement for the recommendations about continuity of care or
circumstances that might complicate/delay processes were:
Processes would need to be put into place to address discharges after-hours, on
weekend or public holidays, or in hospitals without a hospital pharmacist.
Processes to support the continued supply of non-PBS medications and packing
these into a DAA.
Remove the pressure for doctors to churn the patients out of hospital as quickly as
possible.
Waiting for the GP to give the “green light” to dispense from a discharge
prescription was not practical and “superfluous”.
A central database of pharmacies, possible Guild-operated that supplied DAAs
would help so that the hospital could arrange a DAA to ensure continuity of supply.
Ten respondents to the expert panel questionnaire indicated that this recommendation
about continuity of care and the model was feasible but most respondents had
reservations. Two respondents indicated that this was already being done, one saying:
“We have established this with our hospitals after a lot of time and consultation. It
works very well when there is goodwill between the pharmacies and the hospitals.
We are trying to suggest that we are involved in admission as well. It is coming
slowly”.
In general, all respondents felt that something needs to be done to communication with
hospitals but that this recommendation was felt to be too ambitious to implement at this
stage of best practice implementation and much more work, relationship building,
active support from hospitals, IT support and resources would be required. The
reservations generally related to:
The need for all parties to co-operate and participate (3 responses).
Only some hospitals consider this information transfer to be a priority when
allocating resources and establishing systems.
Need major changes in hospital systems (3 responses) including simple things like
notifying the pharmacy that a DAA patient has been admitted.
That it would be difficult to co-ordinate discharge days with packing days (2
responses).
The provision of a DAA by a hospital may be a different kind and confuse the
patient.
Requiring GPs to be intermediaries may delay the process and contribute to
errors/re-work.
Would need more staff resources in some hospitals.
The recommendation could become more feasible with the development of information
and communication technologies (ICT).
Quality Medication Care Group, School of Medicine, University of Queensland121
Because the existing structures and processes needed to support this recommendation
are in their infancy, the impacts of this recommendation on costs and staffing in
establishing the systems were felt to be considerable but future payoffs were
envisaged (4 responses). If recommendations are put in place, success is possible:
“It minimises staff time if done properly as the hospital communicates with us at a
reasonable time to allow discharge procedures to be enacted and we are able to
ensure the patient has the medication available once home. We have a system
that minimises our use of new medication as the hospital arranges discharge
medications to be used until the patient can see their GP. Previously we had
problems with starting new medications on discharge and then the GP did not
agree with the hospital’s recommendation and no one was writing prescriptions for
the medications given on discharge. Once again collaboration between pharmacy
and the hospitals has minimised the waste of medication and the costs to the
community pharmacy”.
Suggestions were made for the improvement of the recommendation:
Make this recommendation an “aspirational standard” rather than one to be met
now.
Make the GP the line of communication with the hospital.
Recommend a database of pharmacy DAA providers on the Guild QCP website.
Integrate or replace with an HMR on discharge from hospital. The timing of the
HMR would be critical.
In using GPs an as intermediary on discharge, change Figure 4.7 to allow s number
of ‘OR’ scenarios “i.e. do this ‘or’ if this cannot be done, do this….”.
To support implementation, the support of state health departments and others would
be required. Implementation of this DAA aspect with existing projects to support
continuity of care would be helpful (e.g. Queensland Health initiatives, the General
Practice Aged Care Quality Use of Medicines (GPAC QUM) project). Funding might
arise through co-ordination of care.
5.2.2.6 Quality control (QC) and quality assurance (QA) for packing, checking
and communication
Recommendations related to assuring quality in the processes related to packing and
checking were included in the community best practice model (and in the RCF model).
The recommendations included quality control and monitoring, patient education about
drug storage and issues related to the stability of medicines packed into a DAA.
All six responding pharmacists felt that record keeping and monitoring would promote
safe and efficient DAA services and act as legal risk reduction. Little change would be
required to implement the recommendation; one pharmacy already fulfilled the
requirements of the recommendation. Another documented errors but did not always
coach the staff member as errors were uncommon and “just silly mistakes”. One
pharmacists reported that implementing the QA and QC recommendation would
require “more paper work” and another felt that monitoring would not be possible in a
community pharmacy. In referring to recommendations about drug stability, this
pharmacist commented that “bodies outside the pharmacy [were needed] to set and
research this for us”.
No improvements were suggested to the proposed QA and QC recommendations in
the models but guidelines and more manufacturer information (2 responses) on short
Quality Medication Care Group, School of Medicine, University of Queensland122
term stability in DAAs (including co-operation between DAA and drug manufactures)
were felt to be helpful in implementing the recommendations.
All thirteen respondents to the expert panel questionnaire agreed that record keeping
and monitoring can be beneficial in promoting safe, effective and efficient DAA
services, and be a form of legal risk-reduction. Other benefits of record keeping and
monitoring recommendations were that:
They encouraged the QA cycle.
They provided a structured framework.
Records can help measure trends (2 responses);
“We have incident reporting and staff procedures in place to identify areas of risk.
In terms of QA we have identified problem drugs e.g. Epilim disintegration; work
issues e.g. ergonomics of bench height etc. We document these issues and
ensure all staff are aware of the procedures we undertake to minimise the
problems”.
The main impacts of this recommendation and the changes that would need to be
made to existing QC and QA processes were:
Costs (5 responses), particularly initial development, documentation and
implementation. On-going costs would be less.
The costs could be off-set (3 responses) by risk reduction and minimisation of
return of unusable medication or rework.
Having appropriate staffing (“preferably a DA who is task-oriented with good
attention to detail and a heart for aged care, properly trained and involved at least
one day per week in packing DAAs and updating records, uninterrupted by other
pharmacy duties”).
Suggestions for improvement in the recommendation and possible tools or process
guides that would be helpful were:
Make the reporting of suspected instability, the use of the ADRAC-type report the
responsibility of TGA not PSA (2 responses). This could be implemented on the
TGA website in a way similar to on-line ADRAC reporting.
Include standards and guidelines and QCPP standards (2 responses). There needs
to be a checklist with specific points stated to reduce the interpretation needed by
assessors (include ‘doing’ words in the standards).
Provide templates for audits (2 responses) “it would then be a matter of determining
what the quality indicators and benchmarks [should be measured and explain] how
to assess gaps and what processes we might suggest to remedy or close out any
identified deficiencies”.
Add another strategy to get stability information along the lines that specific data on
stability of a product in DAAs needs to be provided to TGA by the medication
sponsor and form part of the formal registration/marketing approval process.
Recommend that stability information in texts such as the Australian
Pharmaceutical Formulary (APF) and Australian Medicines Handbook (AMH), etc,
be continually updated.
Workshops including train-the-trainer (for in-house staff training) could be
developed to support this recommendation.
Guidelines or procedures related to ‘education’ of patients/carers that including
what should be provided and how.
Quality Medication Care Group, School of Medicine, University of Queensland123
Other comments in relation to implementation were:
Quality control, assurance and monitoring procedures “should be both pharmacists’
responsibility and DAA suppliers’ responsibility to develop and provide policies and
procedures respectively”.
The costs for QC, QA and monitoring could be averaged across the fee charged.
QCP standards are needed to reduce variation and ad hoc supply of DAAs; “If a
pharmacy is not interested in maintaining high standards of professionalism in
providing a DAA service then they should not be approved to supply DAAs”.
Future developments in both packing technology and information technology could
improve the efficiency of these processes by integrating automation with electronic
networks and would be in line with government policies.
Integration of some of the QC, QA and monitoring activities with existing programs
such as the dispensing software or adapting ePocathary (see
Tell the pharmacy about changes to residents’ medicines
RCF: 89% RCF+GP: 11%
RCF: 63% RCF+GP: 13% GP: 25%
RCF: 57% RCF+GP: 43%
RCF: 14% RCF+GP: 57% GP: 29%
RCF: 33% RCF+GP: 66%
RCF: 66% RCF+GP: 33%
Making sure prescriptions are available to ensure continuity of supply of packed medicines
Pharm: 78% Pharm+GP: 11% Pharm+GP+RCF: 11%
Pharm: 75% Pharm+GP: 25%
Pharm: 71% Pharm+GP: 14%
Pharm+RCF: 14%
Pharm: 71% Pharm+GP: 14%
RCF: 14%
Pharm: 100% Pharm: 66%
GP? (unsure): 33% Inform the doctor when a repeat prescription is required
Pharm: 78% Pharm+RCF: 22%
Pharm: 63% Pharm+RCF: 25% GP: 13%
Pharm: 86% Pharm+RCF: 14%
Pharm: 86%
RCF: 14%
Pharm: 100% Pharm: 100%
Quality Medication Care Group, School of Medicine, University of Queensland133
5.2.3.1 Tendering for DAA services
The first recommendation in the RCF best practice model for DAA services concerned
a tendering process that addressed issues to be considered including those of
payment.
The best practice model of tendering for DAA services was felt to be feasible by seven
of the eight respondents to the DON questionnaire; four indicated that something
similar was already in place. One DON said that systems would improve over time
while another indicated that the clear outline of expectations would “help ensure that
both the RCF and DAA service provider will be satisfied with the outcome”. One DON
questioned the feasibility saying that extra time and money would be needed to
implement the models. The quality assurance (QA) processes were felt to be feasible
by all 9 DONs; several already had systems in place (e.g. a weekly audit for errors,
monitoring pharmacy deliveries) and another felt it was desirable to add QA processes
directly related to the DAA service to the set of QA processes already externally
audited. One DON questioned whether the RCF would audit processes internally and
who would be the external assessors.
The DONs felt the impacts of implementing the model or changes that would need to
be made in their facility were:
Discussion of the model with current providers and if receptive, set up a service
agreement as well as setting up a QA monitoring and external checking service.
The understanding of relatives and carers who could not understand the billing
processes.
The implementation of audits targeting DAA activities potentially had extra costs for
staff to do audits, benchmarking and any education to correct problems identified.
Staff training would be required including training of the medication committee.
More time required to enter QA data (needs to be done daily).
Two DONs felt that there would be little local impact because tendering was done
centrally. Another DON said that the impact on the RCF would be minimal because the
pharmacist already did it “at his own expense”.
Suggested improvements to the best practice tendering model made by DONs
included:
RCF should pay for packing (2 responses). Where an RCF initiates use of DAAs,
the RCF should bear the cost. This stance may be supported by the Aged Care
legislation in relation to high care residents and should be checked.
A recognition that use of DAAs were not just a cost issue but also about quality
care. DAAs helped reduce errors and medication administration time for RNs and
made it easier to audit for mistakes, errors or underhand behaviour of staff.
Limiting resident payment or contribution for DAAs to self-medicating residents.
Funding by the government to ensure all RCFs comply.
That packaging is a pharmacy cost.
Adding a requirement for regular review of procedures and policies for DAA supply.
All community pharmacists indicated that the model and recommendations related to
tendering were feasible and as many pharmacies already tendered, there would be
little cost or staffing impacts for pharmacies. The main impact pharmacists identified
related to RCFs typically having little understanding of the true costs incurred by
pharmacies and a perception that RCFs were not prepared to pay. All community
Quality Medication Care Group, School of Medicine, University of Queensland134
pharmacists also felt that the aspect of the model relating to monitoring and quality
assurance (QA) were feasible with one indicating these activities were already
performed in his pharmacy. Another pharmacist indicated that some additional
resources (extra staff) would be required to perform these activities and that needed to
be streamlined so that they were not too cumbersome in paper. Suggested
improvements to the tendering recommendation and model were:
Government funding for DAA provision.
Some payment by the patient’s family for provision of the DAA service.
A need to focus on quality rather than just price in the tendering process.
From the GP perspective, 3 of the 7 responding GPs felt that participating in the
tendering process for DAA provision would provide no real benefits to them. One
doctor noted that while some overall medical involvement and some training of GPs
about the service would seem appropriate, for the facility to include each individual
doctor would be a ‘nightmare’. In addition, one respondent indicated they “don’t feel
(its) my place to interfere with business negotiations”. Three GPs listed benefits of
involvement namely: medical input/ knowledge of the type of DAA to be used and an
awareness/discussion of the obligations of the doctor. One GP suggested rather than
individualised contracts and service agreements (see also 5.2.3.2) between the
pharmacy and RCF where the GP was to be aware of the variations for each of the
RCFs attended, a standardised statement of service provision across all pharmacies
would be preferred by GPs. These views were similar to those expressed by GPs
associated with the aged care program in a division of general practice. Contracting
was felt to be a commercial decision not involving GPs, however, some input and
consultation with GPs (the most likely scenario would be through GP representation on
a Medication Advisory Committee) would be helpful:
“GPs see some benefit in being involved in discussions regarding the impact on
their work practices – one GP expressed extreme frustration at being asked to re-
write all patient charts when the RCF changed Pharmacy provider”.
“It is widely felt that the GPs should be informed of the Pharmacy service used”.
This information should include who is providing medication supply services and
who is providing Residential Medication Management Reviews (RMMRs)
(especially if different providers), appropriate contact person and contact
information and where to fax or mailing prescriptions and GP requirements with
regard to medication charts.
Six of the eight people who answered the question in the expert panel questionnaire
about the feasibility of the model and recommendations related to tendering felt that
these were feasible:
“It is good to make [the elements] transparent to all parties, so that they see the
true cost of the service”.
“They…provide an education tool that explains the complexities of the whole thing”.
“The model is feasible as long as adequate remuneration and/or a reduction in the
administrative overheads associated with management of prescriptions is
addressed”.
Tendering was already felt to be widespread but only some contracts with RCFs were
done via tender; “The requirement specifically in the Accreditation guidelines is that the
facility should pay for services at a price that reflects the market status. Some RCF’s
have interpreted this as a need to ‘tender’ but that is not a requirement”.
Quality Medication Care Group, School of Medicine, University of Queensland135
Two respondents to the expert panel questionnaire had concerns about the use of
tendering – “Like all ‘tenders’ the emphasis tends to be on price” with the final decision
on the preferred service provider being made by the financial controller (who is not
always aware of the “inadequacies of a substandard service and the potential health
and safety risks to the residents”) rather than someone who fully understands the
implications of service quality. One respondent wrote that RCFs assume all providers
will give a ‘professional’ service as part of their professional obligation so that only price
mattered and, “[RCFs] have an unrealistic view of pharmacists’ profit from prescriptions
and have historically kept the price below the costs of delivering a quality service”. This
“ultimately leads to poor quality service and failure to meet best practice standards and
a dilution of the potential benefits”. Tendering was also felt to have the potential to
create an environment where RCFs would not use local pharmacies, selecting more
distant providers who offered a cheaper service and could mean that only “big players”
in the provision of DAAs would survive.
An improvement suggested to this recommendation about tendering was an increased
emphasis that in any tender, the quality of the service should be the first basis of
assessment; “Ideally the service would be decided on the basis of the focus on patient
care, quality systems and sustainability of the service and not paid out of the RCF
funds”. Sample contracts were also needed.
Other changes suggested included:
Increasing the awareness of RCFs about what constitutes a quality service and that
it is necessary to pay a reasonable price to adequately resource such a service.
“The argument should be put to the facility that in cost-benefit terms this will save
them money”.
Using a term other than “fair price” which assumes the current fee to pharmacy is
not fair. A phrase like “mutually agreed remuneration” would be less presumptive.
Add geographical criteria i.e. that a DAA provider should be within a certain
distance of the RCF to the tendering criteria.
“Design of forms for pharmacist to enter their information, and not feel
overwhelmed when they have to develop and economic framework for working out
their costs”.
The aspect of the model relating to monitoring and quality assurance (QA) was seen as
feasible by five of the 10 people providing comments about this aspect of the
recommendation. Two respondents felt that this was feasible because systems were
already in place; in one instance, reports on these kinds of quality indicators were
made to a facility’s Medication Advisory Committee (MAC). However improvement in
these systems was needed “RCF’s are often in need of improving their quality systems
… Pharmacists need more encouragement and better training in how to keep records
of interventions”. One respondent had reservations about the feasibility of external
assessment based on experiences to date of aged care assessors and QCPP
assessors:
“In the majority they come with their own preconceived ideas and whimsical ways
…. Until we can make assessment and accreditation standards and people who
perform this work into uniform people delivering a role then I am a sceptic of this
process”.
The monitoring and QA aspects were felt to potentially impact on pharmacy costs
(through additional staff time) (2 responses) but this could be minimized by building
Quality Medication Care Group, School of Medicine, University of Queensland136
these activities into existing systems and facilitated by an IT approach. Three
improvements were suggested:
“Nursing Home accreditation should link into this process and should be based on
quality”.
“There needs to be some cross communication between the aged care and
pharmacy assessors with significant penalties applied if parties are non-compliant.
Currently aged care assessors can apply sanctions to RCFs. This may also need to
be considered if pharmacists are non-compliant with standards”.
“In the Community Pharmacy-RCF loop somewhere there should be mention made
of the feedback on the whole DAA/medication management process from the
reporting of the accredited pharmacist doing the RMMRs , especially if this person
is different from the community pharmacy”.
Sample templates were needed.
Other expert panel comments on the tendering recommendation related to the issues
of funding included in the text. Three respondents indicated that RCFs should not
control funding of the DAA service, in particular, based on experiences under Care
Aggregated Module (CAM) funding:
“Even though the money had to be acquitted and refunded if not spent, … there were still RCFs that out of principle refused to pay pharmacists for the services provided. … Inevitably the service suffers to the point of affordability judged by what the RACF thinks the pharmacy should be paid”.
Another respondent felt that patients should not have to contribute to DAA costs
“especially if they are not given any choice or control over the provision of the service”.
Two alternative funding or fee setting strategies were suggested:
That the RMMR payment model could be used as a framework for direct-to-
pharmacy payment from government for DAA provision.
“There needs to an independent arbitrator that sets a fee per bed” particularly
where DAA provider contract changes are seen to be due to price undercutting.
Other improvements to the model in Figure 4.8 and the recommendation overall were:
The model “seems to devolve much of the administration and responsibility to the
pharmacy; Improve by providing flexibility for RCF to do some activities“ (e.g.
prescription management that is already done by some facilities).
The model would be improved if the system of PBS prescriptions was changed so a
“doctor written medication order (on the medication chart) could act as a PBS
prescription for the life of the chart”.
“This document could be converted to a more meaningful working document where,
at the highest level of documentation, the big picture aspect of the proposed
working relationship is detailed and what that entailed”.
There may need to be anti-kickback legislation to support the process to “prevent
pharmacies and nursing home proprietors coming to financial arrangements”.
5.2.3.2 Written agreement for DAA supply
In a process similar to that of the tripartisan agreement for the community model, the
RCF best practice model also included a recommendation for an agreement that
specified obligations and expectations, and promoted mutual awareness about the
DAA service to be provided. As one GP group indicated “one of the basic problems
relates to a lack of understanding and opportunity to discuss the roles and needs of the
Quality Medication Care Group, School of Medicine, University of Queensland137
parties involved”. From the resident and family perspective, this agreement also
included how the service would work, payment and billing, procedures for start up to
streamline the transition after RCF admission, a recognition of privacy of information,
and how the pharmacy would provide counselling and medication information
(consumer medicines information or CMIs).
From the resident perspective, the move into residential care has potential to cause
considerable disruption in the medication management and supply of medication to
individual residents and there is a need for streamlined and explicit processes to
facilitate the transition. Of the residents interviewed for this work, all but one had been
living in their own home (one transferred from another RCF) and all had previously had
managed their own medication supply (several with some assistance from family
members). The move into an RCF required five residents to change their usual GP
and six to change their usual pharmacy. Only four residents recalled that RCF staff
had fully explained to them the medication supply-related processes involved in the
transition. Two other residents had been managing their own medication at the time
and another said the processes were not explained. No problems were recalled when
the responsibility for medication supply and prescription management was transferred
to the RCF; this was attributed to the organisational strengths of the RCF.
None of the residents interviewed had a formal agreement with their pharmacy
covering the details surrounding the provision of DAA services and six residents
indicated that they would be willing sign an agreement covering such points. No parts
of the agreement were felt to require change or omission. When asked to rate such an
agreement as either extremely helpful, somewhat helpful or not at all helpful for people
moving into aged care facilities, 2 residents considered it to be extremely helpful, four
somewhat helpful and one considered it to be not that helpful at all. While this resident
suggested that in her experience a formal agreement would not have been all that
useful, due to the fact that the RCF was so organised, she conceded from a general
perspective such an agreement would be beneficial, provided a resident’s cognitive
function was taken into account.
The provision of medication information for aged care residents forms a key feature of
the best practice model with the added aim of tailoring the information to a variety of
residents with differing involvement in their own medication management. Three
residents had been receiving written medication information from their pharmacy prior
to moving to an RCF and one indicated that although written information was available
they never personally requested it. Six residents indicated that it was still important to
receive medication information; one felt it wasn’t important because the RCF already
had taken care of all features concerning their medications for them. While no
participants had any real family/carer involvement with their medication management,
the potential need to provide medication information to families/carers was identified as
a relevant issue by one resident particularly where a resident had poor cognitive
function.
The provision of medication information to residents did occur. Five residents indicated
that they had received medication information since RCF admission; three were given
information verbally by either their doctor or nurses and two had received written
information. For both of these residents, written information was sourced from the
facility (one facility had a written sheet for every drug used by residents).
Quality Medication Care Group, School of Medicine, University of Queensland138
Since the RCF best practice model agreement included defining how written
medication information was to be provided, residents interviewed were asked about
their preferences for this type of information. Preferences varied based on a resident’s
level of participation in their own medication management. Since there are existing
guidelines for the provision of CMIs [see http://www.guild.org.au/public/currentissues/
mic/mic_when.pdf (last viewed 5/4/06)], residents were asked in which of the following
situations they would like to receive CMIs:
When first starting a new medicine: 5 residents.
When the drug information had been significantly changed: 4 residents.
When the dosage form had been changed: 2 residents.
When there was special information surrounding the use or precautions
surrounding a medication: 3 residents.
When they requested it: 5 residents.
Routine updates on long term treatments: 2 residents.
Verbal information was considered sufficient among several respondents when the
dosage form had been changed and with one respondent when there were special
reasons regarding the use/precautions surrounding a particular medication.
No resident indicated just one situation in which they wanted a CMI or similar, but two
residents said there were no circumstances in which they wanted to receive a CMI.
One resident said that they felt the doctors would tell them all the important information,
while the other indicated that they were happy to let the RCF staff handle all their
medication management issues. Given this variation in resident preferences for written
information, explicitly defining a resident’s expectations in an agreement should aid
appropriate service delivery and increase satisfaction.
From the RCF perspective, eight of the nine responding DONs were willing to enter into
a written agreement for DAA supply, as outlined in the best practice recommendations.
Five DONs considered that all aspects of the agreement were equally important, while
other DONs emphasized the importance of cost (3 of 8 respondents), how to keep
pharmacy resident profiles current and aspects of the pharmacy services such as time
of service availability, prescription management, the availability of medications and
other support services such as staff in service education and participation on the
Medication Advisory Committee (MAC). Clearly defining roles and responsibilities was
felt to fundamental by one DON. None of the DONs interviewed recommended leaving
out any part of the agreement. Suggestions for improvement or changes were:
No change; the guideline was good but needs to allow RCFs to adapt it to suit their
culture.
Adding specification of timeframes for the dispensing and delivery of drugs [Note:
this was included in the recommendation about how the service worked but was not
explicit in dealing with medication changes] and timeframes for doctors writing
prescriptions.
Leaving out visits to hospital (as hospitals have their own systems) and internal
pharmacy issues such as discussions with GPs or the length of time medications
are kept if ceased.
All community pharmacists were willing to have a written agreement as per the
recommendation with one pharmacist indicating that it was essential for parties to
understand their roles for the system to be effective. All pharmacists felt that every
aspect of the agreement was important although, one pharmacist did question the need
to address privacy as this was a “red herring” and another felt that it would be hard to
Quality Medication Care Group, School of Medicine, University of Queensland139
define how quickly a medication change should be done and that this section may
require careful wording. This same pharmacist foresaw possible difficulties where large
numbers of doctors visited an RCF. One area for improvement identified was the need
to explicitly include expectations for DAA service for residents admitted for respite care
(for a fixed, and usually brief period). The example given was: the patient arrives with
their own medication and “the facility would expect us to pack it up, for several weeks,
for nothing, and then, at discharge, justify exactly why certain amounts [of patient’s own
medications brought in on admission] were left over”.
Of the GPs commenting on the RCF model, four of seven indicated they would be
willing to participate in a formal agreement regarding DAA supply as per the best
practice recommendations. The aspects of the agreement felt to be useful or important
were:
Issues regarding medication changes (4 responses) including procedure, time
taken, and the responsibility for communication of changes to relevant parties (3
responses).
Prescription management (3 responses) including when a prescription would be re-
written without seeing a patient and communication about when new prescriptions
are required.
The formalization of how the service will work (2 responses).
Privacy issues.
What happens when RCF staff make an error.
What happens to wasted medications in a DAA.
How much responsibility the RCF and RNs involved are prepared to take.
One of the GPs who did was not willing to be involved in any agreement about a DAA
service stated “the difficulty with the agreement between the pharmacy and the doctor
is the we [the doctors] are employed by the resident; there is no business relationship
between the doctor and the pharmacy”.
On the whole, while written agreements for DAA provision were not embraced as
strongly by GPs in comparison to other stakeholders, many of the points addressed
within were positively received. Two GPs recommended changes to the agreement.
One suggested including a statement that the agreement was able to be read and
understood by the patient and one other family member but it was not clear what would
happen if patient agreement could not be obtained (e.g. if patient had dementia and no
family). The other GP suggested reducing the complexity and specificity of the
document. This GP said that “at present, where there is mutual respect, doctor and
pharmacist can come to agreement about the way medications are dispensed, etc”.
Another GP agreed that there was no need for a written agreement between the
pharmacist and the GP (although this should be a close working relationship); the
agreements should be between the RCF and pharmacy, and the RCF and the doctor.
The preference for a non-written arrangement or understanding also appeared to be
shared by another GP group who indicated that “where there has been an opportunity
for such discussions it seems that a mutually agreeable solution can be reached e.g.
one GP changed their RCF ‘visiting day’ to precede the pharmacy ‘packing day’ “.
Respondents to the expert panel questionnaire were also asked what aspects of the
suggested written service agreement with the RCF and/or patients or doctors were
important or useful. Of the 11 people responding to the questions, six indicated that all
aspects were important:
Quality Medication Care Group, School of Medicine, University of Queensland140
“It makes sense as it covers all aspects of the agreement and leaves nothing to chance”.
“Leave out None – RCFs often have an abysmal understanding of the logistics in
providing a DAA service (which they often expect for nothing) – this agreement
clearly articulates all the problems and conditions involved in the provision of the
service and provides some ammunition for a pharmacy wanting to charge for the
service and provides clear guidelines on who is responsible for what”.
One respondent did recognize that “these agreements will never be completely
inclusive as new issues come up all the time”.
Respondents to the expert panel questionnaire suggested the following changes to the
content of the agreement or the subjects to be negotiated:
Include the role/responsibilities of the accredited pharmacist doing RMMRs; “this
needs to be highlighted in any such document”.
The agreement needs to “recognise patient choice and rights and responsibilities”.
Change the implication that prescription management is the pharmacist’s
responsibility. “I can accept responsibility for notifying the doctor a new prescription
is due but I cannot take responsibility if the doctor fails to write it in a timely
fashion”.
“It should be acknowledged that it is not unusual for residents to be assisted by
their carers with their medication even within RCFs. Therefore there is a need to
involve and inform families/carers in this process just as much as within the
community. They too need medicines information and how this is to be provided to
carers should also be included in the negotiation discussion”.
There was some disagreement about whether the agreement should be a standard
template – two respondents felt a standard agreement was required (one indicating
that stakeholder input was required to develop this template) while two others felt that
the agreement was more a guideline that could be adapted:
Pharmacists and RCFs should be able to start with a broad agreement framework
and then trim it down to individualise it. “I think it is important to start with a
comprehensive agreement and then use the MAC and meetings with management
to tailor the agreement. I find differences between privately owned, corporate
RCFs, rehab centres, jails, etc, that I need to alter my basic agreement to a mutual
level”.
“Rather than provide a ‘sample’ agreement type approach, I feel it would be better
to establish headings which provide a framework. Mandatory and optional; that
way there is scope to clearly flag for all parties the ‘no compromise components’
and the ‘Flexible components’ ”.
Another respondent said that training sessions would be needed for implementation
while another suggested that the agreement should be presented in an acceptable
format; “It should be presented in a folder with clear headings and with explanations of
the importance of each section. It should explain the potential outcomes when certain
elements of that agreement are not adhered to”.
Some respondents felt that service delivery agreements were feasible and had a
positive impact:
“It is feasible and essential from a quality perspective. It is a requirement that all
aged care facilities have formal agreements with external service providers as part
of their accreditation requirements”.
Quality Medication Care Group, School of Medicine, University of Queensland141
“You need the mutual agreement document to ensure you minimise staff time and
costs. A facility with a poor understanding of the agreement and the contractual
procedures is the ones that give you the most grief in tying up staff time. It is always
hard to bring them back to a reasonable level of impost on your staff times if you do
not have a tight agreement at the beginning”.
“A good agreement up front solves may potential problems later on”.
“Once instituted, this would minimise pharmacy staff time as long as RACF staff
were all aware of their responsibilities”.
Another respondent anticipated additional costs to those involved:
“With the implementation of any procedure that guarantees quality there will be costs. The RCF needs to understand that what they are BUYING is REASSURANCE that what is delivered is RIGHT for the RIGHT patient at the RIGHT time”.
Two other respondents had reservations as to the feasibility of the recommendation:
That it would be prohibitive for smaller pharmacies.
That it was not feasible the way it is written, specifying what needs to be addressed
e.g. What happens if pack needs to be changes, rather than spelling out
(prescriptively) how a pack change should be done (e.g. that it is the RCF
responsibility to fax a copy of the changed medication chart to the pharmacy at
least 4 hours before the pharmacy close of business) leaves the pharmacy open to
unreasonable pressure from nursing homes to take on responsibility for all aspects
or to meet unreasonable expectations (e.g. A nursing home could force a
pharmacists to go out to home on the weekend). The document has little sense of
this commercial reality.
5.2.3.3 Communication of medication changes
Problems associated with the communication of medication changes was one of the
main barriers to a safe, effective and efficient DAA service identified in Phases 2 and 3.
While some issues associated with medication changes (e.g. responsibilities and
timeliness) have been included in the recommendation about a written agreement,
maintaining a written record of all communication about medication changes was a
further recommendation.
From the RCF perspective, all DONs indicated that the recommendations for the
communication of medication changes were feasible to implement and beneficial to
patients. All DONs also reported that these changes were likely to incur little or no
extra cost or use of staff time (as most were already using such systems). Some
changes to RCF practices (e.g. always keeping a record of communications to the
pharmacy) and some additional staff training (to ensure procedures are always
followed) may be required. Suggested improvements to the best practice model of
communication of medication changes included:
Increased role of GPs in communicating medication changes to pharmacies (2
responses).
Earlier notification by pharmacists of the need for new prescriptions e.g. by emailing
to RCF.
The use of medication charts ‘online’ to the pharmacy to efficiently communicate
changes “so that changes are immediately obvious to all”.
All three community pharmacists commenting on the RCF best practice model felt that
felt that the recommendations for communication of medication changes were feasible
Quality Medication Care Group, School of Medicine, University of Queensland142
or beneficial with one describing them as ‘vital’. Another pharmacist indicated that in
practice, it might be difficult to implement particularly for communication with specialists
and hospitals – hours could be wasted on the phone clarifying exact requirements. The
concept of a ‘written record’ was interpreted as a paper record by one pharmacist, who
stated a preference for an electronic record showing date of changes and gave details
of who communicated the change. The impact of this recommendation was felt to be
minimal except for more “paperwork”. One pharmacist did identify a change in current
RCF and GP practice where residents visited the GP in the GP’s surgery “patients
would need to take the medication chart to the doctors” so that the change could be
recorded immediately on the chart but there was also a risk that charts would not come
back to the RCF.
The only improvement to the proposed BP model for communication of medication
changes suggested by community pharmacists was the use of a traceable electronic
record or written records.
All seven GPs commenting on the RCF model indicated that they communicated
medication changes; two GP said they amended the medication chart but did not
specify how this information arrived at the pharmacy. Three other GPs wrote on the
medication chart and indicated that the chart was faxed to the pharmacy. It was not
clear for these 4 responses just whose role it was to communicate changes to the
pharmacy; in the other case, the facility faxed the amended chart to the pharmacy. One
GP used the prescribing program Medical Director® to print out a patient drug sheet to
send to the pharmacy. Another GP at times would write a hand-written note and only at
the request of nurses (who presumably communicate this to the pharmacy).
All 10 people responding to the question in the expert panel questionnaire about the
communication of medication changes in RCFs agreed that the recommendations were
feasible and beneficial, with several indicating they were vital or mandatory.
“Proper records enable us to pinpoint where problems arise especially when there
are so many agency nurses now in RCFs. It helps resolve disputes. It is hard to
ensure continuity of care unless these records are kept and proper communication is
mandated”.
Unlike the community setting, the pharmacy practice of asking for changes in writing
was often in place:
“We DEMAND a full medication chart is faxed with every change as that is our QC
[mechanism] to ‘discover’ all of the changes they forgot to send us. We are also not
very lenient on nurse initiated medications and we demand a medication chart
signed by the doctor or a telephone order/confirmation from the doctor”.
Some difficulties and costs were envisaged with implementation (“at the moment it
costs a lot for pharmacy to do this”) but an IT approach would help to minimize costs.
One of the improvements suggested for the recommendation was to explicitly allow
electronic records of communication in the recommendation (as also suggested for the
community model in 5.2.2.4). Other improvements suggested by the expert panel
questionnaire respondents were:
To ensure that the recommendation complied with all state and PBS regulations.
That a standard reporting/communication form be developed.
That the “nuts and bolts” of the process were detailed and agreed upon.
Quality Medication Care Group, School of Medicine, University of Queensland143
5.2.3.4 Continuity of patient care between hospital and RCF
As with the community best practice model, recommendations in the RCF best practice
model included strategies to improve the timely flow of medication regimen information
and supplies of medications when residents were admitted to hospital. The
recommendation covered both information to be provided to the hospital and practices
in hospital to facilitate patient discharge and transfer of care to the RCF.
Six of the nine DONs indicated that the recommendations for continuity of patient care
between hospital and RCFs were feasible. Some reservations were expressed:
The process was too complex (2 responses).
That hospitals would not be keen to be involved.
Use of a 4-week pack means that changes are often needed when a patient is
discharged and would residents pay for new medications?
It was also not clear who would take responsibility; a GP ultimately oversees the
situation on a resident’s return from hospital.
Can poor notification of discharge plans by the hospital be overcome?
Communication between the hospitals and the pharmacies not practical as the
combinations are too large in number and there is a risk of communication
breakdown with the RCF.
The impact of these recommendations was felt to be relatively minor:
No change or minimal changes required (4 responses) e.g. one RCF currently sent
a copy of the medication chart but did not send the DAA, because the model is
largely followed. There are, however, costs such as staff time, photocopying,
faxing, etc.
The hospital needs to communicate with the community pharmacy when patients
are discharged back to the RCF (2 responses). In one RCF, the facility receives the
discharge medication plan and communicates it to the pharmacy.
Staff time would be needed to liaise with the hospital pharmacy plus extra RCF-
pharmacy liaison so that the community pharmacy can liaise with the hospital.
Hospital staff require education about the need for appropriate timeframes so that
medications can be ordered and be on hand in the RCF.
Suggested improvements to the continuity of care recommendations or situations
where processes may be delayed were:
Situations where communication flow is under pressure e.g. emergency situations,
other time where not all parties communicate effectively, poor handwriting (2
responses about poor legibility of discharge orders).
May need explicit strategies to deal with admissions and discharges involving
Schedule 8 medications.
Medications for discharge may not be decided until very close to discharge.
Many medications at discharge may be short term (e.g. antibiotics) that may not
need to be packed or would not be packed in time.
Since routine medication administration times may differ between hospitals and
RCF, the transition may also need to address transition in medication
administration times.
The hospital may need to contact the RCF to find out what and how much medicine
to provide on discharge; the model currently relies on the community pharmacy to
provide this information.
Implementation of an electronic system like “Medical Director”.
Quality Medication Care Group, School of Medicine, University of Queensland144
Community pharmacist commenting on the RCF best practice model reported marked
variation in the receipt of discharge summaries, a step included in the continuity of care
recommendation; one pharmacist rarely received summaries, one sometimes received
summaries (although this was improving) and another always received summaries.
Two pharmacists felt that the model would be beneficial in promoting safe, effective
and efficient DAA services; one pharmacist explicitly indicated that the model was
feasible. All community pharmacists had reservations:
The timing of activities may be a problem; currently “all residents seem to be
discharged at 4.30pm with no medication and [with the RCF] making ridiculous
requests on the pharmacy for a DAA to be provided in 20 minutes (and delivered)”
[Note: a medication round at the evening meal (approximately 5pm) is common
practice in RCFs].
A lot of co-ordination would be needed.
The quality of discharge summaries varied.
There was a danger of a DAA being lost in the hospital if it was sent with a resident.
The medication chart should suffice.
The impact or changes community pharmacists would have to make were minimal, all
said no change would be required but with two qualifications:
No changes were required for patient admission to hospital.
No changes; “if the hospital notified the GP my workload would decrease”.
Two community pharmacists agreed that this recommendation addressed their
reservations about packing a DAA based on discharge information, one saying that
compliance from the hospital with aspects to increase the reliability of discharge
medication information can be a problem. The other pharmacist indicated that the
model was little different to what was already in place and did not address reservations.
Improvements to the recommendation for continuity of care were:
Specifying that the discharge medication list be printed or typed, not handwritten.
Specifying that residents of an RCF should not be discharged after 12 midday to
allow sufficient time for the supply of new packs to the RCF.
The circumstance identified that would complicate the discharge process was the
hospital not complying.
To provide information about the feasibility of this continuity of care model from the GP
perspective, GPs were asked about their current practices and experiences. GP
currently described variation in the extent of communication from hospitals when a
resident was discharged; one GP from a rural town indicated that he never received a
discharge summary, another sometimes received a summary while 4 indicated that
they mostly received discharge summaries for RCF residents (although this was not
always timely).
The best practice model for continuity of care between hospital and the RCF had a step
where the GP confirms with the pharmacy the medication a resident is to take on
arrival at the RCF, so that the pharmacy can pack appropriate medications into a DAA
minimising delays and the potential for rework. Three GPs indicated that, when a
resident was discharged from hospital with new medicines, they communicated with
the pharmacy who packed the DAA about what should be packed or changed. One GP
faxed a ‘changed DAA sheet’ to the pharmacy. One GP only communicated what
should be packed or changes if it was requested by the RCF or pharmacy. Three GPs
Quality Medication Care Group, School of Medicine, University of Queensland145
relied on the RCF to communicate with the pharmacy, one GP saying this was after the
RCF communicated with the GP, while another GP inferred that the RCF were notified
of discharge medications and communicated these changes directly to the pharmacy
without GP involvement (a situation described as “fortunate”).
GPs were asked whether the step of GP confirmation of a drug regimen after hospital
discharge was necessary. Two GPs felt this step was necessary, one saying “mostly
because the patient is seen in hospital in an acute setting, not the overall welfare of the
patient long term”. One GP felt it was unnecessary and that “when the GP is not
available, the pharmacy should be able to change DAA medications as listed on the
hospital discharge”.
To further probe the feasibility of the continuity of care model, GPs were asked to
describe their current procedures including the timeframe for updating a resident’s
medication chart when the resident has returned to the RCF from hospital with new
medication. The following practices were described:
Attempt to make changes within 24 hours (2 responses).
If substantial changes, attempt to charge the chart within the same working day. If
there are difficulties fitting in a visit to the RCF and the GP knows the resident and
staff well and there is only a minor change, the GP would alter the medication chart
in the surgery and fax the amended chart to the RCF.
Usually 24-48 hours.
Same day review of the patient BEFORE DAA change. The pharmacy is then
telephoned and a ‘changed DAA sheet’ is faxed to the pharmacy.
The hospital updates the RCF chart. The GP would normally see the patient within
1 week if the GP wanted to change any medication.
In an attempt to streamline the transition process, one GP practice in a rural town
tried to encourage visiting medical officers at the local hospital to send an amended
medication chart with the resident on discharge back to the RCF.
The new medication list is sent by the hospital to the RCF and the RCF send the
chart to the GP to amend [Note: this GP also worked at the local, rural hospital].
Circumstances that GPs felt complicated or delayed the transition of care process
included:
The GP being away or otherwise unavailable (e.g. family commitments) (4
responses).
Notification of discharge late in the day or weekend/public holiday discharges (3
responses).
When GP notification about resident admission or discharge is not timely e.g. “a
letter arrives on desk one week after discharge”.
Poor documentation from the hospital.
Pharmacy unwilling to make a DAA change on verbal order from the GP so that
faxing an instruction can cause a delay.
From the responses about continuity of care recommendations for RCF residents made
by the people completing the expert panel questionnaire, it is evident that opinions on
feasibility varied and many of the same problems with feasibility arose. For example:
“It seems fine in theory but very cumbersome and getting the
commitment/involvement of the GP ant the various stages will not be easy”.
“The model appears clear but may come into conflict with hospital policies which
differ from public to private hospital systems. Hospital pharmacists are not
Quality Medication Care Group, School of Medicine, University of Queensland146
specifically funded to coordinate with community pharmacists and may follow their
usual procedures and not worry about DAAs, leaving that for the community
pharmacist to worry about. This can either lead to double prescribing or repacking
by the community pharmacist (for which they are not funded!)”.
“Not feasible. It assumes all hospitals have the same resources”
“Doctors could make this process very difficult. It assumes a Doctor is there ready
to help the patient on discharge. I could see in some cases a DAA could take 5
days post discharge to be delivered”.
Another respondent indicated that the model in Figure 4.10 was “an ideal model”. Two
respondents indicated that the recommendations and model were largely being
followed and that there were benefits from the time invested in setting up these
processes:
“It comes down to following the process to completion and having a safety net
available so the resident does not miss doses”.
“We have established this in our community and it works well…. the time wasted in
packing for patients that have gone to hospital is minimised by effective
communication”.
Suggestions to assist implementation were:
Using an IT approach to facilitating continuity of care
“The roles and responsibilities might be “thrashed out and agreed upon at MAC or
GP panel meetings”.
5.2.3.5 Quality Control (QC) and Quality Assurance (QA) – Monitoring
procedures for packing, checking and communication
As with the community best practice model, the RCF model also included a
recommendation about quality control and quality assurance (where drug stability in a
DAA was included). All community pharmacists indicated that this recommendation
was feasible. The impacts of implementing this part of the model were minimal. One
pharmacist indicated that the pharmacy already had extensive QC and QA processes
while another felt that the costs would be higher and that help and guidelines would be
needed to assist implementation. The development of standards based on good
manufacturing practice was the only tool suggested as helpful to the adoption of the
recommendation.
The respondents to the part of expert panel questionnaire on the Quality Control,
Quality Assurance and monitoring had already commented on much of this
recommendation (5.2.2.6). For the RCF model, respondents were asked about the
differences between the recommendation in the RCF model and the similar one in the
community model, namely, educating RCF staff about medication storage and involving
them in capturing evidence of drug stability. Five of the 10 responses indicated that this
aspect of the recommendation was feasible. Such education could be part of normal
staff education and quality assurance. One respondent indicated that there was a
potential to integrate this activity with QUM responsibilities related to staff education in
the proposed changed model for RMMRs. Another respondent indicated that “I rely
heavily on the nursing staff alerting me to problems like this”.
Quality Medication Care Group, School of Medicine, University of Queensland147
Some of the difficulties foreseen were:
Capturing evidence of drug instability is not funded for any providers.
Unless instability is obvious (e.g. dose form visibly degraded or a pack has
malfunctioned), it can be difficult to identify.
It can be difficult to provide education in a situation where staff constantly change.
That calling for staff education about medication storage had the potential to create
huge expectations of pharmacy (as staff education was normally an RCF
responsibility) [Note: the recommendation does not indicate who should provide the
medication storage education].
5.2.3.6 Review of charting systems
As a record of a resident’s current medication is maintained in three places, regular
checks of agreement or concordance between these records was recommended in the
RCF best practice model. All community pharmacists were willing to check
concordance between the pharmacy packing profile, GP records and the RCF charts
every 6 months. Two indicated that this was already done (in one case, more
frequently) while another indicated that cost would be a problem.
All seven GPs were willing to conduct a 6-monthly concordance check between their
records, the resident’s chart and the pharmacy packing profile. One GP indicated that
this would be “an excellent idea” while two were already undertaking these
concordance checks (in one case, only checks between the GP records and the RCF
charts, but many of the charts were generated by the pharmacy).
Five respondents to the expert panel questionnaire supported 6-monthly concordance
checks. Three respondents indicated that these were already being done three-monthly
(“just part of the unremunerated extended service”);
“We find the check invaluable. It ensures our records are accurate and up to date
and it helps identify weaknesses in the communication between the RCF and the
pharmacy”.
There were concerns that there was no funding model for this 6-monthly check. One
respondent suggested that “the RMMR pharmacist could perform part of the activity
and provide information to the community pharmacist to collate and forward to the
doctor, however this would occur once a year only”. Two other respondents suggested
that the 6-monthly check could be incorporated into the proposed QUM model for
RMMRs. Other implementation strategies were suggested:
Using IT to facilitate the check.
That there needed to be accountability that the check happens (e.g. QCPP or Aged
Care assessors).
That the check could occur in conjunction with “the [APAC residential care
guidelines and Aged Care standards] recommendation for Doctors to do a six
monthly assessment of each resident. The RCFs push their doctors to do this as
they [the RCF] lose points at accreditation if this standard is not met”.
5.2.3.7 Efficiency in pharmacy procedures
The RCF best practice model also included recommendations aimed at increasing
efficiency in the pharmacy include ensuring that prescriptions are available when
required, reminder systems for doctors and patients, and DAA packing practices
(including use and training of non-pharmacist staff).
Quality Medication Care Group, School of Medicine, University of Queensland148
All three community pharmacists responding to the RCF model felt that the
recommendations for pharmacy procedures promoted safe and efficient DAA services.
The impact of implementing these recommendations was unknown by two pharmacist,
one because this was already done and the cost was part of a “global costing”. The
other pharmacist indicated that implementation would be associated with higher cost.
All three identified that a change in the way prescriptions were written for RCF
residents would improve these recommendations i.e.:
“Allowing the pharmacist to dispense [and presumably get paid] directly from the
medication charts”.
Need to develop some way other than a prescription to supply RCFs – “getting
scripts from some doctors is nothing short of a miracle”.
“Repeat availability is difficult in the current situation”.
In Phases 2 and 3, ensuring that a valid and current prescription was available in time
for the DAA packing cycle generated problems for both GPs and pharmacists. The best
practice model included recommendations that involved the inclusion of explicit mutual
understanding of the procedure and practices associated prescription management in a
DAA service agreement (4.6.3.2) and in improving the efficiency of pharmacy
procedures (4.6.3.7.1). To inform the question of feasibility of these recommendations
from the GP perspective, GPs were asked about current prescription writing practices,
their preferences related to prescription reminders and how the recommendation about
prescription management (4.6.3.7.1) might be improved.
There was variation in GP views about prescription management. GPs were prepared
to write a prescription without seeing a resident in the following circumstances:
For continuing medications for one of the GP’s patients (6 responses).
Changes requested by the hospital or specialist.
Usually only order over the telephone, and only for temporary medications if the
staff on duty and the resident are well known to the GP.
For certain antibiotics, initiate on the advice of the nurse (patient seen later).
In Phases 2 and 3, pharmacists indicated that some GPs were not willing to write
prescriptions without having a consultation with the resident (and hence a consultation
fee). Some GPs charged the pharmacy for having to write new prescriptions for
ongoing medications if the resident had not been seen. Among the six respondents to
the RCF model however, none charged for prescription writing when a resident was not
seen, although one GP indicated that a fee of $2-3 would be fair.
Prescription reminders from the pharmacy to the GP were included in the best practice
model. This task was considered to be a pharmacy task in the model as discontinuity of
prescriptions has a big impact on the efficiency of pharmacy procedures related to DAA
packing and can disrupt the packing schedule. The prescription reminders, if well done,
were included to increase efficiency. GPs views on what should be included on a
prescription reminder to promote best practice were:
The medication, strength, form and dosage (4 responses).
Date last repeat issued/used (2 responses).
Date the prescription is needed by or next pack due date (2 responses).
Resident’s name.
Any recent changes of dose as if the change is made on the RCF chart, this is often
not transferred to the doctor’s computer in the surgery.
Quality Medication Care Group, School of Medicine, University of Queensland149
Pharmacies needed to track requests for repeats; one GP experienced repeated
requests for prescriptions for the same medication for a patient when they could not
have been needed. Another group of GPs also expressed similar frustrations with
prescription and medication chart requests and writing:
“GPs feel at times overwhelmed by the requests for chart and script re-writes and
are frustrated by the perceived need to do both”.
“GPs report frustration at and with repeated requests for scripts that have already
been written and perceived or real inefficiencies (both RCF and pharmacy) in the
recording and monitoring of script requests and the processing of scripts once
received by the pharmacy. Those GPs who have addressed these issues with the
relevant pharmacy services report improvements in the streamlining of these
processes and therefore a cessation of the need for ‘owed scripts’”.
“Timeliness of requests for scripts is also an issue and while the flowcharts appear
to address the multidisciplinary complexities, this area appears to cause much
frustration”.
Four GPs felt that the prescription management recommendation did not require any
addition or change. Suggested improvements to the model were:
Ensuring reminders were timely (from the GP’s perspective).
A discussion between relevant parties prior to commencement, determining the
best timing of reminders.
Another GP commented that there was also a need for the RCF to know what
prescriptions were due, so that when the GP visited the RCF, these could be requested
at that time, not the day or so later when the pharmacy generated the reminder. At the
present time, when prescriptions are stored outside the RCF, the GP has no record of
when the last prescription was ordered and cannot monitor prescription usage as there
is no record of prescription written unless these were done on the computer in the
doctor’s surgery. Electronic communication between the GP, the RCF and pharmacy
would improve the co-ordination and continuity of prescriptions.
Given variation in GP and pharmacy practices, it is likely that policies and procedures
related to medications in DAAs need to be explicit and to accommodate preferences
where possible, recognising that the goal is a safe, effective and efficient service (from
a whole of health system perspective).
The comments made by respondents to the expert panel questionnaire about efficiency
in pharmacy procedures are to be found in section 5.2.2.7 as these recommendations
are essentially the same in the two models.
5.2.3.8 Facility receipt of packed medication
The RCF best practice model includes a recommendation that an RN check packed
medicines against the medication chart at the time of delivery for any packs where
residents are self-medicating or where ENs or PCAs assist residents to take their
medications from the packs. This check reduces the risk of errors and is an opportunity
to monitor and improve service quality.
There were mixed views on this recommendation. Five felt that they already had
procedures in place that addressed the recommendation or did not have PCAs or self-
medicating residents. Four DONs indicated that the impact of this recommendation was
cost for staffing e.g. extra workload when an endorsed (to give medications) EN was on
duty or costs in the order of 8-12 hours/week @ $25. One RCF had an alternative audit
Quality Medication Care Group, School of Medicine, University of Queensland150
process in place where 10% of medication deliveries were checked against the
medication chart. Another RCF had said minimal changes would be needed as a
medication endorsed EN already checked self-medicating residents on a weekly basis
and should check the DAA at the same time, but a tighter policy might be required to
endure this was done. One DON felt that it was not necessary to do the second check
because:
It duplicated what the pharmacist did.
The RCF had procedures clearly stating that RNs, ENs and PCAs must check
packs against the drug order before administering drugs plus ENs had proven
competencies and PCAs (in low care) had been credentialed.
5.2.3.9 Role of GPs in RCF DAA best practice
The RCF DAA best practice model involves a resident’s GP in a number of key steps.
Community pharmacists were asked to what extent those GPs the pharmacist
interacted with in relation to DAAs were already following the best practice processes
described in the preliminary model. One pharmacist said GPs rarely followed best
practice processes because there were “too many different GPs wanting to do their
own thing”. Two pharmacists indicated that GPs sometimes followed best practice, one
saying that the main problem was a lack of awareness among GPs that a prescription
was needed to dispense medications not just an order in the RCF medication chart.
When asked about the feasibility of the recommendations involving GPs, one
community pharmacist said they were not feasible, although the GP role was crucial to
the delivery of a safe, effective and efficient DAA service. This pharmacist suggested
including regular meetings between the GPs, pharmacist and senior RCF staff as a
change to the best practice document to improve the interaction between pharmacists
providing DAAs to a RCF and the visiting GPs. Another pharmacist did not offer a
judgment on feasibility but that she would like to see the GP participate as per the
model (having previously noted the large number of GPs pharmacists must deal with
makes it exceptionally difficult to get all to comply). The other pharmacist felt that the
GP role was limited to writing medication charts and prescriptions, and the “most GPs
are lukewarm about DAAs”.
In the RCF best practice model, there is no requirement for a GP to approve a
resident’s pharmacy packing profile prior to the pharmacy packing the patient’s
medicines into a DAA. GPs were asked about the advantages and disadvantages of
GP approval prior to packing. The additional workload and possible delays were GP
approval to be required was balanced against the opportunity to double check a
medication regimen (Table 5.6).
Another group of GPs also commented on caveats on the GP role as the prescriber,
saying “while GPs may recommend a dosing regime for the patient/resident, this
remains dependant on the availability of nursing staff to administer the medication “.
The comments on this topic from respondents to the expert panel questionnaire can be
found in section 5.2.2.10. One respondent made a comment specific to a GP’s role in
RCFs:
“I feel that the GP panel idea for RCFs is a poor joke and should be reconstructed
to involve pharmacists as well and to extend to all managed care practices like
DAAs, chronic disease care programs, etc. It is under funded and poorly designed
and will fail in its present format”.
Quality Medication Care Group, School of Medicine, University of Queensland151
Table 5.6 GP views on the advantages and disadvantages of GP approval of a
pharmacy packing profile prior to packing for RCF residents
Advantages Disadvantages
Another point of cross-checking to minimise errors
Finding time for the doctor to approve profile may hold up process
Interaction may facilitate learning for both GP and pharmacist Professional assurance that it is being done properly
Coordinating times would be difficult and may delay commencing medication
Ability to check times and frequency of dose Very time-consuming for doctors with more than 10 nursing home patients
Place responsibility on doctor More work Less errors and wastage Nil significant Time consuming and with adequate
communication at earlier stages, unnecessary
5.2.4 IMPLEMENTATION ISSUES
Community and hospital pharmacists, DONs, GPs and the expert panel were asked to
identify the main barriers and implications to implementing the recommendations in
their DAA service, plus any strategies or support that would assist widespread adoption
of the best practice model.
5.2.4.1 Main barriers to the models
Among community pharmacists, no major problems were envisaged (3 responses) for
the community best practice model, although one community pharmacist felt education
was needed to stress to all parties the importance of best practice and the need to
work better together. Possible barriers identified by 4 other pharmacists were:
Finding time to implement the model (2 responses).
Payment to GPs for input.
Patient willingness to pay and accept the service.
Increased paperwork taking up more time. A way to avoid duplication of information
was needed.
Gaining GP co-operation.
Trying to communicate with large numbers of GPs.
Gaining co-operation from hospitals.
Lack of real remuneration.
The lack of a fee structure to pay for GP and pharmacy time was a barrier also
identified by the GP who reviewed the community best practice model.
From a community pharmacist perspective, the main barriers to and implications for
implementing the RCF best practice recommendations were seen as:
Variable compliance by GPs (2 responses) “the biggest hurdle is likely to be getting
the doctors to agree”.
RCFs saying they could not afford staff time to undertake the activities in the model.
The time it would take.
The cost of extra pharmacist cover in the pharmacy while a pharmacist visits the
RCF.
In addition to those barriers already mentioned in sections 5.2.2.1-5.2.2.3 and 5.2.2.5,
the main barriers and implications to implementing the best practice recommendations
in the respondents’ hospital pharmacy practice were:
Quality Medication Care Group, School of Medicine, University of Queensland152
Resources: Time and staffing, therefore costs. Concerns over who pays, who will do
the work and cost shifting.
All parties have to be motivated; only as good as the weakest link.
That it was not feasible for the hospital pharmacy to contact the community
pharmacy of each patient admitted who used a DAA as it was not always possible to
know a patient used a DAA or which pharmacy packed it, for instance. [Note: the
tripartisan agreement and patient template were included in the best practice models
to overcome these problems].
The amount of potential paperwork. The formal patient DAA needs assessment was
another form to fill in while the tripartisan agreement was cumbersome; both seemed
time-consuming.
The need for a database of community pharmacies.
A formal process may delay the start of the services. An outreach pharmacist would
often set up a DAA or organize it with a community pharmacy at the time of an
outreach visit.
The need for a GP to agree to DAA filling after discharge.
Hospital pharmacists were asked whether the DAA best practice models integrated
with hospital continuity of care initiative. All five who responded indicated that DAA best
practice models did integrate, that they were an extension of the APAC continuity of
care model currently being implemented in hospitals and supported communication
between parties and advanced planning for discharge:
“Continuity of patient care depends heavily on transfer of accurate and
comprehensive medication related information in a timely manner. Having a
structured process that dovetails between healthcare settings is crucial.”
“As the use of electronic records progresses to some community providers, this
could be extended to community pharmacy”.
However, one pharmacist from a small regional hospital had practical reservations
“until staff issues improve, the models are a nice goal rather than complementing local
hospital practices”.
From an RCF perspective, the main barriers to and implication for implementing the
recommendations were seen as:
Time needed (2 responses).
Staff attitudes (2 responses) e.g. staff dislike of change in any process.
Unclear roles in the models.
Awareness of the legal implications.
Acceptance of the model by the contracted pharmacists. The pharmacist may
decide to increase the fee to the RCF to cover costs.
The model shifts the responsibility from the facility. Pharmacy is already stretched
and doesn’t know the resident.
That effective communication is required particularly with GPs and hospitals.
Poor facilities (e.g. lack of good fax with electronic memory) and practices (e.g.
poor recordkeeping about compliance).
The need for good working relationships with GPs and community pharmacy. In
some regional/remote areas, there is limited choice in pharmacists and GPs if
communication and relationship problems arise.
Need to have better ways of identifying tablets and capsules in packs e.g. detailed
and readable description of contents or MIMS on all computers [this text has photos
of, or shows the markings on the various products].
Quality Medication Care Group, School of Medicine, University of Queensland153
From the GP perspective, there were fundamental concerns about whether DAAs
should be used at all (irrespective of best practice or not) with some GPs citing lack of
proof that the benefits exceeded costs. Two GP were concerned about wastage of
previously packed medication when packs were changed and that a medication change
takes too long to implement if repacking is involved. One of these GPs also identified
problems when non-packed medications are forgotten (e.g. schedule 8 medications or
medications not suitable for packing). A different GP had experienced difficulties
because the pharmacy who was contracted to the facility and packed the DAAs were
located a considerable distance away and could not provide a timely service when
medication was available when needed. The pharmacy was seen as unwilling to deliver
medication the same day and the RCF did, at times, obtain medication from a local
pharmacy. Another GP felt that the cost to the community might be “better spent by
employing more RNs, not less i.e. give RNs the same duty of care they have in the
hospital setting”.
The main barriers to and implication for implementing the RCF best practice
recommendations were seen by GPs as:
Time constraints (2 responses).
The document itself was too much for a busy GP to read but
“the RCF should read it”.
There would be problems if each RCF had a different guideline. One document and
procedure for the whole state would be better.
Difficulty for pharmacy dealing with multiple doctors.
Computers in nursing homes would make it easier to keep track of medication
dosage changes.
Inability/unwillingness of the pharmacy staff to speak with the GP.
Lack of co-operation from some practitioners; “what will be done if practitioners
choose not to be involved?”.
Two GPs felt that there were no real barriers to implementation. One GP suggested
that a Medicare fee for participating in DAA best practice would seem appropriate.
Respondents to the expert panel questionnaire indicated several ‘main’ barriers to
implementation. Lack of or inadequate funding was cited as a main barrier by 4
respondents:
“Lack of commitment on the part of pharmacists for a poorly-remunerated service”.
“Funding issues are the main barriers to both community and RCFs. Only the most
essential service will be delivered (i.e. supply of the medications) if funding is not
available for the required expanded service with quality safeguards”.
“Payment for service is a major issue and may determine the quality of the service,
particularly in RCFs. Currently with a free market situation the supply of DAAs to
RCFs has resulted in a substandard service in many cases due to a low or no price
demanded by the RCFs. Many pharmacists who are prepared to supply a quality
service are unable to do so at the price demanded and therefore opt out of
servicing RCFs. …. Pharmacists who are supplying at no or low cost may be forced
to adopt poor standards and inadequate staffing to reduce their costs”.
“The fact some providers do not change is not an indication that the service should
not be paid for. We all have a right to fair remuneration for services provided”.
One respondent suggested that the costs to pharmacy of the additional activities
performed to ensure a quality service is delivered (i.e. safe, effective and timely, and
efficient) could be averaged across the fee when the fee structure was developed.
Quality Medication Care Group, School of Medicine, University of Queensland154
Four respondents to the expert panel questionnaire identified general lack of
awareness of what was involved in DAA service (i.e. what the pharmacist contributed
and what each professional’s obligations were):
That DAAs were essentially a risk reduction strategy, not just a supply service.
“There needs to be acceptance by all parties that the pharmacist is the key player
in medication management”.
“Professionals understanding the costs involved at each stage and patients and relatives understanding the work and costs involved in providing competent medication management”.
“RCFs understanding the legal requirements of the medical and pharmaceutical
professions”.
One of these respondents also identified the need for “tangible evidence that it was a
good thing to be doing for everybody” as a barrier.
The need for improved communication between health professionals was felt to be a
main barrier by two respondents to the expert panel questionnaire. “In the community
setting up and maintaining the tripartisan communication of the ‘living prescription’ is
vital and requires good communication between all parties”.
Two respondents felt that any suggestion that the GP should control access to DAA
services was likely to make the models unworkable (increasing delays and costs).
Two respondents felt that the comprehensive nature of the document and what was
required for best practice had the “potential to ‘scare’ local providers off service
delivery”, while the potential administrative costs to pharmacy of the model (and current
practices) were a barrier cited by two respondents to the expert panel questionnaire:
In the community “It may be [that] a threshold has to be reached to fully implement
otherwise the whole thing may be administratively heavy”.
In RCFs a major barrier is “the costs of administrative overheads in relation to
prescription management”.
One respondent commented that “establishing a good practice model also needs a
good business model attached as for many pharmacists it is something they feel they
should do for their patients but in reality it costs them money”. Two other respondents
suggested that a standardised approach with standard templates (including electronic
versions) with “appropriate training and resources” was needed to ensure smooth
implementation. Standard forms or frameworks suggested for the community model
included:
An initial assessment form (this could include similar assessment questions to
those used in the patient survey in the Phase 2 DAA study and include medication
knowledge at baseline) and could be recorded on-line or in some central system.
A letter for the pharmacist to advise the doctor that their patient was to commence
a DAA. This would include the initial template of the patient’s current medications
completed at assessment to allow the first concordance check with GP records.
A communication framework (Pharmacist-doctor/doctors and pharmacist-
patient/carer with feedback).
A re-assessment at 6 months form that would include similar key indicators from
the initial survey plus a check on the dispensing computer of repeat intervals. In this
assessment, concordance between pharmacy records and patient-reported
medication use would be performed.
Quality Medication Care Group, School of Medicine, University of Queensland155
A 6 month concordance check with doctor form where the findings of the pharmacy-
patient concordance check were reported.
The DAA patient profile maintained by the pharmacy.
The patient template. It was suggested, however, the combining the role of a
packing template and a communication/notification of changes function may
compromise both functions. A separate ‘communication of regimen change form (a
doctor note to pharmacist not via the patient) may be better so that the patient held
record is a different form.
A working party would be needed to develop and test these tools before widespread
implementation.
The need for a significant change in hospital practices especially at discharge was
identified as a barrier by one respondent;
“Hospital discharge requires major changes to hospital systems in relation to DAAs
as part of co-ordinated care of the patient. There needs to be far better
communication between GPs, community pharmacists and hospital staff and
management procedures set up across all the times when discharges and
admissions occur, not just Monday to Friday from 9 to 5. Patients may be assessed
in hospital and discharged on a DAA with a referral to pharmacists in the area able
to continue the service”.
Another potential barrier to implementation of the best practice model was the need to
implement accountability and quality safeguards. The lack of mechanisms, capacity or
a willingness to enforce quality standards were cited as potential problems. The
enforcement of quality standards was considered essential by one respondent.
“The inadequacies of a substandard service and the potential health and safety
risks to the residents may be picked up by aged care assessors [in RCFs], but
medication management is just one of 44 standards”.
Mechanisms or strategies suggested as supporting accountability and quality
safeguards were:
To incorporate the best practice approaches to nursing home accreditation
requirements and also as mandatory standard for community pharmacies via
QCPP as both involved external assessment.
To strengthen the emphasis on the quality assurance process (e.g. including the
QA cycle into Figure 4.5 and Figure 4.8) as the “QA process is integral to the
acceptance and sustainability as well as risk minimisation”.
“If remuneration is tied to accountability there may be less chance people will put in
unrealistic tenders [for RCF DAA provision]”.
The final implication for the best practice model raised was that the notion of ‘best
practice’ was not static and would change with time. One respondent to the expert
panel questionnaire identified a need to future proof each best practice document by
adding a recommendation that the models be continually reviewed in the light of any
streamlining and, technological and legislative changes.
5.2.4.2 Supporting widespread adoption
Strategies to support widespread dissemination of the community best practice model
suggested by community pharmacists were:
Hard work by the PSA, the National Prescribing Service (NPS) and the Guild.
Making best practice a QCPP requirement.
Quality Medication Care Group, School of Medicine, University of Queensland156
Making best practice a requirement for government reimbursement.
Pharmacy bodies to provide documentation to support the model practices.
Better software for community use.
Lobbying the government to provide a fee for the service.
To support implementation of the RCF model, two community pharmacists suggested a
need for government funding as a strategy necessary for widespread adoption. One of
these pharmacists indicated government funding would be needed “if the facility will not
pay”. Implementation would also be supported if the systems implemented were
streamlined including the automation of the doctor prescription request process.
From the hospital pharmacist perspective on the implementation of the continuity of
care part of the models, the two respondents from organizations representing hospital
pharmacists emphasized stakeholder involvement and extensive consultation.
Suggested parties include state health departments, a GP advisory council, Divisions
of GP, consumer groups, APAC, SHPA, PSA and the Pharmacy Guild. An integrated
and collaborative approach between the various pharmacy stakeholders (e.g. PSA,
SHPA, AACP and PGA) was also called for to minimise responses like “I’m not going to
be told what to do by…”. A top down approach was suggested to facilitate systematic
implementation that dovetailed with the current medication safety movement in health
care. Other hospital pharmacists suggested:
Incentives such as remuneration for DAA packing (2 responses).
Establishment of a community pharmacy database that summarized the key DAA
service aspects (e.g. that the service was provided, details of the pack type used
and usual packing schedule) and that is up-to-date and widely available to
hospitals.
Promotion to patients of the importance of a medication list.
Support strategies suggested by DONs for the RCF model were:
Education of those involved (3 responses) e.g. training workshops especially for
hospital staff, RCF, pharmacy and discharge RNs.
Peak bodies (2 responses) (e.g. Queensland Nurses Union, Aged Care
Queensland, unions) who were said to be very open to supporting best practice
models.
Introduction of the model in-house through the medication committee and joint
homes medical advisory meetings to encourage co-operation of GPs and
pharmacists.
One DON felt that the best practice model should not be a stand alone document
and must be supported by policies and procedures of the individual RCFs.
The facility could pay for the packing service. This DON indicated that their
pharmacist provided an extended and excellent service (including profile
maintenance, work on the medication committee, liaising with doctors) but did not
get paid for packing DAAs, his time or the service.
Respondents to the expert panel questionnaire suggestions of strategies that might
lead to widespread adoption of the best practice models included funding, integration
with existing program (e.g. aged care standards and assessment, QCPP) and systems
(e.g. dispensing software), wider stakeholder consultation and, involvement and
support of peak bodies.
Quality Medication Care Group, School of Medicine, University of Queensland157
Funding or adequate payment for the service was felt to be “the single most relevant
criteria that will define the extent and quality of the service provided is if the pharmacist
receives a payment from the RACF to offset the extra costs involved”. To support this,
one respondent indicated that the pharmacy peak bodies “need to ensure peak RCF
and doctor groups become aware of what is actually involved in providing a DAA
service”. Reform of other aspects of the payment system were also suggested
including how pharmacists seek payment for these services (i.e. charging practices)
and how the PBS funds the supply of medications in RCFs and indeed, in other chronic
care situations (i.e. use of a medication chart as a PBS claiming mechanism).
Integration with many programs or guidelines was suggested: PSA standards and
guidelines, APAC guidelines and QCPP, Aged Care Standards.
“QCP accreditation for DAA service is vital, and could provide govt with assurance
that capitation fee is justified and cost-effective”.
Updating the PSA protocols (guidelines and standards), the development of standards
with standard templates or tools and including a list of accredited DAA pharmacies on
Guild and PSA web sites were also suggested to support this integration as was
appropriate training and other resource development.
Integration with existing IT (e.g. dispensing, prescribing and DAA software) and future
ICT (information and communication technology) initiatives (e.g. HealthConnect and
SmartCard) would also support wider adoption of best practice models
“It needs a great deal of involvement from the software suppliers of dispensing
programs and packing software as I am sure that better integration of packing profiles
and medication profiles and a lot of the QC reporting could be performed more easily
than is current. For example in the packing process the batch number and expiry dates
should be able to be entered for the current bottles being used.”
Wider stakeholder consultation and involvement would be needed for more complete
buy-in including “meetings at the highest levels”. Stakeholders and peak bodies
suggested for inclusion were:
Divisions of General Practice.
The Standards Agency (for aged care).
Pharmacy organisations (including PSA, QCPP, the Australian Association of
Consultant Pharmacy (AACP) and the Australian College of Pharmacy Practice and
Management (ACPPM)).
TGA with the involvement of bodies like the Australian Self-Medication Industry
(ASMI) would be needed to address the true medication issues like stability in
DAAs.
Consumer bodies including those for carers.
The Australian Government Department of Health and Ageing.
The Department of Veterans’ Affairs.
State health departments and other hospital organisations.
Organisations concerned with healthcare-related safety, quality and QUM.
In all, the introduction of the best practice model represents a substantial change so
“change management strategies need to be considered and implemented”.
Quality Medication Care Group, School of Medicine, University of Queensland158
PART B - ECONOMICS OF PROVIDING DAAS TO
COMMUNITY PATIENTS
6. COMMUNITY PATIENT FOLLOW-UP
6.1 METHODS
6.1.1 INTERVIEW/QUESTIONNAIRE DEVELOPMENT
Telephone survey instruments were developed to measure quality of life, willingness to
pay, direct consequences of Adverse Drug Events (ADEs) and heath resource use.
Validated instruments used in the patient interview were the EQ-5D (Interview Part C)
and the EQ-VAS (Interview Part C). All other questions were adapted from literature
and our own previous work (Ientile et al. 2004).
Contingent valuation methods were used to assess willingness to pay in this phase.
We utilised an open-ended approach which involved directly asking participants their
maximum willingness to pay given three different attributes of DAAs. The same method
was used in Phase 2 but in that case only one attribute of DAA use was highlighted.
While this method suffers from a number of biases, it was preferable to other
approaches such as payment scales, dichotomous choice techniques and bidding
game formats (Bayoumi 2004; Ryan 2004; Shiell 2003; Whynes 2004). The direct
approach was also thought to be the simplest to explain to the participants, easier to
use over the telephone and translatable to a written survey format.
A mail-out version of this survey was produced to maximise response rates for
participants who did not wish to complete the interview over the phone. Both the
telephone interview and mail-out version have been included in Appendix H.
6.1.2 PARTICIPANT FOLLOW-UP
The participants were recruited in Phase 2 through their community pharmacy. For Phase 3, each Phase 2 participant was sent a letter which detailed the Phase 3 study and included a visual analogue scale to help complete the telephone interview (see Appendix H). This letter was then followed up with a phone call by project staff further explaining the Phase 3 study and obtaining verbal consent to conduct the telephone interview (see telephone interview script in Appendix H).
Where the participant was unwilling to complete the telephone interview they were asked whether they were willing to do a mail-out survey. Participants who received a mail-out survey were followed up with a phone call to increase response rates.
Where the participant did not complete the interview, an outcome or explanation was
recorded (see Figure 6.2). The results of the follow-up were entered into a purpose-
designed Microsoft Access database, which was prepared to capture all information
collected via the patient interviews and questionnaires. Checks of data validity have
been undertaken and some open-ended responses were recoded prior to analysis.
To examine the predictors of outcome at 1 year follow-up, the characteristics recorded
in Phase 2 were compared for the outcome groups at Phase 3 (still living in the
community, dead and RCF admission).
Quality Medication Care Group, School of Medicine, University of Queensland159
6.2 RESULTS
6.2.1 RESPONSE RATES AND OUTCOMES
Of the original 342 community patients in Phase 2 that could be matched to patient
consent forms and contact details, 196 (57.31%) completed the follow-up interview. Of
these patients, 149 (76.02%) completed the interview via telephone and 45 (22.96%)
completed or partially completed the written survey. A further 2 (1.02%) partially
completed the telephone interview (see Figure 6.1).
refused to participate
lost to follow-up
mail survey returned complete
mail survey returned partly completed
interview partly completed
interview completed
0 30 60 90 120 150
Frequency
refused to participate
lost to follow-up
mail survey returned complete
mail survey returned partly completed
interview partly completed
interview completed
0 30 60 90 120 150
Frequency
Figure 6.1 Results of the follow-up of 342 community patients from Phase 2
Differences in response rates were noted between patients using DAAs packed by the
pharmacy, patients using original packs and patients packing their own DAAs (see
Figure 6.2). Overall, pharmacy supplied DAA patients were more likely to be lost to
follow-up due to death, RCF admission and ill health.
Given the differences in rates of death and RCF admission at follow-up (approximately
1 year after the Phase 2 data collection), adjustment for heath status measures
collected in Phase 2 was be required.
Quality Medication Care Group, School of Medicine, University of Queensland160
Pharmacy supplied DAA group
Completed 80 (47.3%)
Refused 30 (17.8%)
Lost to follow-up 59 (34.9%)
• Deceased 16 (9.5%)
• RCF admissions 14 (8.3%)
• In hospital 4 (2.4%)
• Too sick 13 (7.7%)
• Unable to contact 12 (7.1%)
DAA not packed by pharmacy
Completed 27 (73.0%)
Refused 4 (10.8%)
Lost to follow-up 6 (16.2%)
• Deceased 3 (8.1%)
• RCF admissions 0 (0%)
• In hospital 0 (0%)
• Too sick 0 (0%)
• Unable to contact 3 (8.1%)
Original Packs (no DAA)
Completed 89 (65.4%)
Refused 26 (19.1%)
Lost to follow-up 21 (15.4%)
• Deceased 6 (4.4%)
• RCF admissions 1 (0.7%)
• In hospital 1 (0.7%)
• Too sick 5 (3.7%)
• Unable to contact 8 (5.9%)
Phase 3 follow-up of community patients N=196 completed interviews (57.3%)
Phase 2 community patients with consent formsN=342
Pharmacy supplied DAA n=169 (49.4%)
DAA not supplied by pharmacy n=37 (10.8%)
Original Packs (no DAA)n=136 (39.8%)
Figure 6.2 Patient outcomes for the Phase 3 community patient follow-up
6.2.2 CHANGES IN DAA USE STATUS
In the course of the year between the Phase 2 interview and the Phase 3 follow-up, 18
participants changed DAA use status. The majority of these changes were patients
who were using original packs who began to pack their own DAA (n=14, 77.8%). An
additional four participants who had been using original packs began having their
medicines packed in a DAA by the pharmacy (see Table 6.1). The main reasons for
starting to use a DAA included: not wanting to forget doses, making a complex
medication regimen simpler and in response to a serious health event such as
hospitalisation or on the recommendation of a doctor, hospital pharmacists or
community pharmacist.
Table 6.1 Comparison between DAA use in Phase 2 and at Phase 3 follow-up
Phase 2 DAA status Phase 3 DAA status DAA not packed
by pharmacy Original Packs Pharmacy packed
DAADAA not packed by pharmacist 28 14 0 Original Packs 0 70 0 Pharmacy packed DAAs 0 4 80
Shading indicates no change
6.2.3 COMMUNITY PATIENT MEDICATION USE
Patient reported medication use for the four weeks prior to the follow-up interview was
consistent with medication use data collected in Phase 2 home visits (see Table 6.2).
The number of regular medications used in the four weeks prior to follow-up was
significantly correlated (p<0.001) with the total number of current medicines and current
solid medicines found in patient homes at Phase 2 (Pearson’s correlation r=0.443 and
0.432 respectively).
Quality Medication Care Group, School of Medicine, University of Queensland161
Table 6.2 Community patient medication use in Phase 2 and Phase 3
Community Patient Medication Use N Mean Median Range
There was also no difference between groups on the number or types of symptoms
experienced in either the four weeks or the 2 days prior to follow-up.
Table 6.6 Comparison between groups on symptoms experienced in the last 4 weeks
or 2 days
Non-pharmacy DAA
Original Packs Pharmacy packed DAA
Symptoms and conditions
N % N % N %
Chi-Square
Falls in past 4wks 1 3.57 12 13.95 14 18.18 0.087Falls in past 2 days 0 0.00 1 1.16 5 6.49 0.249Dizziness in past 4 wks 11 39.29 24 27.91 29 37.66 0.326Dizziness in past 2 days 6 21.43 14 16.28 19 24.68 0.644Headaches in past 4 wks 13 46.43 32 37.21 24 31.17 0.402Headaches in past 2 days 5 17.86 12 13.95 8 10.39 0.685Fever in past 4 wks 5 17.86 11 12.79 4 5.19 0.129Fever in past 2 days 4 14.29 3 3.49 2 2.60 0.384Depressed past 4 wks 9 32.14 37 43.02 36 46.75 0.335Depressed past 2 days 5 17.86 19 22.09 17 22.08 0.875Pain in past 4 wks 23 82.14 66 76.74 52 67.53 0.400Pain in past 2 days 16 57.14 54 62.79 46 59.74 0.153Fatigue in past 4 wks 17 60.71 50 58.14 47 61.04 0.845Fatigue in past 2 days 13 46.43 35 40.70 38 49.35 0.773Shortness of breath 4 wks 14 50.00 36 41.86 34 44.16 0.683Shortness of breath 2 days 8 28.57 25 29.07 22 28.57 0.348Nausea, constipation, diarrhoea 4 wks
12 42.86 32 37.21 29 37.66 0.868
Nausea, constipation, diarrhoea 2 days
5 17.86 16 18.60 13 16.88 0.770
Anxious past 4 wks 9 32.14 40 46.51 34 44.16 0.337Anxious past 2 days 8 28.57 29 33.72 24 31.17 0.631Vision problems past 4 wks 14 50.00 42 48.84 40 51.95 0.961Vision problems past 2 days 13 46.43 37 43.02 37 48.05 0.780Trouble sleeping past 4 wks 17 60.71 48 55.81 38 49.35 0.651Trouble sleeping past 2 days 14 50.00 32 37.21 29 37.66 0.706Memory trouble past 4 wks 16 57.14 29 33.72 34 44.16 0.071Memory trouble past 2 days 10 35.71 23 26.74 26 33.77 0.237Hearing problems past 4 wks 13 46.43 32 37.21 39 50.65 0.189Hearing problems past 2 days 13 46.43 29 33.72 33 42.86 0.154
On the quality of life measure EQ-VAS (possible range 0-100) there was no difference
between groups (see Table 6.7). There were also no significant differences based on
the method of data collection (i.e. whether participants completed this measure in
questionnaire form (indicating their response by marking the visual analogue scale),
Quality Medication Care Group, School of Medicine, University of Queensland164
over the phone with the EQ-VAS picture or over the phone without the EQ-VAS picture
(F=1.036, p=0.368)). Population comparisons for patients aged 70-79 (based on New
Zealand population statistics, as Australian data is not currently available) indicate
lower self-ratings of health in this sample. Devlin et al (Devlin et al. 2000) reported
mean EQ VAS scores of 76.8 (std dev 19.0) for respondents aged 70-79.
Table 6.7 Mean EQ-VAS scores for patients from the different DAA groups
N Mean Median Range S.E. Mean ANOVA
Non-pharmacy DAA 28 64.39 70 30-100 3.41
Original Packs 88 66.76 72.5 5-100 2.05 P=0.463
Pharmacy packed DAA 77 63.17 70 2-100 2.10
Pharmacy packed DAA patients were significantly more likely to have some problems
with mobility, personal care and conducting their usual activities compared with original
packs and self-packing DAA patients (Table 6.8).
Table 6.8 EQ-5D responses by DAA group
Non-pharmacy DAA
Original Packs
Pharmacy packed DAA
Chi-square*
Dimension Response
N % N % N % Sig.
no problems 16 57.14 45 50.56 19 24.68 0.001
some problems 12 42.86 44 49.44 55 71.43 0.001†
Mobility
more problems 0 0 0 0 3 3.90
no problems 22 78.57 79 89.77 47 61.04 <0.001
some problems 4 14.29 6 6.82 22 28.57 <0.001†
Self care
more problems 2 7.14 3 3.41 8 10.39
no problems 13 46.43 53 60.23 23 29.87 0.001
some problems 14 50.00 30 34.09 48 62.34 <0.001†
Usualactivities
more problems 1 3.57 5 5.68 6 7.79
no problems 9 32.14 34 38.64 25 32.05 0.627
some problems 18 64.29 44 50.00 41 52.56 0.345†
Pain
more problems 1 3.57 10 11.36 12 15.38
no problems 20 71.43 62 69.66 45 57.69 0.248
some problems 6 21.43 26 29.21 30 38.46 0.123†
Anxiety/Depress-ion more problems 2 7.14 1 1.12 3 3.85
* where dimension recoded as no problems or any problems † Pharmacy DAA vs OP
The results across the EQ-5D dimensions for original pack patients was consistent with
the percentage of reported problems for the New Zealand population of patients aged
70+ (Devlin et al. 2000) (Figure 6.3). However for the pharmacy packed DAA group,
the percentage of patient reporting problems was substantially higher. The finding that
pharmacy packed DAA patients have a poorer health status with respect to needing
greater assistance is consistent with the findings from Phase 2.
Quality Medication Care Group, School of Medicine, University of Queensland165
0
10
20
30
40
50
60
70
80
Mobility Self care Usual activities Pain Anx/Dep
EQ-5D Dimensions
% r
ep
ort
ing
pro
ble
ms
DAA not packed by pharmacy Original Packs
Pharmacy packed DAA Population comparison (70-79 years)
Figure 6.3 Sum of the proportion of level 2 and level 3 problems for each of the 5 EQ-
5D dimensions for the three DAA status groups and for a comparison
sample aged 70-79 (Devlin et al. 2000)
The scores on the five dimensions were transformed using the standard EuroQol
algorithm to produce a weighted health index score (range -0.2 to1). Based on this
measure, pharmacy packed DAA patients had a significantly poorer quality of life
(p<0.001) than original pack or non-pharmacy packed DAA patients (see Figure 6.4).
Interestingly, the Weighted Health Index scores were significantly correlated with the
OARS-IADL score collected in Phase 2 (Rho=-0.405, p<0.001, n=174).
0.6
0.62
0.64
0.66
0.68
0.7
0.72
0.74
0.76
0.78
0.8
0.82
Pharmacy DAA Non-pharmacy DAA Original pack
Weig
hte
d H
ealt
h In
dex
Figure 6.4 Comparison of Quality of Life measured by EuroQol between the DAA
groups
Quality Medication Care Group, School of Medicine, University of Queensland166
6.2.5 HEALTH SERVICE USE
Pharmacy packed DAA patients were more likely to have used support services in the
four weeks prior to follow-up, and were more likely to have been hospitalised and/or
used respite services in the 12 months prior to follow-up, than original pack and self-
packing DAA patients (see Table 6.9 and Table 6.11). The costs reported by patients
did not differ (Table 6.11).
Table 6.9 Percentage of DAA status groups reporting health service use
% or group using health service Used service in past 4 weeks
DAA not packed
by pharmacy
Original
Packs
Pharmacy
packed DAA
Chi-
square
GP services 75.0 80.9 79.7 0.802
Community Nursing 7.1 13.5 15.2 0.517
Other health professionals 32 29.2 30.4 0.955
Emergency Department treatment 0 6.7 7.6 0.144
Hospital clinic or outpatients
treatment
3.6 10.1 11.4 0.404
Other health care i.e. pathology 35.7 28.1 22.8 0.405
Support services (i.e. home help,
Meals-on-wheels, home nursing)
39.3 37.1 60.8 0.006
Used service in past 12 months
Public hospital 17.9 20.2 29.1 0.302
Private hospital 10.7 22.5 20.3 0.354
Any hospitalisation 25.0 37.1 49.4 0.051
Other care facilities 3.6 0.0 8.9 0.004
6.2.6 ADR AND HEALTH CONSEQUENCES
In Phase 3, overall lower rates of adverse drug reactions (ADRs) in the previous 12
months were reported compared to Phase 2 (27.5% vs. 39.3%). While in Phase 2,
significantly fewer pharmacy packed DAA patients reported ADRs compared with
original pack patients, in Phase 3 the difference in ADRs between DAA groups was not
significant (see Table 6.10). However, there was a significant correlation between
experiencing an ADR in 12 months prior to Phase 2 and Phase 3 (Spearman’s rho=
0.216, p=0.003). The number of patients needing hospitalisation, respite care or
community nursing as a result of ADRs was too low to allow for statistical testing but
the frequency of these consequences are reported in Table 6.10 below.
Table 6.10 Patient reported adverse drug events and consequences by DAA group
Deidenfied Provider ID for prescriber if possible.
Deidentified dispensing pharmacy approval number.
7.1.2 DATA RECEIPT
MBS records for 237 patients and PBS records for 235 community patients were
retrieved by HIC. No matching data was found for 2 patients in the MBS data and 4
patients in the PBS data. An examination of the date ranges for services revealed that
the service data for a number of patients was incomplete (data was missing at the
beginning or at the end of the periods, or in some cases, in the middle).
As of 10 June 2005, data had not been received for the 71 DVA gold card holder nor
any service use by the 7 white card holders from DVA records (some data on these
patients was included in the HIC data). Despite frequent follow-up of this request by
the reachers, the data release request was misplaced and data release was not grated
in time to include veterans in this analysis.
7.1.3 DATA PREPARATION
The HIC data files were imported into a Microsoft Access database for cleaning and
aggregation and merging with Phase 2 data.
Patients with apparently incomplete MBS or PBS were excluded from the analyses
where data was missing. If either MBS or PBS data were missing, the patient was
excluded. Zero cost for MBS or PBS costs are likely to represent incomplete data
collection for these patients (since they were expected to have medicines and to see a
doctors in the time periods). These patients were also censored from analyses of
service use. Three other patients were excluded, one because HIC data was
requested for the wrong patient, another whose MBS service use (associated with
cancer treatment and including $26,517 of hospital MBS costs , 3 times the next
closest value) was an extreme outlier and another whose PBS service use was an
extreme outlier.
Variables of interest derived from the raw HIC data were:
Total number of items dispensed in year (for the cross-sectional comparison up to
and including day of home visit) Exclude ATC V04* (glucose testing strips etc),
Exclude stomal and wound dressing products (ATC V07AS and V07AY).
Sum of benefit received (cost to government).
Total number of different items dispensed.
Quality Medication Care Group, School of Medicine, University of Queensland172
Estimation of number of different regular medications – Total of number of different
items dispensed where more than 1 script per item was filled or if >1 script per item
filled, not an antiinfective agent.
Number of different prescribers.
Number of different pharmacies filling prescriptions.
Total number MBS items.
Total cost of MBS items (sum of benefit).
Total number MBS items provided in hospital (those to be included in each analysis
period that had no hospital items were coded as having zero items rather than
missing).
Total cost of MBS items (sum of benefit) provided in hospital (all those with no
costs were left null as mean costs were only to be calculated for those who were
actually incurred any costs).
Number of GP items not provided in hospital (those to be included in each analysis
period that had no GP items were coded as having zero items rather than missing)(item numbers 1-4, 20, 23-5, 33, 35-7, 44, 47, 50-54, 57, 59-60, 65, 83, 87, 89-91, 97-8, 160-162, 173,
Pharmacy DAA vs OP R Squared = 0.409 (Adjusted R Squared = 0.356)
0.595
*All models significant to the p<0.0001 level #p value for DAA and control group as an independent predictor for other covariates
The details of the cost models can be found in Appendix I.
The modelled mean costs for each group were calculated. These retransformed costs
are those predicted by the models after adjusting for covariates in the models. The
adjusted means based on the model were converted back to dollars (from Ln dollars)
as were the confidence intervals for the estimated mean. Figure 7.1 to Figure 7.6
compare adjusted costs (3 group and 2 group comparisons) with unadjusted costs in
the period.
Quality Medication Care Group, School of Medicine, University of Queensland184
Users of medicines in original packs had lower PBS costs than either users of
pharmacy supplied DAAs or non-pharmacy DAA users, adjusted or unadjusted (Figure
7.1). Adjusting for age, health status, symptom frequency, the number of current
medicines at home, the number of different dispensing pharmacies and any
hospitalisation in the previous 12 months suggested that non-pharmacy DAA users
actually had higher PBS costs than they should have given their characteristics in the
model. Pharmacy DAA users actually had slightly lower PBS costs than might be
expected given the model. However, the models explained between 30 and 35% of the
variance in PBS costs so that there may be further differences in characteristics
(unmeasured) between the groups for which adjustment would be required.
0
500
1000
1500
2000
2500
3000
3500
4000
Pharmacy DAA Non-pharmacy DAA Original pack
$ P
BS
be
ne
fits
in
ye
ar
pri
or
to h
om
e v
isit
Adjusted PBS benefits - 3 group
Adjusted PBS benefits - Pharmacy DAA vs OP
Unadjusted PBS benefits
Figure 7.1 Adjusted and unadjusted PBS costs (mean and 95%CI) for the year prior to
home visit
For GP costs (Figure 7.2), there were no significant differences between the groups
and the effect of adjusting for covariates (living alone, the number of GP visits in the
previous 2 weeks, health status, age, prior ADR, number of different prescribers, any
hospitalisation in last 12 months, gender and disorganised non-adherence) slightly
lowered OP and pharmacy DAA costs while increasing non-pharmacy DAA (i.e. given
the characteristics of the non-pharmacy DAA users, they should have incurred higher
GP costs that they did).
Adjusted MBS costs in hospital were significantly lower for pharmacy DAA users
compared to OP users (Figure 7.3). Adjusting for age, OARS IADL score, the number
of different prescribers and dispensing pharmacies and any hospitalisation in the
previous 12 months suggested that pharmacy DAA users actually had higher MBS
hospital costs than they should have given their characteristics in the model and OP
users actually had lower costs than might be expected given the model. However, the
models explained about 40% of the variance in costs so that there may be further
differences in characteristics (unmeasured) between the groups for which adjustment
would be required. Further, it is possible that uncaptured hospital costs (public and
patient/insurer paid) were not equally distributed between the groups.
Quality Medication Care Group, School of Medicine, University of Queensland185
0
100
200
300
400
500
600
700
Pharmacy DAA Non-pharmacy DAA Original pack
$ c
ost of G
P ite
ms b
enefits
in y
ear
prior
to h
om
e v
isit Adjusted GP costs - 3 group
Adjusted GP costs - Pharmacy DAA vs OP
Unadjusted GP costs
Figure 7.2 Adjusted and unadjusted MBS costs for nonhospital GP services (mean
and 95%CI) for the year prior to home visit
0
500
1000
1500
2000
2500
3000
3500
4000
Pharmacy DAA Non-pharmacy DAA Original pack
Co
st $
MB
S ite
ms in
ho
sp
ita
l in
ye
ar
prio
r to
ho
me
vis
it
Adjusted MBS Hospital costs - 3 group
Adjusted MBS hospital costs - Pharmacy DAA vs OP
Unadjusted MBS hospital costs
Figure 7.3 Adjusted and unadjusted MBS costs for services provided in hospital for
those who used these services (mean and 95%CI) for the year prior to
home visit
The costs of monitoring/pathology services that might assist monitoring of medication
management was higher in the pharmacy DAA group compared to the other two
groups (Figure 7.4). After adjusting for health status, the number of doctors seen, the
number of illnesses reported by the patient, any ADR, memory related non-adherence,
the number of GP visits in the last 2 weeks, comparative health rating, impaired IADL,
living alone and regular use of community health services, the predicted cost of
monitoring/pathology services was higher in the pharmacy DAA group than actual
costs, however the variance on costs explained by these models was the lowest
among the cost models (15-18%).
Quality Medication Care Group, School of Medicine, University of Queensland186
0
50
100
150
200
250
300
Pharmacy DAA Non-pharmacy DAA Original pack
$ c
ost of M
BS
monitoring/p
ath
olo
gy ite
ms
(no
nh
osp
ita
l) in
ye
ar
prio
r to
ho
me
vis
it
Adjusted cost of monitoring/pathology items - 3 group
Adjusted cost of monitoring/pathology items - Pharmacy DAA vs OP
Unadjusted cost of monitoring/pathology items
Figure 7.4 Adjusted and unadjusted MBS costs for nonhospital monitoring/ pathology
services for those who used these services (mean and 95%CI) for the year
prior to home visit
Other MBS costs incurred outside hospital were highest for OP users but adjusting for
health status, the number of GP visits in the last 2 weeks and regular use of community
health services brought the predicted service use for OP and pharmacy DAA users
closer together (Figure 7.5).
0
200
400
600
800
1000
1200
1400
1600
1800
2000
Pharmacy DAA Non-pharmacy DAA Original pack
Cost $ o
ther
MB
S ite
ms (
nonhopsital,
no
t G
P o
r m
on
ito
rin
g/p
ath
olo
gy)
Adjusted cost other MBS items - 3 group
Adjusted cost other MBS items - Pharmacy DAA vs OP
Unadjusted cost other MBS items
Figure 7.5 Adjusted and unadjusted other nonhospital MBS costs for those who used
these services (mean and 95%CI) for the year prior to home visit
Since it was possible that pharmacy DAA users had higher service use prior to starting
a DAA as well as more disability and poorer health, the total HIC cost (Ln PBS+MBS
benefits) in the post-DAA period (extending from 3 to 6 months after the home visit
date) was modelled with adjustment for covariates (any hospitalisation in the 12
Quality Medication Care Group, School of Medicine, University of Queensland187
months prior to the home visit, health status, the number of GP visits in the last 2
weeks, the number of current medicines in the home, gender and age) and the cost of
service use in the pre-DAA period. The model explained 56.6% of variance in HIC
costs (R Squared = 0.604, Adjusted R Squared = 0.566) and DAA use was not a
significant predictor of cost in the post-DAA period (p=0.107). The adjusted mean and
unadjusted mean post-DAA costs are shown in Figure 7.6.
0
1000
2000
3000
4000
5000
6000
Original pack Pharmacy DAA
$ M
BS
+P
BS
benefits
Baseline period
Unadjusted post
Adjusted post
Figure 7.6 Unadjusted baseline, unadjusted post-DAA and adjusted post-DAA HIC
costs (mean and 95%CI) for subset in the pre-post analysis
Baseline costs were a strong predictor of post-DAA costs (accounting for 22.6% of
variance after adjusting for other variables in the model) followed by the person’s rating
of their health status (10.3% of variance after adjusting for other variables in the
model).
8. CHARACTERISING COMMUNITY PATIENTS
WHO USE A DAA
The question of just who will benefit from using a DAA has not been fully defined. The
literature and guidelines suggest patients with a range of characteristics benefit but the
nature of “benefit” varies from the view that certain patients who are nonadherent will
benefit through to the assumption that patients who can use the device successfully will
have improved adherence. Few studies have examined benefit in terms of outcomes.
In this section, the characteristics that distinguished between Phase 2 community
patients who continued to use pharmacy-provided DAAs and those who used original
packed (OP) medicines are examined. While the study conducted in Phase 2 was a
cross-sectional study and was not designed to establish a causal link between outcome
and DAA use, the results may be helpful in developing more defined criteria for
potential community-based recipients of a DAA service. A randomised controlled trial
and/or a longitudinal study would allow patient characteristics that predict likely benefit
from a DAA to be better defined, however, a comparison of two groups of community
patients who self-selected to use either medicines in original packs or have medicines
packed by a pharmacy into a DAA can still inform the question of who would benefit
Quality Medication Care Group, School of Medicine, University of Queensland188
from a DAA. Continued DAA use by patients implies some perception of ‘benefit’ when
merely using original packs could reduce patient cost and inconvenience.
8.1 METHODS
In Phase 2, information was collected by data collectors who visited participating
community patients in their homes. The data was collected using a combination of
patient interview and patient questionnaire.
To summarise the patient recruitment (described in full in the Phase 2 Final Report,
section 3.3), patients were recruited by pharmacists who were participating in the
study. Pharmacists were asked to recruit four community based DAA users and four
matched non-DAA users. Pharmacists were given the following criteria for matching:
age, gender, number and type of medications and living arrangements. Participants
also had to be regular customers of the pharmacy, living in the community (i.e. not
residing in a hostel or residential care facility) and capable of answering interview
questions and completing a questionnaire (with the help of a carer if necessary). In
total, 353 community patients were recruited by 75 pharmacies, with eight pharmacies
unable to recruit any community patients.
Usable patient demographic (age, gender, living arrangements), clinical (use of
medications, compliance, health status and some health service use) and functional
(Instrumental Activities of Daily Living (IADL) and ability to use medicines) were
available for 348 Phase 2 community patients. The patient subset for whom models
were developed for the economic analyses included 312 patients who gave valid
consent for the retrieval of Health Insurance Commission (HIC) data and for whom
there were sufficient details to match to an identifier used by HIC for data retrieval.
8.1.1 PREPARATION OF MODEL VARIABLES
Only patient characteristics collected in Phase 2 were used in models because data
from Phase 3 would not be available for people who had died, been admitted to
residential care or were otherwise lost to follow-up. A number of dichotomous
variables were taken from the interview and questionnaires:
Living alone.
Having a regular carer.
Community health worker visits on a regular basis (e.g. nurse, meals-on-wheels, therapist).
Gender.
Any Adverse Drug Reactions (ADR) – symptoms or health problems patient thinks were
caused by medicines.
The four Meichenbaum questions: Do you ever forget to take your medications; Are you
careless about taking your medicines; When you feel better do you stop taking your
medicines; If you feel worse do you stop taking your medications.
Continuous variables were:
The number of doctors (GPs or specialists) usually seen.
The number of times a GP was seen in the last 2 months.
The number of different illnesses reported by the patient.
The number of different prescribers of current medicines (taken from packs) – median
substitution (median=1) was used for 58 cases.
The number of different dispensing pharmacies of current medicines (taken from packs).
Several ordinal variables were also used:
Quality Medication Care Group, School of Medicine, University of Queensland189
Self-rating of health (poor, fair, good or excellent).
Comparative health (worse than others, about the same, better).
Frequency of illness symptoms in the past week (4 levels: not at all to most days (4-
7days/week)) – also recoded into a high/low dichotomous variable.
Rating of illness symptom severity in the past week (5 levels: nil to severe) – also recoded
into a high/low dichotomous variable.
Some variables were calculated or recoded from interview or questionnaire responses:
Age in years was calculated as at the home visit.
The number of days in hospital and the number of admissions in the last year was recoded
to “any hospitalisation in the last year”.
Older Americans Resource Scale for Instrumental Activities of Daily Living (OARS-IADL)
score (range 0-14).
The days since the last medication change were calculated.
The total number of current medicines, the number of solid (potentially packable) medicines
and the number of non-solid current medicines were derived from the list of medicines
identified at home visit. Only two of these variables were used in a model at one.
A Principal Components analysis with varimax rotation was also conducted in relation to
several compliance measures to determine factors for participants resulting in 3 variables
(the sum of scores for the questions loading on a given factor): memory-related non-
adherence, deliberate non-adherence and disorganised non-adherence. Each variable was
later recoded as dichotomous variables: any non-adherence or no non-adherence.
The use of drugs in various therapeutic categories were aggregated from the current
medicines in the home for 306 patients (Table 8.1). These classes are indicative of mental
health problems (where noncompliance is a problem), dementia (where cognitive
impairment may affect medication management ability), physical limitations to open
medicine packs, treatments for cardiovascular diseases (hypertension and dyslipiaemia,
asymptomatic conditions where treatment reduces morbidity and mortality, and heart failure,
a symptomatic condition), diabetes type II (where good glycaemic control through good
compliance to drugs and diet prevents complications) and respiratory medicines (because
many of these are in dose forms that cannot be packed into a DAA). The drug classes,
beta blockers and drugs acting on the angiotensin system were aggregated separately as
these medicines can be used to treat hypertension and/or heart failure.
Table 8.1 Aggregated variables for use of specific classes of drugs
Drug use marker ATC codes % Frequency
Any antianxiety medicines (yes/no) starting N05B Y=13.4 Any antidementia medicines starting N06D Y=2.3 Any antidepressant medicines starting N06A Y=27.8 Any antipsychotic medicines starting N05A Y=9.5 Any medicine used to treat rheumatoid arth-ritis or Parkinson’s disease (a potential for physical problems with packs)
starting N04 or M01C, L01B01, A07EC01, L04AA11 to 14, L04AA01, L04AX01, P01BA01
Y=6.5
Any lipid lowering medicines starting C10 Y=42.8 Any oral hypoglycaemic medicines starting A10B Y=11.8 The number of respiratory medicines starting R03A or R03B or
R03CC or R03D 1=9.8; 2=4.2; 3=2.6; 4=1.3; 5=1.3; 6=0.3
Any beta blocker medicines starting C07A Y=27.1 Number of medicines acting on angiotensin II (a potential marker of disease severity)
starting C09 1=52.0; 2=5.2; 3=0.7
Any other antihypertensive medicines starting C02 or C08, C03AA03 Y=31.7 Number of other medicines used to treat heart failure (a potential marker of severity)
C01AA05, starting C03C, C03DA01
1=21.9; 2=7.2; 3=1.0
A square root transformation of age and the total number of current medicines was
used the propensity model as the economic analysis also involved multiple linear
Quality Medication Care Group, School of Medicine, University of Queensland190
regression where variables whose distribution approaches the normal distribution.
Untransformed age and numbers of medicines were used in other models as logistic
regression more robust to violations the assumption of normality for continuous
variables.
8.1.2 LOGISTIC REGRESSION MODELS
Binary logistic regression (SPSS for Windows version 12.0.1) with backward
elimination was used to develop a number of models to predict the use of pharmacy-
provided DAAs versus medicines in original packs. A significance level of 0.05 was
used for all analyses. All of the medication related risk factors were considered a priori
to potentially impact on health outcomes.
The various models included slightly different variables initially, before backward
elimination of variables until a model that was parsimonious but maximised the
explanatory (Cox & Snell R square and Nagelkerke R square) and predictive power of
the model (the percentage of cases correctly classified). The goal of the first model
used to generate propensity scores for the economic analysis was to rank cases on
variables that were related to health service use. As Phase 2 was a cross-sectional
study where causation could not be attributed, a number of other logistic models were
developed that omitted some variables that may well be the consequence of DAA use
rather than a predictor of DAA use:
Any ADR.
Number of GP visits in the last 2 months (DAA users have fewer visits just to get a
prescription written).
Number of different pharmacies and different prescribers – DAA users have a packing
relationship with a pharmacy and a prescription writing relationship with a GP to ensure
continuity of supply (often managed by the pharmacy).
The self-reported non-compliance (Meichenbaum) questions “When you feel better do you
stop taking your meds?” and “If you feel worse, do you stop taking your medicine?” were
excluded in some cases as having medicines packed in multi-dose DAAs makes if more
difficult to stop taking a specific medicine.
The time since the last medication change was excluded from some models as this
variable was missing for 49 cases. The variables included initially in each model are
shown in Table 8.2.
Table 8.2 Variables initially included in the logistic regression models
Propensity Model
Model 2
Model 3
Model 4
Model 5
Model 6
Living alone Having a regular carer Regular community health worker visits No. doctors usually seen Self-rating of health No. times a GP was seen in last 2 months No different illnesses reported by patient Patient reports any hospitalisation in last year Comparative health Frequency of illness symptoms in past week score Severity of illness symptoms in past week score Symptoms in past week (high/low) Symptom severity in past week (high/low) Any ADRNo. different prescribers (median substitution)
Quality Medication Care Group, School of Medicine, University of Queensland191
No. different dispensing pharmacies OARS-IADL score Square root (94-Age) AgeGender Total No. of current medicines Square root total No. of current medicines Total No. of current solid medicines Total No. of current non-solid medicines Memory-related non-adherence score Any memory-related non-adherence Deliberate non-adherence score Any deliberate non-adherence Disorganised non-adherence score Any disorganised non-adherence Any antianxiety medicines Any antidementia medicines Any antidepressant medicines Any antipsychotic medicines Any rheumatoid or Parkinson’s disease medicine Any lipid lowering medicines Any oral hypoglycaemic medicines No. respiratory medicines Any beta blocker medicines No. medicines acting on angiotensin II Any other antihypertensive medicines No. other medicines used to treat heart failure Days since last medication change Do you ever forget to take your medications Are you careless about taking your medicines When you feel better do you stop taking your medicines If you feel worse do you stop taking your medicines
8.1.3 NON-LINEAR MACHINE LEARNING MODELS
The logistic regression models presented in 8.2.2 assume that the relationship between
the decision to use a DAA and the predictors is linear. An alternative modelling
strategy, machine learning, was trialled to overcome the limitations of a logistic model
and to better reflect the combination of patient characteristics that might describe an
individual DAA user. The machine learning technique trialled was that of a decision
tree as this was felt to lend itself better to ‘eligibility’ assessment than other types of
machine learning.
Non-linear models were generated using the same initial combination of variables as
those included in the logistic regression models 2, 5 and 6 to illustrate the technique.
Thus , the Phase 2 data was used as an initial training set for decision tree generation.
Note that because of the cross-sectional nature of the study in Phase 2, model 2
includes variables that may be ‘caused’ by DAA rather than predicting their use;
models 5 and 6 exclude these variables.
Quality Medication Care Group, School of Medicine, University of Queensland192
An additional tree was generated that incorporated the probability score from the
logistic model 2 to examine the validity of the tree i.e. whether the tree and the logistic
regression model the same underlying construct.
All trees were generated using Ross Quinlan’s C5 algorithm, available in the See5
package for Windows on the Rulequest website
(http://www.rulequest.com/download.html).
One of the options available when building trees is that of boosting. When a tree is
generated it makes some errors of prediction on the training set. Boosting generates a
succession of trees giving higher weightings to cases that have been previously
misclassified in an attempt to rectify the errors made. When a new case is to be
classified, each of the trees generated votes for a predicted class. These votes are
counted and the final prediction is made. This gives a better overall predictive accuracy
but at the expense of interpretability as many trees are being consulted to give the final
prediction. For each tree generated a boosting of 50 was used in conjunction was cross
validation to try and assess the best possible predictive accuracy in real world
conditions.
In each case, the overall accuracy of the tree is predicted by using cross-validation.
With all machine learning techniques there is a danger of over-learning on the training
set in which case the models produced may not generalize well into the greater
community. Therefore to check this we used ten-fold cross validation. This technique
splits the data into 10 random parts. The model is trained on 9 parts and tested on the
10th. Therefore there are 32 randomly chosen test cases held out from training on the
model and used to test predictive accuracy.
For each case the pruning was left at 25%. Fuzzy thresholds were not used.
Both decision trees and equivalent rule sets were generated.
8.2 RESULTS
8.2.1 LOGISTIC REGRESSION MODEL USED TO DERIVE PROPENSITY SCORES FOR
ECONOMIC ANALYSIS
The logistic regression model (Chi-square=98, df 12, p<0.001, n=249) included the
variables collected in Phase 2 (Table 8.3) and correctly predicted 72.3% of cases
(probability cut point 0.5) as belonging to DAA (77.4%) or OP (66.1%) groups (Cox &
Snell R Square=0.326 and Nagelkerke R Square=0.436). The variables, living alone
and self-rated health status were included as they improved the predictive ability of the
model.
The logistic regression model used to adjust for between group differences in the
propensity of using a DAA over using medicines in original packs also provided
information on the characteristics of community patients electing/choosing to use a
DAA after adjusting for the effect of any covariates in the characteristics. In addition to
willingness to use a DAA (these patients made the choice to start the DAA), the
characteristics of community patients that independently predicted DAA use compared
to use of original packs were:
Quality Medication Care Group, School of Medicine, University of Queensland193
Having a regular carer increased the odds of using a DAA by 313% compared to
not having a regular carer. This determinant may point to other benefits of a DAA
in easing carer burden.
Having community health workers visit on a regular basis increased the odds of
using a DAA by 226% compared to not having a regular carer, suggesting that DAA
users make greater use of community support.
Seeing fewer doctors. For each additional doctor (GP or specialist) regularly
visited, a community patient is 24% less likely to use a DAA. This could reflect a
more interventionist approach to treatment in the less disabled OP group or that the
pharmacy is more likely to act as the primary medical care provider for DAA users.
If the patient indicated that they had been to hospital at least once in the preceding
year, the odds of using a DAA were increased by 351%, compared to those
reporting no recent hospital admission.
Community patients experiencing an adverse reaction to medications were 66%
less likely to be DAA users.
Increasing age - older community patients were more likely to use a DAA.
Community patients who were 79 years (50th percentile of the people in this study
OP, pharmacy DAA and non-pharmacy DAA) were 37% less likely than someone
85 years old (approximately 80th percentile) to use a DAA; a 69 year old (~ 20th
percentile) was 37% less likely to use a DAA than a 79 year old; a 58 year old (~ 5th
percentile) was 37% less likely to use a DAA than a 69 year old.
The number of medicines used and the number of solid medicines (that could
be packed into a DAA). The effect of these variables on the probability is opposite
and reflects the underlying effects of the number of non-solid medicines a person
might take. Increasing the total number of solid medicines increases the odds of
using a DAA but this probability is ameliorated by how many non-solid medicines
are taken. For example, moving 1 unit on the square root of the number of current
medicines at home (includes non-solid medicines) from 1.41 to 2.45 (a move from 2
to 6 medicines at home) and all of these 4 extra medicines were solid medicines,
the odds of using a DAA are 5.8 times (1.48x4 – 0.14) more likely than not using a
DAA but if one of the extra medicines was non-solid, the odds of using a DAA are
4.3 times more likely than not using a DAA. If only 1 of the extra medicines was a
solid medicine, the odds of using a DAA would be only 1.34 times more likely than
not using a DAA for a community patients where all other characteristics are
similar.
Community patients who had visited the GP more frequently in the previous 2
months were less likely to use a DAA. For each additional GP visit, the odds of
using a DAA was 21% less. Prescription management by pharmacies may reduce
the number of GP visits to obtain new prescriptions.
Having medicines prescribed by more doctors reduce the likelihood that the
community patient would use a DAA. For each additional prescriber (as reported by
the patient), the odds of using a DAA was 47% less. Again, the prescription
management by pharmacies may rationalise the prescribing responsibilities (as the
logistic regression adjusts for the number of doctors seen regularly).
Living alone and self-rated health status approach significance as DAA predictors:
Community patients who lived alone were 176% more likely to use a DAA than
those not living alone
Quality Medication Care Group, School of Medicine, University of Queensland194
Better self-rated health reduced the likelihood of using a DAA. A change in health
rating from ‘poor’ to ‘fair’ or from ‘fair’ to ‘good’ or from ‘good’ to ‘excellent’ reduced
the odds of using a DAA by 35%.
Table 8.3 Logistic regression model of using a pharmacy supplied DAA (versus OP)
Variables in the logistic regression
model
B Std.
Error
Wald
statistic
df Sig. OR OR 95% CI
Having a regular carer 1.141 0.344 10.990 1 <0.001 3.131 1.594 - 6.148
Regular community health visits 0.815 0.374 4.756 1 0.029 2.260 1.086 - 4.704
Any current antidepressants 1.480 0.451 10.784 1 0.001 4.392 1.816 – 10.623
No. other current medicines to treat CCF
0.901 0.317 8.079 1 0.004 2.462 1.323 – 4.582
Do you ever forget to take your medications
1.011 0.403 6.304 1 0.012 2.749 1.248 – 6.052
Any current anti-rheumatic or Parkinson’s disease medicines
2.080 0.909 5.238 1 0.022 8.003 1.348 – 47.506
Constant -0.715 2.128 0.113 1 0.737 0.489
Quality Medication Care Group, School of Medicine, University of Queensland198
Comparison of models 4 to 6 shows that there is more than one potential set of
characteristics to predict DAA use. Indeed, model 5 does not include a variable about
medication burden i.e., the total number of current medicines (either solid or non-solid).
The variables included and retained in the various models is summarised in Table 8.9.
Table 8.9 Variables initially included in the logistic regression models
Propensity Model
Model 2
Model 3
Model 4
Model 5
Model 6
Living alone Having a regular carer
Community health worker visits on a regular basis
No. doctors usually seen Self-rating of health No. times a GP was seen in last 2 months No different illnesses reported by patient Patient reports any hospitalisation in last year Comparative health Frequency of illness symptoms in past week score Severity of illness symptoms in past week score Symptoms in past week (high/low) Symptom severity in past week (high/low) Any ADRNo. different prescribers (median substitution) No. different dispensing pharmacies OARS-IADL score Square root (94-Age) Age
Gender Total No. of current medicines Square root total No. of current medicines Total No. of current solid medicines Total No. of current non-solid medicines Memory-related non-adherence score Any memory-related non-adherence Deliberate non-adherence score Any deliberate non-adherence Disorganised non-adherence score Any disorganised non-adherence Table 8.9 continued Propensity
Model Model
2Model
3Model
4Model
5Model
6
Any antianxiety medicines Any antidementia medicines Any antidepressant medicines Any antipsychotic medicines Any rheumatoid or Parkinson’s disease medicine
Any lipid lowering medicines Any oral hypoglycaemic medicines No. respiratory medicines Any beta blocker medicines No. medicines acting on angiotensin II Any other antihypertensive medicines No. other medicines used to treat heart failure
Days since last medication change Do you ever forget to take your medications
Are you careless about taking your medicines When you feel better do you stop taking your
Quality Medication Care Group, School of Medicine, University of Queensland199
medicines If you feel worse do you stop taking your medicines
Included initially but removed; Retained in the final model
8.2.3 NON-LINEAR MACHINE LEARNING MODELS
8.2.3.1 Interpreting output
Two types of output of the decision tree approaches to machine learning, decision
trees and a rule set technique, were generated. To read a decision tree, start with the
left most variable and it’s conditional statement. As extra branches are added to the
tree the conditional statements in all parent branches must be met. There are numbers
at the end of each leaf node, (n/m) or just (n). ‘n’ is the number of cases which fall
under that branch and ‘m’ is the number of cases incorrectly classified under that rule.
Fractions can exist because “(A non-integral number of cases can arise because, when
the value of an attribute in the tree is not known, See5 splits the case and sends a
fraction down each branch.)” (See 5 instructions)
Consider the leaves of the tree below in italics as an example. IADL <= 10: :...HVNUMRES > 2: : :...LIVING = 0: 0 (5/0.4) : : LIVING = 1: 1 (2.6/0.3) : HVNUMRES <= 2: : :...REGULAR CARER = 1: 1 (74.4/11) : REGULAR CARER = 0: : :...NUMBER OF NON-SOLID MEDICATIONS > 1: 0 (2) : NUMBER OF NON-SOLID MEDICATIONS <= 1: : :...LIVING = 1: 1 (9.9/1.4) : LIVING = 0: : :...TOTSOLID <= 6: 0 (3.9) : TOTSOLID > 6: 1 (2.7)
This would be read as:
If independent activities of daily living (IADL) is less than or equal to 10 and the
number of respiratory medications is greater than 2 (HVNUMRES) and the patient is
living alone then a dose aid should be assigned; 2.6 cases were classified under this
branch with 0.3 errors.
Or
If independent activities of daily living score (IADL) is less than or equal to 10 and the
number of respiratory medications is less than or equal to 2 and there is no regular
carer (CARER) and the number of non-solid medicines (NUM_NSME) is less than or
equal to 1 and the patient is living alone (LIVING) and has more than 6 solid
medicines (TOTSOLID) then a dose aid should be assigned: 2.7 cases were correctly
assigned to this branch of the tree.
For each leaf node on the tree a value of:
0 indicates that the person would be assigned to no dose aid group.
1 indicates that the person would be assigned to the dose aid group.
Another option that could lead to easier interpretation is rule-sets. For every tree
model that is learned an equivalent rule-set can be generated. These rule sets could be
easier to read and assess by health care professionals. Here is an example of a rule
set generated in the same way as the trees above. For each rule the following are
given:
Quality Medication Care Group, School of Medicine, University of Queensland200
Statistics (n, lift x) or (n/m, lift x) that summarize the performance of the rule.
Similarly to a leaf, n is the number of training cases covered by the rule and m, if it
appears, shows how many of them do not belong to the class predicted by the rule.
The rule's accuracy is estimated by the Laplace ratio (n-m+1)/(n+2). The lift x is the
result of dividing the rule's estimated accuracy by the relative frequency of the
predicted class in the training set.
One or more conditional statement(s) that must all be satisfied for the rule to be
applicable.
A class predicted by the rule.
A value between 0 and 1 that indicates the confidence with which this prediction is
made. (Note: If boosting is used, this confidence is measured using an artificial
weighting of the training cases and so does not reflect the accuracy of the rule.) (Copied from See5 help files)
8.2.3.2 Decision trees
The following figures show the decision trees generated from the same variables
included initially in the logistic regression models 2, 5 and 6.
Tree 1 used all the logistic regression model 2 variables and included 311 cases in the
training data. A tree with 20 leaf nodes was generated (Figure 8.1). Overall, 17.7% of
cases were miss-classified i.e. correct classification for 82.3% with correct classification
for 71.0% OP users and 91.3% pharmacy provided DAA users. The first node was
IADL score with a cut point of less than or equal to 10 versus greater than 10. People
who had a lower IADL score and did not stop taking their medicines if they felt worse
were assigned to the DAA group (87 cases met these criteria with 18% error). The
node where IADL is low, the person stops taking their medicines if they felt worse
(deliberate non-adherence) but have not been to hospital in the last 12 months were
also classified to the DAA group (10% error) makes less sense when taken in
conjunction with the logistic regression. This may be because the variable “when feel
worse, stop taking medicines” is potentially caused by DAAs rather than predicting
DAAs. Similarly, the variable “stop medicines if feel better” further down the tree may
be misleading as the DAA group responses do not represent their behaviour prior to
starting a DAA. Were patient behaviours prior to starting a DAA included in the tree,
the result may have been different.
activities of daily living <= 10: :...when feel worse stop taking medications = no: 1 (87/16) : when feel worse stop taking medications = yes: : :...hospital visit in last 12 months = no: 1 (3/0.3) : hospital visit in last 12 months = yes: 0 (10.4/3.5) activities of daily living > 10: :...when feel better stop taking medications = yes: 0 (3.1) when feel better stop taking medications = no:
:...ever forget to take meds = yes: 1 (46.1/13.6) ever forget to take meds = no: :...age > 86: 1 (16.3/2.8) age <= 86: :...regular community health workers visits = yes: :... when feel worse stop taking medications = yes: 0 (5.9/0.8) : when feel worse stop taking medications = no: : :...number of illnesses <= 4: 1 (19.7/5.3) : number of illnesses > 4: 0 (4.4/0.2)
Quality Medication Care Group, School of Medicine, University of Queensland201
regular community health workers visits = no: :...antirheumatoid or parkinsons meds = yes: :...number of angiotensin II meds <= 0: 1 (5.7/0.4) : number of angiotensin II meds > 0: 0 (3.4/0.3) antirheumatoid or parkinsons meds = no: :...when feel worse stop taking medications = yes: 0 (11.7/0.6) when feel worse stop taking medications = no: :...number of regular doctors visited > 4: 0 (7.3/0.1) number of regular doctors visited <= 4: :...memory related non-adherence > 0: 1 (9.5/3.8) memory related non-adherence <= 0: :...antidepressant meds = no: :...age <= 71: 1 (5.9/1.3) : age > 71: : :...number of illnesses <= 0: 1 (4.1/1.1) : number of illnesses > 0: 0 (53.5/8) antidepressant meds = yes: :...activities of daily living <= 12: 1 (3.8/0.3) activities of daily living > 12: :...number of regular doctors visited <= 1: 1 (2.9/0.3) number of regular doctors visited > 1: 0 (7.2/0.6)
Figure 8.1 Decision tree generated using the same variables as Regression Model 2
Tree 2 (Figure 8.2) excludes the potentially confounding ‘causal’ Meichenbaum
variables related to deliberate non-adherence. It included 311 cases in the training
data. A tree with 28 leaf nodes was generated. Overall, 83.9% of cases were correctly
classified with correct classification for 77.5% OP users and 89.0% DAA users. Again,
IADL was the first node. For people with a low IADL, the next node was number of
respiratory medicines where people with more than 2 were classified as OP users
(error 35%). This makes teleological sense as respiratory medicines are more likely to
be dose forms (aerosols, nebulisers) that cannot be packed into a DAA, so that in this
case the difficulties associated with using a DAA when lots of other medicines are not
packed may well override any advantage of DAAs. Note that the number of non-solid
medications appears further down the tree with a similar effect.
activities of daily living <= 10: :...number of respiratory drugs > 2: 0 (7.5/2.6) : number of respiratory drugs <= 2: : :...regular carer = yes: 1 (74.4/11) : regular carer = no: : :...number of non-solid medications > 1: 0 (2) : number of non-solid medications <= 1: : :...living alone = yes: 1 (9.9/1.4) : living alone = no: : :...total number of solid medications <= 6: 0 (3.9) : total number of solid medications > 6: 1 (2.7) activities of daily living > 10: :...ever forget to take meds = yes: 1 (48.2/15.6) ever forget to take meds = no:
:...number of illnesses <= 0: 1 (8.3/1.2) number of illnesses > 0: :...regular carer = no: :...number of non-solid medications > 2: 0 (15.3/0.5)
Quality Medication Care Group, School of Medicine, University of Queensland202
: number of non-solid medications <= 2: : :...number other heart failure meds > 0: : :...regular community health workers visits=yes: 1 (4.6/0.4) : : regular community health workers visits = no: : : :...oral hypoglycaemic meds = no: 0 (11/3.2) : : oral hypoglycaemic meds = yes: 1 (3.3/0.2) : number other heart failure meds <= 0: : :...other antihypertensive meds = yes: 0 (22/2.1) : other antihypertensive meds = no: : :...hospital visit in last 12 months = no: 0 (29.8/8.1) : hospital visit in last 12 months = yes: : :...age <= 79: 0 (5.1/1.2) : age > 79: 1 (3.9/0.2) regular carer = yes: :...number of illnesses <= 1: 1 (6.1/0.2) number of illnesses > 1: :...oral hypoglycaemic meds = yes: 1 (5.1/1) oral hypoglycaemic meds = no: :...number of angiotensin II meds > 1: 0 (3/0.2) number of angiotensin II meds <= 1: :...community health workers visit regularly = yes: :...number of illnesses <= 3: 1 (8.3/0.8) : number of illnesses > 3: 0 (3.2) community health workers visit regularly = no: :...deliberate non-adherence > 2: 1 (2) deliberate non-adherence <= 2: :...number other heart failure meds > 1: 1 (4/0.8) number other heart failure meds <= 1: :...number of regular doctors visited > 2: 0 (9/0.5) number of regular doctors visited <= 2: :...memory related non-adherence > 0: 1 (2.1/0.1) memory related non-adherence <= 0: :... activities of daily living > 12: 0 (7.1) activities of daily living <= 12: :...living alone = no: 0 (4.4/0.7) living alone = yes: 1 (5/1)
Figure 8.2 Decision tree generated using the same variables as Regression Model 5
Tree 3 (Figure 8.3) started with the same variables as tree2 except that regular carer
and community health visits were excluded – these two variables overshadowed the
effect of IADL in the logistic regression models and use of these services may be
driven, in part, by service availability rather than direct need for a carer or community
health visitor. It included 311 cases in the training data. A tree with 18 leaf nodes was
generated. Overall, 80.1% of cases were correctly classified with correct classification
for 70.3% OP users and 87.9% DAA users. Again, IADL was the first node but all other
nodes after the number of respiratory medicines were omitted (due to the exclusion of
carer and community health worker), there were 18.3 errors in the low number of
respiratory medicines arm compared to 12.4 in tree 2.
activities of daily living <= 10: :...number of respiratory drugs <= 2: 1 (92.9/18.3) : number of respiratory drugs > 2: 0 (7.5/2.6) activities of daily living > 10: :...ever forget to take meds = yes: 1 (48.2/15.6) ever forget to take meds = no:
:...number of illnesses <= 0: 1 (8.3/1.2)
Quality Medication Care Group, School of Medicine, University of Queensland203
number of illnesses > 0: :...number of respiratory drugs > 1: 0 (12.9/1.1) number of respiratory drugs <= 1: :...oral hypoglycaemic meds = 1: :...other antihypertensive meds = no: 1 (14.7/4.8) : other antihypertensive meds = yes: 0 (2.3/0.2) oral hypoglycaemic meds = no: :...hospital visit in last 12 months = no: :...past week self-rated symptom severity level = 1: 0 (23.9/3.8) : past week self-rated symptom severity level = 2: 0 (9.5/3) : past week self-rated symptom severity level = 3: 0 (9.8/1.6) : past week self-rated symptom severity level = 4: : :...age <= 80: 0 (9.6/1.3) : : age > 80: 1 (6/0.4) : past week self-rated symptom severity level = 5: : :...number of angiotensin II meds <= 0: 1 (4.2/1) : number of angiotensin II meds > 0: 0 (7.7/1.9) hospital visit in last 12 months = yes: :...number of regular doctors visited > 2: 0 (19.6/3.6) number of regular doctors visited <= 2: :...number of non-solid medications > 1: 0 (6.7/1.8) number of non-solid medications <= 1: :...number of illnesses <= 4: 1 (21.3/3.4) number of illnesses > 4: 0 (5.9/1.5)
Figure 8.3 Decision tree generated using the same variables as Regression Model 6
Also of interest are the medication classes included in the nodes – these nodes are
printed in grey in the figures. In tree 1 (Figure 8.1), three different drug classes were
included in the model whereas tree 2 included 6 classes of drugs including oral
hypoglycaemics, not included in any logistic model. Here a person with type II diabetes
with a higher IADL score, no a regular carer or community health visits, low numbers of
non-solid medicines, at least one ‘other heart failure medicine’ were classified as DAA
users (error 6%).
The predictive accuracy of the trees was examined through a process of cross
validation and boosting with results shown in Table 8.10. Overall, these figures
suggest that including all of the relevant variables (Model 2) produces the most
accurate prediction when cross-validated but overall there is little variation in the cross-
validation figures. Note also, the potentially misleading results associated with the
deliberate non-adherence variables that had been included in the tree from Model 2.
In terms of utility, another point that is important to consider is the size of the tree. A
smaller tree is likely to be easier to interpret. Using a reduced set of variables (Model 6)
the tree that is generated has only 18 leaf nodes whereas including all variables
produces a slightly larger tree with 20 leaf nodes, and tree 2 had 28 nodes.
Table 8.10 Predictive Accuracy of Decision Tree Models
Initial Tree Cross-validated 10 Fold Boosted 50 + Cross-validated 10 x
Model2
Size Errors20 55(17.7%)
(a) (b)<-classified as ---- ---- 98 40 (a): class 0 15 158 (b): class 1 Sensitivity: 87%
Mean 27.9 35.1% SE 2.0 2.1%
(a) (b)<-classified as ---- ---- 85 53 (a): class 0 56 117 (b): class 1 Sensitivity: 60%
Mean Errors 29.9% SE 1.9%
(a) (b)<-classified as ---- ---- 87 51 (a): class 0 42 131 (b): class 1 Sensitivity: 67%
Quality Medication Care Group, School of Medicine, University of Queensland204
(a) (b)<-classified as ---- ---- 107 31 (a): class 0 19 154 (b): class 1 Sensitivity: 85% False Alarm: 17%
Mean Errors 21.5 36.9% SE 1.8 3.3%
(a) (b)<-classified as ---- ---- 75 63 (a): class 0 52 121 (b): class 1 Sensitivity: 59% False Alarm: 34%
Mean Errors 32.4% SE 2.3%
(a) (b)<-classified as ---- ---- 80 58 (a): class 0 43 130 (b): class 1 Sensitivity: 65% False Alarm: 30%
Model6
Size Errors18 62(19.9%)
(a) (b)<-classified as ---- ---- 97 41 (a): class 0 21 152 (b): class 1 Sensitivity: 82% False Alarm: 21%
Mean Errors 21.6 38.0% SE 1.4 2.6%
(a) (b)<-classified as ---- ---- 72 66 (a): class 0 52 121 (b): class 1 Sensitivity: 58% False Alarm: 36%
Mean Errors 31.5% SE 2.4%
(a) (b) <-classified as ---- ---- 83 55 (a): class 0 43 130 (b): class 1 Sensitivity: 65% False Alarm: 30%
8.2.3.3 Rule sets
As an example, a rule set was generated using the same variables (and 311 training
cases) as in the logistic regression model 5 (and tree 2). The rules are presented
below and sorted by utility (5 bands). The rule set classified 79.1% of cases correctly
(71.7% OP and 85.0% DAA) – this is less than tree 2.
Rule 1: (155/48, lift 1.2) CARER = 1 If have regular carer -> class 1 [0.688] then assign dose aid.
Rule 2: (195/89, lift 1.2) IADL > 10 If activities of daily living are more than 10 -> class 0 [0.543] then no dose aid.
Rule 3: (76/22, lift 1.3) BMQ1A1 = 1 If ever forget to take medicine -> class 1 [0.705] then assign dose aid.
Rule 4: (71/21, lift 1.3) NUM_NSME <= 2 If number of non-solid medicines is <=2 HVNUM_OT > 0 and number of other heart failure medicines is >0 -> class 1 [0.699] then assign dose aid.
Rule 5: (72/23, lift 1.5) LIVING = 0 If don’t live alone CARER = 0 and don’t have a carer -> class 0 [0.676] then no dose aid.
Rule 6: (10, lift 2.1) CARER = 1 If have a carer COM_HLT = 0 and community health workers don’t visit regularly NUM_ILL > 0 and the number of illnesses is greater than 0
Quality Medication Care Group, School of Medicine, University of Queensland205
IADL > 12 and activities of daily living are more than 12 F1_MEM <= 0 and there is no memory related non-adherence F2_DELIB <= 2 and the rating of deliberate non-adherence is <=2 HVNUM_OT <= 1 and number of other heart failure medicines is <=1 -> class 0 [0.917] then no dose aid.
Rule 7: (27, lift 2.2) COM_HLT = 0 If community health workers don’t visit regularly NUM_GP > 2 and the number of regular doctor is greater than 2 NUM_ILL > 0 and the number of illnesses is greater than 0 IADL > 10 and activities of daily living are more than 10 HVORAL_H = 0 and oral hypoglycaemic medicines are used HVNUM_OT <= 1 and number of other heart failure medicines is <=1 BMQ1A1 = 0 and don’t forget to take your medicines -> class 0 [0.966] then no dose aid.
Rule 8: (28/6, lift 1.4) COM_HLT = 1 If community health workers do visit regularly HVNUM_OT > 0 and number of other heart failure medicines is >0 -> class 1 [0.767] then assign dose aid.
Rule 9: (14/4, lift 1.5) HVNUMRES > 2 If number of respiratory medicines is greater than 2 -> class 0 [0.688] then no dose aid.
Rule 10: (14/1, lift 1.6) NUM_ILL <= 0 If the number of illnesses is less than or equal to 0 -> class 1 [0.875] then assign dose aid.
Rule 11: (36/10, lift 1.6) CARER = 0 If don’t have regular carer NUM_NSME > 1 and the number of non-solid medicines is >1 -> class 0 [0.711] then no dose aid.
Rule 12: (33/2, lift 1.6) LIVING = 1 If you live alone IADL <= 10 and activities of daily living are <= 10 NUM_NSME <= 1 and the number of non-solid medicines at home<=1 -> class 1 [0.914] then assign dose aid.
Rule 13: (11/2, lift 1.7) IADL > 10 activities of daily living are more than 10 HV_NUM_A > 1 number of medicines acting on angiotensin II is >1 -> class 0 [0.769] then no dose aid.
Default class: 1 (dose aid)
The following rule utility summary shows that including rules 1 to 10 correctly classifies
77.5% of cases with the last 3 rules only improving the correct classification by 1.6%
shows the cost difference between DAAs and OPs when a bare bones - just packing
and checking DAA service ($12.57) is provided or a full DAA service including
prescription and account management, and delivery ($20.05) is provided. Table 9.2
shows the incremental cost of a DAA (the cost per DAA customer minus the cost per
OP customer) ranges between $12.57 and $29.27 depending on variations in packing
time, the type of pack used, who packs and the level of additional services provided.
Table 9.2 Description of the sensitivity analysis variations for the cost of DAA and
OP medication provision and results
Area of sensitivity Description of Variation Total Cost Cost/ customer
CostDifference
1. Pharmacy salaries
DAA base case Award wages plus 25% on-costs $1,070.50 $35.75
OP base case Award wages plus 25% on-costs $543.88 $18.13
$17.62
DAA salary on-costs Award wages plus 40% on-costs $1,169.40 $38.98
OP salary on-costs Award wages plus 40% on-costs $597.04 $19.90
$19.08
2. Packing Time
DAA base case ~11 min/customer $1,070.50 $35.75 $17.62
Single packing time excludes interruptions
Mean packing time of 9.66 minutes $1,059.72 $35.32 $17.20
packing time 75th percentile – 11.26 minutes $1,089.63 $36.32 $18.20
packing time 25th percentile – 4.05 minutes $954.87 $31.83 $13.71
Single packing time include interruptions
Mean packing time of 12.62 $1,115.04 $37.17 $19.05
packing time 75th percentile – 16.23 minutes $1,182.51 $39.42 $21.30
packing time 25th percentile – 4.94 minutes $971.69 $32.39 $14.27
3. Checking Time
Single checking time exclude interruptions
Mean checking time - 3.44 minutes $1,049.20 $34.97 $16.85
checking time 75th percentile – 3.95 minutes $1,060.25 $35.34 $17.22
checking time 25th percentile – 0.81 minutes $992.20 $33.07 $14.95
Single checking time include interruptions
Mean checking time - 4.67 minutes $1,075.85 $35.86 $17.74
checking time 75th percentile – 5.14 minutes $1,086.04 $36.20 $18.08
checking time 25th percentile – 1.11minutes $998.70 $33.29 $15.17
4. Who packs
Pharmacist only 100% pharmacist packing $1,151.93 $38.40 $20.28
Dispensing technician only
100% dispensary technician packing
$1,030.17 $34.34 $16.22
5 Type of Pack
Blisters only 100% of blister packs used $1,051.50 $35.05 $16.93
Dosett type only 100% of Dosett-type boxes used $1,421.66 $47.39 $29.27
Automated only 100% automated packing used $1,068.00 $35.60 $17.48
Quality Medication Care Group, School of Medicine, University of Queensland211
Table 9.2 continued
Area of sensitivity Description of Variation Total Cost Cost/ customer
CostDifference
6. Level of services
DAA pack only - no other services
DAA service included dispensing meds, packing and checking and additional cost
$921.09 $30.70 $12.57
DAA pack as part of a complete service
100% of customers receive all services i.e. as above plus prescription management, delivery, counselling and account management
$1,145.28 $38.18 $20.05
9.4 POTENTIAL COST SAVINGS FROM PREVENTING ADRS
The rate of adverse drug reactions (ADRs) was lower in both Phase 2 and Phase 3 for
the DAA sample compared to the OP group. It is hypothesised that using DAAs will
decrease costs associated with ADRs and that these savings will off-set the cost of
supplying DAAs in the community setting. The purpose of this section is to model the
potential cost-savings due to reduced ADRs reported for DAA patients compared to OP
patients, from the perspective of the health service funding body (namely the Australian
Government). These costs savings will then be considered in terms of the cost-
effectiveness of DAAs in avoiding an ADR and the net social cost of DAAs compared
with OPs.
9.4.1 METHODS
The rate of adverse drug reactions (ADRs) was lower in both Phase 2 and Phase 3 for
the DAA sample compared to the OP group. In Phase 2, 48% and 33% of OP and DAA
patients, respectively, suffered from an ADR (p=0.007). In Phase 3, the corresponding
proportions were 32% and 22% (p=0.147). This apparent improvement is due in part to
the greater proportion of ADR experienced by people who had exited the study by
Phase 3 - 67% of OPs who exited the study to enter an RCF reported an ADR in Phase
2 and 50% of those who died had reported an ADR in Phase 2 compared with 17% and
31% respectively for DAA users.
The decision analytic model developed in Phase 2 to simulate the potential costs and
outcomes of using the intervention (DAA) versus no intervention (OP) over a one year
period was adjusted to reflect the consequences of ADRs as reported by community
patients in the Phase 3 follow-up (see Phase 2 Final Report for methodology details
(Ientile et al. 2004)). The model was designed using a “branching tree format”, with the
beginning decision (using DAA or OP) branching into further pathways or nodes
(Drummond et al. 1997). End points determined for this model were the percentage of
GP visits, medication changes, hospitalisations, community nurse visits, and RCF
admission due to adverse drug reactions (ADRs), with or without the use of DAAs.
These end-points and pathways differ from those chosen in Phase 2 because of the
availability of data collected in the follow-up that was not available in the literature on
the consequences of ADRs. Medication changes were included in this model as they
were a frequently reported consequence of an ADR but were not included in the Phase
2 model as there was not data available on medication changes in the literature. The
Phase 2 model also included the percentage of patients that died as a result of an ADR
(mean frequency of 7%, range 0.2% to 18.2%) (Classen et al. 1997; Dartnell et al.
1996; Naranjo et al. 1981; White 1999). However, as this revised model is based on
Quality Medication Care Group, School of Medicine, University of Queensland212
self-report, patients who died from an ADR were not available to complete the Phase 3
follow-up. Consequently this branch has been excluded from the revised decision
analytic model.
The model was applied to a sample population of 30 community customers (to reflect
the costs of one pharmacy providing DAAs), with the probability of each event being
determined using actual data obtained from the current study (i.e. rate of ADRs in DAA
and OP customers). The rates of adverse events and probability of service use as a
result of an ADR were obtained from the community patient follow-up interview.
Contributions of each end point to the total cost were then calculated by multiplying the
final probability in population terms by the average cost of each end point of the
decision-analytic model. The value of the resources utilised in the provision of
healthcare as a result of an ADR are presented in Table 9.3.
Table 9.3 Resources utilised in the provision of healthcare resulting from an ADR
Cost item Value of cost item Description Cost Reference
Hospital $774.16/day 3.29 days (avg. visit length) $2,547 1 GP visit $57/consultation 2 long consults (MBS items 36) $115 2 Nurse visit $56/hour Weekly visits @ 30 min/visit for
6 months $728 3
RCF $69/day Mid year admission (180 days) $12,417 4 Medication change $19.89 per
medicine Mean dispensed price for medication use by this sample
$19.89 5
1. NHCDC:Cost Report - Round 4 (1999-2000); 2. Medicare Benefits Schedule - updated July 2004; 3. Rates of visit length = community nurse survey4. Commonwealth Department of Health and Aging 2002 5. PBS listing and Phase 2 data set
9.4.2 RESULTS
In the sample of community customers followed up in Phase 3, ADRs were reported by
17 of 78 DAA pts (21.8%) compared with 28 of 88 OP pts (31.8%) ( 2 =102 p=0.147).
For the purpose of this model, it was assumed ADRs were experienced only once
within the year. The frequency of outcomes was used to populate the decision analytic
model presented in Figure 9.1. This diagram shows the likelihood of an event and the
number of community customers experiencing a particular event for both the DAA and
OP branches based on 30 community customers in each arm. The events included in
this model were GP visits, medication changes, hospitalisation, community nursing
services and RCF admission. The Phase 2 model was altered to reflect the reported
health service use resulting from ADRs such that all patients who received treatment
utilised GP services and some experienced medication changes, hospitalisation, RCF
respite and community nursing services (in addition). Unlike the Phase 2 model based
on the literature, different consequences of ADRs were reported by DAA and OP
patients. Further, the treatment path in Phase 3 stemmed from GP visits rather than in
Phase 2 where literature estimates for treatment included either GP, hospital or RCF
visits.
The difference in the number of customers experiencing outcomes between the DAA
and OP branches stems from the 10% difference in the rates of ADRs but was
mediated by the increased likelihood of DAA patients seeking treatment for an ADR,
compared with OP patients. In addition, DAA patients were less likely to report
additional treatment (other than GP services) for an ADR compared with OP patients.
Quality Medication Care Group, School of Medicine, University of Queensland213
Table 9.4 compares the health and cost implications for both DAA and OP customers
and the difference. The total cost of treating ADRs for OP customers minus the cost of
treating ADRs for DAA customers ($5,040) represents the potential cost savings of
DAA for the government, based on 30 community patients over 12 months.
DAA
ADR
No treatment
GP visit
32
68
56
44
30
30
9.6
20.4
5.4
4.2
5.4
Community customers on
multiple medications
OP
No ADR
Treatment
Med changes
Hospital7
5.0
0.4
RCF admission7 0.4
Comm. Nurse0 0
ADR
No treatment
22
78
82
186.5
23.5
5.4
1.2
No ADR
Treatment
100
93
GP visit
5.4
Med changes
Hospital0
3.8
0
RCF admission0 0
Comm. Nurse7 0.4
71
100
Figure 9.1 Phase 3 decision analytic model comparing the frequency of outcomes as a
result of an ADR based on patient report - DAA versus OP customers
Table 9.4 Comparison between 30 DAA and OP customers of the costs to the
healthcare system incurred as a result of ADRs
Outcome DAA cost OP cost Difference
ADR + conseq + GP $618.11 $608.40 $9.71
ADR + conseq + GP+ Med change $76.53 $98.43 -$21.90
ADR + conseq + GP+ Hosp $0 $905.16 -$905.16
ADR + conseq + GP+ RCF $0 $4,412.71 -$4,412.71
ADR + conseq + GP+ nurse $290.31 $0 $290.31
Total cost $984.96 $6,024.70 -$5,039.74Key: conseq=consequence, hosp=hospitalisation, GP=GP visit, med change=medication changed,
nurse=Community nursing care, RCF=admission to a residential care facility
9.4.3 SENSITIVITY ANALYSIS OF HEALTHCARE COST SAVINGS
Sensitivity analysis has been used to explore the uncertainty in the proportion of
customers experiencing ADRs, hospitalisation, and RCF use. Sensitivity analysis 1
explores variation in the proportion of patients experiencing an ADR. There was a
greater difference in the rates of ADRs reported by community patients in Phase 2
compared with the sample followed-up in Phase 3. In Phase 2, 33% of community
patients using DAAs reported experiencing an ADR compared with 48% of OP
community patients ( 2=7.39, p=0.007). Sensitivity analysis was preformed using this
data to evaluate the cost implications of a higher rate of ADRs.
Sensitivity analysis 2 explores variations in the cost of hospitalisation based on data
extracted from HIC on treatment costs in hospital in addition to the accommodation
costs in hospital utilised in the base case. Sensitivity analysis 3 explores variations in
the cost of RCF care. The costs of RCF care were varied by multiplying the daily rate
by 30 days to reflect respite care only compared with the base model where permanent
Quality Medication Care Group, School of Medicine, University of Queensland214
admission was assumed to occur mid-year (180 days). Finally, sensitivity analysis 4
examines the uncertainty around the probability of different consequences between
DAA and OP customers. The difference in outcomes was not significantly different due
to the small sample size for Phase 3 and the lower rates or reported ADRs for both
groups. In sensitivity analysis 4, the consequences of ADRs are assumed to be the
same for both groups. The results of the analysis are presented in Table 9.5.
Sensitivity analysis 1 indicates that savings in healthcare costs resulting from a
reduction in ADRs at the rate observed in Phase 2 would be an additional $2,184.
Similarly increases in the cost of hospitalisation when treatments are included in
addition to accommodation, increases the potential cost savings from DAAs by
$824/year per 30 patients. Sensitivity analysis 3 indicates that a reduced length of stay
in an RCF (respite rather than permanent admission) reduces the total cost savings to
$1,362. Sensitivity analysis 4 indicates that if there is no difference in the likelihood of
hospitalisation and residential care admissions, the savings due to reduced ADRs in
the DAA group translates to just $370/year per 30 customers.
The cost difference varied between $370 and $7,583 (base model $5,039) when ADR
rates and consequences, hospitalisation and RCF cost were varied. The model was
most sensitive to change in the rates of RCF admission and hospitalisation (Table 9.5).
Table 9.5 Sensitivity analysis of healthcare cost savings
Area of sensitivity Description of Variation Total Cost Cost Difference
1. Frequency of ADRs
DAA base case 22% experience ADRs $984.96 OP base case 32% experience ADRs $6,024.70
-$5,039.74
33% experience ADRs $1,472.92 Rates of ADRs in Phase 2 data 48% experience ADRs $9,055.99
-$7,583.07
2. Hospitalisation
DAA base case $2,547 per admission $984.96 OP base case $2,547 per admission $6,024.70
-$5,039.74
DAA hosp costs $4,867 per admission $984.96 OP hosp costs $4,867 per admission $6,849.19
-$5,864.23
3. RCF admission
DAA base case Mid-year admission (180 days) $984.96 OP base case Mid-year admission (180 days) $6,024.70
-$5,039.74
DAA days in RCF Respite care (30 days) $984.96 OP days in RCF Respite care (30 days) $2,347.44
-$1,362.48
4. Same consequence
DAA base case Med =71%, Hosp& RCF=0%, CN=7% $984.96 OP base case Med =93%, Hosp& RCF=7%, CN=0% $6,024.70
Sensitivity analysis was performed to explore the impact of variation in WTP (lower and
upper 95%CI), variations in cost of healthcare due to ADRs as examined in Section
9.4.3 and the cost of providing DAAs as examined in Section 9.3.2. The cost benefit
ratio (Benefits – Costs) varied between -$5,746 and -$17,414 (base model -$13,697)
when the benefits to the patient and the healthcare system and the costs to the
Quality Medication Care Group, School of Medicine, University of Queensland217
pharmacists were varied. The model was most sensitive to variations in the cost of
DAAs, as seen in Table 9.7.
Table 9.8 Sensitivity analysis of cost-benefit ratios
Area of sensitivity Description of Variation Value Benefit-Cost
1. WTP
DAA base case $5.87 per week per customer $9,157.20 -$13,291 WTP (lower 95%CI) $4.57 per week per customer $7,129.20 -$15,319 WTP (upper 95%CI) $7.16 per week per customer $11,169.60 -$11,279
2. Cost of healthcare
OP-DAA base case ADR: 22% DAA & 32% OP $5,039.74 -$13,697OP-DAA Phase 2 rates ADR: 33% DAA & 48% OP $7,583.07 -$11,153OP-DAA days in RCF Respite care (30 days) $1,434.77 -$17,302OP-DAA hospital cost $4,867 per admission $5,936.52 -$12,8003. Cost of DAAs
OP-DAA base case See 9.3.1 $27,487.94 -$13,697OP-DAA packing time 25th percentile – 4.05 minutes $21,387.60 -$7,524DAA pack only DAA service includes dispensing
meds, packing and checking only $19,609.20 -$5,746
DAA pack – complete service
100% of customers receive all serv-ices i.e. as above plus prescription management, delivery, counselling and account management
$31,278.00 -$17,414
9.5 COST SAVINGS AND ECONOMIC EVALUATION
ASSOCIATED WITH PATTERNS OF HEALTH SERVICE
UTILISATION
The benefits of DAAs are thought to be more far-reaching than reducing ADRs. It was
hypothesised that savings from reduced health service use for DAA patients may offset
the cost of supplying DAAs in the community. The purpose of this section is to model
the potential cost-savings due to differences in service use of DAAs compared to OP,
from a wider health care perspective.
9.5.1 METHODS
A costing model was constructed based on patient data on resource use and outcomes
collected in Phases 2 and 3, and health service use, Medical Benefits Schedule (MBS)
and Pharmaceutical Benefits Scheme (PBS) data extracted from the Health Insurance
Commission (HIC) (now Medicare Australia). Additional cost data (not available
through HIC) were also included in the cost model to provide a more complete picture
of health service use. These additional costs included residential care and community
care. The following resource use was accounted for in this model: PBS drug costs, GP
service items not provided in hospital, MBS Monitoring and Pathology costs that might
be incurred in monitoring medication management (not provided in hospital), other
MBS services (not provided in hospital), MBS hospital costs (cost of MBS items
received in hospital not including accommodation costs or other patient costs),
community care (including community nursing, home care and meals-on-wheels), and
residential care. These resources were assumed to be utilised over a 12-month period
and cost estimates reflect costs to the health care system, i.e., no patient co-payments
are included.
Quality Medication Care Group, School of Medicine, University of Queensland218
9.5.1.1 HIC data
Health service use data (MBS and PBS service) was obtained from the HIC for
consenting community patients from Phase 2, to inform the economic model of DAA
provision to community patients by pharmacies. The details of data retrieval and data
analysis can be found in section 7.1.
Due to the differences observed between the DAA group and the original pack groups
both at baseline (Ientile et al. 2004) and at follow-up (see Community patient results
section 6.2), propensity score methods have been utilized to adjust for sampling bias.
This method involves comparing service costs and outcome probabilities from a
matched sample of DAA and OP patients. Matching involves pairing together treatment
and comparison units that are similar in terms of their observable characteristics. When
the relevant differences between any two units are captured in the observable (pre-
treatment) covariates (i.e., outcomes are independent of assignment to treatment,
conditional on pre-treatment covariates), matching methods can yield an unbiased
estimate of the treatment impact (Dehejia et al. 1998).
Propensity scores were calculated from an algorithm using the Beta weights derived
from a logistic regression model to predict group membership (DAA or OP). The logistic
regression model (Chi-square=98, df 12, p<0.001, n=249) included the variables
collected in Phase 2 (see Table 9.9) and correctly predicted 72.3% of cases overall
(probability cut point 0.5) as belonging to DAA (77.4%) or OP (66.1%) groups (Cox &
Snell R Square=0.326 and Nagelkerke R Square=0.436). The variables, living alone
and self-rated health status were included as they improved the predictive ability of the
model.
Table 9.9 Logistic regression model of using a pharmacy supplied DAA (versus OP)
Variables in the logistic regression model
B Std. Error
Wald statistic
df Sig. OR OR 95% CI
Having a regular carer 1.141 0.344 10.990 1 <0.001 3.131 1.594 - 6.148Having regular community health worker visits
0.815 0.374 4.756 1 0.029 2.260 1.086 - 4.704
Number of GPs/doctors regularly visited
-0.279 0.131 4.562 1 0.033 0.756 0.585 - 0.977
Any hospital in last 12mths reported by patient
1.256 0.369 11.610 1 <0.001 3.512 1.705 - 7.235
Experienced adverse reactions - any symptoms/problems caused by medicines
-1.069 0.335 10.183 1 0.001 0.343 0.178 - 0.662
Square root of (94 – age (in years)) -0.459 0.163 7.938 1 0.005 0.632 0.459 - 0.870Square root of the total number of current medicines in home
-1.971 0.707 7.782 1 0.005 0.139 0.035 - 0.556
Total number of current solid medicines
0.392 0.134 8.545 1 0.004 1.481 1.138 - 1.926
Number of times visited GP in past 2 months
-0.235 0.085 7.731 1 0.005 0.790 0.670 - 0.933
How many different doctor prescribing medicines per patient (with median substitution (1) of 41 missing values)
-0.637 0.242 6.906 1 0.009 0.529 0.329 - 0.851
Living alone 0.568 0.343 2.731 1 0.098 1.764 0.900 - 3.458Self-rated health status (poor-excellent)
-0.424 0.225 3.535 1 0.060 0.654 0.421 - 1.018
Constant 6.320 1.660 14.497 1 <0.001 555.606
Quality Medication Care Group, School of Medicine, University of Queensland219
The cases included in the model were divided into 7 equal groups based on the
probabilities derived from the model. Within each of the septiles, the DAA and OP
groups were compared to test equality of the covariates (from variables collected in
Phase 2) and probability score per the model. In strata 5 only was there any significant
difference – the number of illnesses reported by the patient was (OP 4.8, DAA 2.9,
p=0.048). The stratum was considered to be matched however, as there was no
significant relationship between the number of illnesses and outcome (RCF admission
or death) or HIC costs.
The number of cases in each of the Pharmacy DAA and OP groups were examined for
each stratum (Figure 9.2). In the lowest and highest strata, there were insufficient
cases with overlapping probability for comparison.
1 2 3 4 5 6 7Probability of DAA strata
0
10
20
30
40
No
. c
ase
s i
n e
ac
h s
tra
ta
Original pack
Pharmacy DAA
Figure 9.2 The number of cases in each stratum of the propensity score based on the
probability of using a pharmacy supplied DAA after adjusting for covariates
The pharmacy-supplied DAA and OP groups were then compared on HIC costs and
outcome at 1 year follow-up. The mean of the natural logarithm of each cost type
(where a patient incurred a cost) was calculated for the DAA and OP groups within the
strata 2 to 6. The mean was then converted back into dollars to make the result more
interpretable for use in the cost models. The lower and upper 95% confidence intervals
of the mean were also converted into dollars (giving a non-symmetrical confidence
interval when expressed at the non-transformed value). There was no significant
difference in the Ln costs for any cost in any stratum. The number of patients admitted
to RCFs by 1 year follow-up was significantly higher (Fisher’s exact test p=0.045) for
stratum 4 but there were no other significant differences in the outcome of RCF
admission or death by 1 year follow-up.
A sub-sample of patients with a propensity score within the strata 2 to 6 were selected
from the sample of patients with HIC data. The probability of service use and mean
costs were calculated for both the DAA and OP groups based on this sample matched
on propensity scores (see Table 9.10).
Differential costing for the DAA and OP arms have been included to reflect the health
service use data that was collected for this sample in this study. The methods used to
Quality Medication Care Group, School of Medicine, University of Queensland220
derive these costs and the results of service use for the full sample are shown in
section 7.1. The mean cost per patient per year was calculated after excluding
veterans for whom cost data was not available.
Table 9.10 Resources use and cost data assumptions for matched sample
(strata 2-6): Mean and 95% CI
Service/ Resource
Description Group $Mean Cost
Percentile5th
Percentile95th
OP $2,089.54 $1,796.73 $2,430.07PBS Drug Costs
Mean cost for 12 month supply of medicines listed on PBS DAA $2,423.46 $2,096.54 $2,801.37
OP $504.51 $423.80 $600.58GP services (MBS)
Mean cost for 12 months of GP services. DAA $510.10 $431.51 $603.00
OP $145.17 $106.38 $198.10Pathology(MBS)
Mean cost for 12 months of pathology (excluding those who had no pathology).
DAA $189.93 $148.32 $243.22
OP $978.79 $720.95 $1,328.85Other MBS costs
Mean cost for 12 months of other MBS services (not pathology, GP services, or MBS services delivered in hospital – excluding those who had no additional MBS services)
DAA $577.16 $237.62 $1,401.86
OP $1,117.19 $737.85 $1,691.54Hospital Costs
Mean MBS item hospital costs for 12 months (excludes those who had no hospital services)
DAA $844.48 $501.20 $1,422.87
9.5.1.2 Patient reported service use and outcomes
Additional cost data (not available through HIC) were also included in the cost model to
provide a more complete picture of health service use. These additional costs included
residential care and community care.
Residential care
In assigning a cost for residential care admission, admission was assumed to occur at
the beginning of the 12 month period (i.e. service use for a full 365 days). The costs of
residential care in nursing homes and hostels was estimated as the average basic
subsidy cost per bed-day (Resident Classification Scales 1 to 7) in the 2004 financial
year of $68.98 multiplied by 365 days. When compared with the literature, the cost
estimates for residential care appear to be highly conservative even after excluding
patient contributions. Reported costs for residential care have ranged from $64 to $180
per bed day (not adjusted for inflation); between 72% and 87% of the cost is born by
the Commonwealth government (Liu 1999; Smith et al. 1993; Zinn 2002). The actual
amount that the government pays for each aged care facility place depends on the
residents care needs, extra service requirements, the facilities location (urban/rural),
the facilities policy regarding accommodation bonds and charges and the resident’s
income and assets. For simplicity, however, the average daily care fee has been used
as the basis for residential care costs in this model, i.e., $68.98 x 365 = $25,178 (with
rounding).
Community care
Community care resource use is an aggregate measure of a number of different
community services including community nursing, home care and meals-on-wheels. In
the current study, patients were asked about their use of community care services in
Quality Medication Care Group, School of Medicine, University of Queensland221
the previous four weeks. The cost data was extrapolated to reflect a 12 month period
(see Table 9.11).
Table 9.11 Valuation of different types of community support services
Cost item Value attached to cost item
Description Cost/year Reference
Cleaning $40.00 Once/week for 12 months $2,080 Patient reported use and cost*
Meals $30.00 Meals provided 5/week for 12 months@$30/week
$1,560 Patient reported use and cost*
Personalcare
$200. 1 visit per day @$200/month
$2,400 Patient reported use and cost*
Community Nurse visit
$56 Weekly visits @ 30 min/visit for 12 months
$1,456 Department of Health and Aging**
* Patients were asked to describe support service use and costs for the past 4 weeks. For the purposes of this model it
was assumed that the rates of service use and cost were constant over the full 12 month period (i.e. reported cost per
month multiplied by 12).
A model was constructed to reflect differences in both the probability of support service
use and the type of services used (see Table 9.11 and Table 9.12). The frequency of
use within the sub-set of the OP and DAA groups between strata 2 to 6 was multiplied
by the cost to produce an aggregate measure of support service use that reflects the
fact that many of the patients receiving community care used a variety of services. A
mean cost per patient receiving community care was derived for both the OP and DAA
groups (see Table 9.12). The assumptions underlying this model are that 47.1% of OP
patients received support services and 58.9% of DAA patients (as per patient reports
for strata 2-6) therefore in a sample of 30 OP and 30 DAA patients, 14.13 OP patients
and 17.67 DAA patients would incur some cost.
Table 9.12 Probability of using support services and total cost of service for matched
sample (strata 2-6)
OP % Number Cost Cost of service Cleaning 75.0% 10.6 $2,080.00 $22,042.80 Meals 12.5% 1.8 $1,560.00 $2,755.35 Personal care 4.2% 0.6 $2,400.00 $1,424.30 Community Nursing 16.7% 2.4 $1,456.00 $3,435.74 $29,658.19
Mean cost per OP patient receiving care = $29, 658.19/14.13 = $2,098.95 DAA % Number Cost of service Cleaning 72.7% 12.8 $2,080.00 $26,719.87 Meals 30.3% 5.4 $1,560.00 $8,352.26 Personal care 6.1% 1.1 $2,400.00 $2,586.89 Community Nursing 9.1% 1.6 $1,456.00 $2,341.20
$40,000.22
Mean cost per DAA patient receiving care = $40, 000.22/17.67 = $2,513.78
9.5.1.3 Health service use consequences
The probabilities of occurrence assigned to each event are presented in Table 9.13.
The probability of an event occurring was based on the observed frequency of the
event occurring during the 12 months prior to Phase 2 data collection for health service
use or in the 12 months post Phase 2 data collection with respect to community care,
RCF care or death for strata 2-6.
Quality Medication Care Group, School of Medicine, University of Queensland222
9.5.2 RESULTS
Results of the cost analysis model are presented in Table 9.13. The cost per patient
per year in the OP arm was $5,156 compared with $7,966 per patient using a DAA.
Over a 12-month period, the DAA use strategy resulted in 0.7 fewer deaths but cost an
additional $45,040 in health service and support costs. In this model, the cost of PBS
drugs was the highest service cost for the original pack arm of the model but residential
care was the highest service cost for the DAA arm. The biggest difference between the
two arms in costs for a single resource was also in residential care use, with the DAA
arm estimated to cost 1.9 times more than the original pack arm reflecting the fact that
7.4% of DAA patients were admitted to a RCF compared with 3.8% of OP patients.
Overall, 83% of the total difference in costs between the groups was accounted for by
non-medical support (RCF care and community care).
Table 9.13 Results of the cost model comparing DAA use to OP with 30 patients in
each arm
ARM EVENT Probability No. of patients Cost per patient Total Cost
OP PBS Drugs 100.0% 30.0 $2,089.54 $ 62,686.20 GP services 100.0% 30.0 $504.51 $15,135.30 Pathology 87.9% 26.4 $145.17 $3,828.13 Other MBS 22.4% 6.7 $978.79 $6,577.47 MBS Hospital 24.1% 7.2 $1,117.19 $8,077.28 Community Care 47.1% 14.1 $2,098.95 $29,658.19 RCF admission 3.8% 1.1 $25,178.35 $28,703.32 Death 7.6% 2.3
Total Cost for all Events $154,665.90
DAA PBS Drugs 100.0% 30.0 $2,423.46 $72,703.80 GP services 98.5% 29.6 $510.10 $15,073.46 Pathology 85.3% 25.6 $189.93 $4,860.31 Other MBS 20.6% 6.2 $577.16 $5,218.89 MBS Hospital 23.5% 7.1 $844.48 $5,953.58 Community Care 58.9% 17.7 $2,263.74 $40,000.22 RCF admission 7.4% 2.2 $25,178.35 $55,895.94 Death 5.3% 1.6
Total Cost for all Events $199,706.19
9.5.2.1 Sensitivity analysis
Sensitivity analysis has been used to explore the uncertainty in the probability and
mean costs of health service and support service use. Sensitivity analysis 1 explores
variation in the proportion of patients and cost by utilising the unadjusted values or raw
data for the full sample instead of the matched sample data from septiles 2 to 6 and
then by examining the model using data from single strata between septiles 2 and 6.
Sensitivity analysis 2 explores variations in RCF costs where the cost is based on 6
months care in a RCF instead of a full year. Sensitivity analysis 3 examines the
consequences of including hospital accommodation costs in the model.
Sensitivity analysis 1 (Table 9.14) indicates that the modelled health care costs for the
DAA group exceed the OP group in all comparisons with the exception of the sample of
patients that fall in septile 6. When the unadjusted mean costs and probabilities are
used, the difference between the DAA and the OP group (where the DAA group cost
more) is more than twice the difference observed in the base model (using adjusted
Quality Medication Care Group, School of Medicine, University of Queensland223
data and selective sampling (strata 2-6 only). In sensitivity analysis 2, reducing RCF
costs reduces the total costs for both groups and also the difference between the
groups. Similarly increases in the cost of hospitalisation when accommodation is
included, increases the total costs for each group and also the difference between the
DAA and OP arms such that the DAA arm cost an additional $57,227.
The cost difference (DAA arm minus OP arm) varied between -$37,472 and $111,652
(base model $45,040) when sample selection, adjustment for group differences, and
RCF and hospitalisation costs were varied. The model was most sensitive to sample
selection, as seen in Table 9.14. For patients with higher propensity scores (reflecting
higher care needs), DAA users had lower modelled healthcare resource use than OP
users.
Table 9.14 Sensitivity analysis of healthcare resource use
Area of sensitivity Description of Variation Total Cost Cost Difference
1. Selection of sample
DAA base case (2-6) $199,706
OP base case (2-6) $154,666
$45,040
DAA all cases $252,969
OP all cases $141,318
$111,652
DAA Septile 2 $154,612
OP Septile 2 $129,097
$25,515
DAA Septile 3 $155,772
OP Septile 3 $96,624
$59,148
DAA Septile 4 $175,957
OP Septile 4 $127,051
$49,906
DAA Septile 5 $260,158
OP Septile 5 $258,229
$1,929
DAA Septile 6 $151,181
OP Septile 6 $188,652
-$37,472
RCF Costs
DAA base case (2-6) 365 days @ $69 per day $199,706
OP base case (2-6) 365 days @ $69 per day $154,666
$45,040
DAA RCF costs 180 days @ $69 per day $171,375
OP RCF costs 180 days @ $69 per day $140,118
$31,358
Hospital costs
DAA base case (2-6) MBS hospital costs only $199,706
OP base case (2-6) MBS hospital costs only $154,666
$45,040
DAA + hosp. accom $222,231
OP + hosp. accom
Any patient reported hospitalisation
for 3.29 days@ $774 per day $165,003
$57,227
9.5.2.2 Health service use consequences for patients still living in community
setting
Data from Phase 3 (where patients had valid HIC cost data) suggest that 92% (N=83)
and 75% (N=79) of OP and DAA patients, respectively, were still living in the
community. Given that RCF admission was a key cost driver in results for the DAA
arm, the cost analysis model outlined above has been re-analysed using data for only
Phase 2 patients who remained in the community. Table 9.15 outlines the probabilities
and associated cost estimates for these patients. The cost per patient per year living in
the community in the OP arm was $3,461 compared with $3,163 per patient using a
DAA. Over a 12-month period, the DAA use strategy cost $8,950 less in health service
Quality Medication Care Group, School of Medicine, University of Queensland224
and support costs. In this model, the cost of PBS drugs was the highest service cost for
both original pack and DAA.
Table 9.15 Results of the cost model comparing DAA use to OP with 30 patients still
living in the community
ARM EVENT Probability No. of patients*Cost per patient Total Cost
No DAA PBS Drugs 100% 28 $2,285 $63,205.91 GP services 100% 28 $578 $15,997.50 Pathology 92% 25 $231 $5,853.82 Other MBS 27% 7 $1,131 $8,291.52 MBS Hospital 28% 8 $1,368 $10,491.22 Community Care 47% 13 $2,099 $27,293.37 RCF admission 0% 0 $25,178 $0.00 Death
Total Cost for all Events $103,839.96
DAA PBS Drugs 100% 22 $2,805 $62,708.97 GP services 99% 22 $674 $14,877.33 Pathology 84% 19 $278 $5,191.10 Other MBS 20% 5 $910 $4,119.33 MBS Hospital 23% 5 $1,569 $7,993.61 Community Care 59% 13 $2,264 $29,862.09 RCF admission 0% 0 $25,178 $0.00 Death
Total Cost for all Events $94,890.33 * people still in the community from base model of 30
9.5.3 COST BENEFIT ANALYSIS OF DAAS USING HEALTH SYSTEM DATA FOR ALL PATIENTS
When the cost benefit analysis is done utilising a full range of health service use data
extracted from HIC and patient data as collected in Phase 3 follow-up (see sections
7.1.4 and 6.2.5), the costs of providing DAAs to 30 community customers outweigh the
benefits of DAAs by $63,371 per year (see Table 9.16). ADRs are not included in this
analysis given that the costs associated with treatment are explicitly covered by the
inclusion of HIC costs.
Table 9.16 Results of cost-benefit analysis utilising HIC data and patient outcomes
Formula Value Perspective
Total cost OP $28,281.93 to pharmacist
Total cost DAA $55,769.87 to pharmacist
Net cost of DAA -$27,487.94 to pharmacist
Total HIC and support costs OP $154,665.90 to healthcare system
Total HIC and support costs DAA $199,706.19 to healthcare system
Net HIC and support costs -$45,040.29 to healthcare system
Willingness to pay for DAA $9,157.20 to patient
Net social benefit (or cost) -$63,371.03
9.5.4 COST BENEFIT ANALYSIS OF DAAS USING HEALTH SYSTEM DATA FOR PATIENTS
LIVING IN THE COMMUNITY
For patients living in the community, when the cost benefit analysis is done utilising a
full range of health service use data extracted from HIC and patient data as collected in
Phase 3 follow-up (see section 6.2.5) the costs of providing DAAs to 30 community
customers outweigh the benefits of DAAs by $9,381 per year (see Table 9.17).
Quality Medication Care Group, School of Medicine, University of Queensland225
Table 9.17 Results of cost-benefit analysis utilising HIC data and patient outcomes
Formula Value Perspective
Total cost OP $28,281.93 to pharmacist
Total cost DAA $55,769.87 to pharmacist
Net cost of DAA -$27,487.94 to pharmacist
Total HIC and support costs OP $103,839.96 to healthcare system
Total HIC and support costs DAA $94,890.33 to healthcare system
Net HIC and support costs $8,949.62 to healthcare system
Willingness to pay for DAA $9,157.20 to patient
Net social benefit (or cost) -$9,381.12
9.6 DISCUSSION OF ECONOMIC ANALYSIS
9.6.1 OVERVIEW OF PURPOSE
This evaluation utilises a systematic approach to explore the issues of cost and effect
of pharmacists supplied DAAs and builds on the work done in Phase 2 by addressing
many of the limitations identified (see section 9.9.2 in the Phase 2 Final report (Ientile
et al. 2004)). We have utilised accepted guidelines for costing and innovative
techniques such as decision analytic modelling to allow comparisons between
medication provision by the pharmacy in DAAs or OPs. In the process of evaluating
these alternatives, we have produced a detailed and flexible model that can be
improved as more data becomes available in the future.
9.6.2 LIMITATIONS
In conducting these analyses a number of methodological challenges were
encountered.
A wider social viewpoint is often preferred in economic evaluations. A societal
perspective incorporates costs (and benefits/consequences) to patients, health
professionals, carers, the health care system and other non-health sectors. In
considering the components of economic evaluation in health care (Drummond et al.
1997) (Figure 2.2), costs are those for the health care sector, patients and family, and
other sectors. The consequences can be:
Health state change that can be lead to a measurable effect or measured by WTP.
Other value created measured by global WTP for a whole program.
Cost savings between two alternatives to the health care sector, patients and
family, and other sectors.
While the study proposal indicated that a societal perspective was to be used in a cost
benefit analysis, the detailed study plan and the decision analytic model developed in
Phase 2 was focused on better understanding the healthcare use for DAA and OP
users. In the economic evaluation in this study, the perspective was primarily
concerned with resources used by the pharmacy in preparing and dispensing OPs and
DAAs combined with costs and (potential) benefits to the healthcare system (from the
commonwealth budget perspective) and to patients (health benefits (ADEs), willingness
to pay and various out of pocket expenses for health care). Other predominantly
indirect and intangible costs (and hence related cost savings) were not readily
captured, were limited by the time, resource and study design constraints and
complicated the model. Costs and benefits not reflected in the model included:
Quality Medication Care Group, School of Medicine, University of Queensland226
Carer indirect costs such as travel and time spent (including lost productivity), their
opportunity costs and intangible benefits on carer burden and peace of mind.
Phase 2 showed DAA users were more likely to have a carer overall but where care
needs (IADL) and ill health were similar (any hospitalisation in the previous year),
there was not difference the probability of having a carer. Further, carer costs and
burden were likely to be lower for DAA patients as pharmacies were more likely to
deliver medications (see section 9.3.4 in the Phase 2 Final report (Ientile et al.
2004))and carer burden was lower compared to OP.
State public hospital and health insurer costs although reports of patient
hospitalisation were used as a proxy. Patients were from various states and
potentially used various insurers that made collecting this data impractical.
Patient expenses not covered by MBS or PBS although some costs were captured
by self-report
GP administrative costs in communicating with the pharmacy. The opportunity
costs of this activity would potentially be balanced by fewer visits by DAA users to
GPs, thus allowing the GP to consult with other patients.
Intangible benefits to the patient such as increased confidence in being able to
manage at home and reduced concern over RCF admission and increased
productivity although these aspects may have been included in part, in WTP data.
Productivity gains for community nursing. In phase 2, community nurses reported
that DAA use reduced the frequency of visits and the duration of visits. The
proportion of patients who had received a community nurse visit in the preceding 4
weeks in phase 3 was not different for DAA and OP users.
Benefits in delayed or prevented RCF admission or deaths. These effects would be
delayed and not measurable in this study.
Given resource constraints and the exploratory nature of this research, we believe the
limited perspective (that of the main payer, the main provider and the main beneficiary)
used in the current study was appropriate particularly since these perspectives were
likely to reflect those of decision makers and possible payers (George et al. 1999) and
sheds light on a number of pertinent issues to be addressed in subsequent studies.
Another limitation was that data was missing for some of the Phase 2 patients. The
data on patient medicine and healthcare costs collected in Phase 3 (see 6.2.5) was not
available for those lost to follow-up (a greater number of DAA users had died or moved
into residential care) and not all that collected was included as the response rate to
some of these items was low and responses were often ambiguous. A review of this
data in 6.2.5 suggests that DAA patients incur greater costs than OP patients and the
inclusion of these costs would likely indicate that the societal cost of DAAs is even
higher than the results presented above. These costs however, were unadjusted for
health status but after adjusting for differences such as care needs and age using
propensity scores (see 8.2.1), use of these services was not different.
The lack of baseline data and the non-random nature of the study impacts on the
usefulness of comparison between OP and DAA patients. As the study was cross-
sectional and patients in the DAA group had already started using DAAs, change in
costs and benefits could not be assessed. This difference between groups is reflected
in the fact that there appears to be a fundamental difference (in this study) between
people who use a DAA and people who do not. We found that DAA users (in this study)
were generally sicker and had a greater need for assistance and care as reported in
Section 6.2.
Quality Medication Care Group, School of Medicine, University of Queensland227
The comparison of alternative strategies for enhancing medication provision to
community customers was restricted to DAAs and OPs. Other studies that have
assessed the cost-effectiveness of pharmacy based medication management
interventions in the community have examined Home Medicine Reviews (HMRs)
(Sorensen 2004), payment and training for pharmacist in providing proactive clinical
interventions (Benrimoj 2000) and pharmacy based education and monitoring for at-risk
community patients (Sturgess 2003). In comparison to DAAs, HMRs appear to be a
more cost-effective intervention (incremental cost-effectiveness in reducing ADEs was
AUS$69 (1999 figures) compared with $9,163 in the current study and $6,028 in Phase
2. This is in part explained by differences in intervention costs (one HMR per year is
less than a third of the cost of providing DAAs for a year) and also because the
apparent difference in effectiveness in reducing ADEs (the HMR intervention group
appear to have a 27% reduction in ADEs compared to a 10% reduction in the DAA
group at Phase 3 and 15% in Phase 2). The authors also observed a trend in savings
in health service use for the intervention group of a similar magnitude to the cost of the
intervention. Others (Benrimoj 2000; Sturgess 2003) have also reported trends in
savings in health service use but did not report the total costs of the interventions. A
review of 16 studies on the economic benefits of clinical pharmacy services found that
for every $1 invested in clinical pharmacy services, more than $4 in benefit is expected
(U.S dollars) (Schumock 2003).
Finally, the HIC data included in the models did not include service use for the DVA
gold card holders in the sample. These data were not received in time for this study
due to delays in providing approval for the release of the data and then delays in
extraction of the data. Veterans represented 23.6% of the Phase 2 sample. As
veterans have access to a potentially wider range of subsidised services, their
healthcare costs may be higher than the community patients who were not veterans.
There were some differences in characteristics of veterans and non-veterans (Table
9.18) but from these data, it does not seem that the veteran’s would have a significantly
higher burden of disease than the people included in this analysis as there was no
difference in the proportion of veterans and non-veterans using a DAA (p=0.380) and
no significant difference between veterans and non-veterans overall or within the DAA
and OP groups for the propensity score used to create the septiles (intended to adjust
for differences in care needs) used in the economic analyses. Any higher cost,
therefore, may be related to access to services.
Logistic regression found no significant difference between veterans and non-veterans
overall or within the DAA and OP groups for the following variables: Number of
GPs/doctors regularly visited; How would you rate your health status (poor to
excellent)?; Number of times visited GP in past 2 months; Number of different illnesses
patient reports having; Compared to other people your age how would you rate your
health?; Over the PAST WEEK, how frequently did you experience symptoms?; Over
the PAST WEEK, how would you rate the symptoms you experienced?; How many
different doctors prescribing?; How many different pharmacies have dispensed
medicines?; Total number of current medicines in home; Number of current non-solid
medicines at home; and EQVAS z-scores.
Quality Medication Care Group, School of Medicine, University of Queensland228
Table 9.18 Veterans compared to non-veterans: differences in Phase 2 and certain
Phase 3 variables (shaded variables used in propensity score for economic
analysis)
Overall DAA OP More likely to live alone (p=0.014) NS More likely to live alone
(p=0.038) No difference for regular carer (NS) or regular community health visits (NS)
NSNS
NSNS
Any hospital in last year (NS) NS NSADR in Phase 2 (NS) NS NSADR in Phase 3 (NS) NS NS Fewer females (p=0.056) Fewer females (p=0.012) NS OARS IADL (NS) NS NS More likely to have health problems limiting independence (Phase 3 EQ-5D) (p=0.004); worse mobility (p=0.039) and usual activity score (p=0.074); a lower EQ-5D score (p=0.057)
NS for any problems, NS for sub-scales; NS for EQ-5D score
More likely to have health problems limiting independence (Phase 3 EQ-5D) (p=0.037); worse mobility (p=0.045); a lower EQ-5D score (p=0.087)
No difference in service use (including any hospitalisation) reported in Phase 3 except veterans more likely to have been in a private hospital (p=0.003)
Same pattern as overall (private hospital p=0.016)
NS for any service type
Less likely to take antianxiety and antidepressant medicines (p=0.003 and p=0.045)
encouraging pharmacists to educate patients and other health care workers on the
importance of not only maintaining proper storage conditions, but also to be vigilant
and monitor that the integrity of the DAA is maintained throughout the dosage
period.
Quality Medication Care Group, School of Medicine, University of Queensland238
Attention to risk management and further improvement in the available stability
information at will improve risk assessment. Many research questions have been
raised concerning the repackaging of medications in DAAs. It will take some time and a
significant amount of work before our understanding of the complex physicochemical
stability issues pertaining to the repackaging of medications in DAAs is sufficient to
enable us to develop and implement evidence-based “current Good Packing
Practices”. The following strategies, however, will allow incremental improvement in the
quality of packing decisions and provide a foundation for the development and
implementation of evidence-based “current Good Packing Practices”.
10.2.1 IMPROVING DRUG STABILITY IN DAAS IN THE CURRENT SITUATION
In the short term, pharmacists can use currently available information in more
structured decision making. The risk assessment-risk management approach to drug
stability can be applied to the current level of information. In section 4.3.2 and
Appendix E, the science of the physicochemical stability issues surrounding the
repackaging of medications into DAAs is revised to support pharmacists’ professional
judgement in this area and to assist in improving their packing methods/ procedures to
by considering the various stability issues. A risk assessment decision-tree has been
developed in this study, based on the science relating to physicochemical stability
issues and currently available information (e.g. package inserts; CMI’s) for repackaging
medications in DAAs, to provide some guidance to the pharmacists repackaging drug
products into DAAs. Risk management strategies have been described in 4.3.3 and
4.3.5. Patient education on the preferred storage and use of DAAs is also important.
10.2.2 GATHERING MORE EVIDENCE OF DRUG INSTABILITY AND DRUG STABILITY
In preparing this report, it is evident that many research questions have been raised
concerning the repackaging of medications in DAAs. It will take some time and a
significant amount of work before sufficient evidence on the physicochemical stability of
medications packaged in DAAs is available. The following are three strategies to
increase the information base.
10.2.2.1 Gather reports of suspected physicochemical instability observed in
DAAs
In solid dosage forms, evidence of physicochemical instabilities often present with an
associated change in appearance (e.g. colour changes/mottling, tablet softening/
hardening, odour, disintegration) and can thus be detected upon simple visual
inspection of the product (Aulton 2002). For example, photochemical processes usually
take place on the product surface and photochemically induced interactions in tablets
may frequently lead to discolouration even when chemical transformation is modest or
undetectable (Tønnesen 2001).
Information about apparent changes in appearance should be gathered from
pharmacists and pharmacy staff, other health professional and patients/carers
(especially where patients are encouraged to report any visual or other problems,
which may be related to the medication’s stability, to their health care provider) . This
information, if centrally collated, could become a reference for other pharmacists in the
same way that voluntary adverse drug reaction reporting has informed prescribers and
other health professionals about medication risks. Such a register could also identify
targets for DAA “in-use” stability testing.
Quality Medication Care Group, School of Medicine, University of Queensland239
It is certainly possible for pharmacies to report suspected physicochemical instability
for medicines in DAAs, plus any risk management strategies a pharmacist has put into
place to address possible instability. A brief list offered by pharmacy staff who pack a
heat-sealed blister multi-dose pack in a coastal tropical area of Australia included
(personal communication with Karalyn Huxhagen, 25/10/05 & 28/10/05): Salt tablets (for renal patients) are hygroscopic. They go ‘bumpy’ and leach out to ‘salt’
neighbouring products. The maximum time before this occurred was 2 weeks.
Bioglan melatonin capsules – go transparent after 2 to 3 weeks in a DAA.
Bioorganics cranberry capsules – the powder in the capsule changes colour to dark maroon
and shrinks away from the sides of the capsule after about 3 weeks in a DAA.
Slow K™ tablets go ‘slimey’ after 2-3 weeks in a DAA.
Halved Solprin™ tablets are better protected by the DAA blister pack than wrapping the
halved tablet back in foil and storing it in the original box.
Other observed physical changes that lead to practice changes (deblistering
immediately prior to repackaging) for a DAA provider using sachets include (personal
communication with Dawn Woodward, APHS, 7/11/05): Enalapril tablets – these go soft and crumbly after 2 weeks in a sachet
Hiprex ™ tablets - absorb moisture
Cavedilol
Nicorandil (Ikorel)
Omeprazole (Hexal brand, has 4 layer coating that breaks down/gets cracks after 21 days)
– not seen with Acimax™ or Losec™ brands
Interestingly, the UK manufacturer of Slow K ® and Nicorandil indicated that Slow K
“May absorb water; probably stable in an airtight compliance aid for 14 days” and
Ikorel® was “Stable for at least 7 days in a dry environment” (Church & Smith 2006),
suggesting that caution may be required in interpreting data from the UK situation.
Other drugs (including some of those mentioned above) have been reported as being
sensitive to moisture (personal communication with John Parke, Queen Elizabeth II
Hospital, 14/11/05) – changes were observed when patients using various DAAs
brought the packs into hospital: Amoxycillin/clavulante – note that these products are packaged in a tropicalised pack by the
manufacturers and according to the risk assessment in 4.3.5.3, should not be packed)
Aspirin 100mg non-enteric coated
Felodipine
Moclobemide
Perindopril
The provision of an easily accessible reporting tool for pharmacists, nurses and other
health care workers would be valuable for acquiring evidence of potential medications
and/or combinations of medications that have displayed apparent physicochemical
instability when packed in DAAs. Figure 10.1 shows a possible reporting form.
The form in Figure 10.1 is modelled on the form used by the Adverse Drug Reactions
Advisory Committee’s (ADRAC) “Report of Suspected Adverse Reaction to Drugs and
Vaccines” and the Australian and New Zealand Medical Device Incident Report
Investigation Scheme’s “Medical Device Incident Report” (DAAs are not considered a
device according to the TGA definition (http://www.tga.gov.au/devices/devinfo2.htm)).
The proposed form contains questions to collect information specifically relevant to
assessments of instability. It could be web-based and potentially redesigned to be a
web-based database searchable by pharmacists.
Quality Medication Care Group, School of Medicine, University of Queensland240
Since TGA administers drug and device reporting schemes and is responsible for
assessing drug stability data submitted by manufacturers in the regulatory process and
therefore has the expertise to evaluate reports, it would be reasonable that TGA act as
a repository for reports of suspected DAA drug instability. In the interim, a pharmacy
organisation like PSA or the Pharmacy Guild of Australia under the auspices of QCPP
could develop the web-based databases and store reports of suspected instability. The
data could be derived from voluntary reports of instability. Alternatively, many
pharmacies providing DAAs have their own in-house list of medicines with stability
concerns that relate to the environmental conditions in which a specific type of DAA is
used – a survey of pharmacies providing DAAs would allow collation of empirical
evidence of instability.
10.2.2.2 Stability studies on the most commonly packed solid dosage forms
There are a number of techniques for assessing the stability of medicines but these
can be time-consuming and costly, particularly where stability in each type of DAA
would need to be assessed (even without the complication of testing for possible
interactions within a DAA blister, compartment or sachet). It would therefore be prudent
to prioritise any such studies to target medicines or medicine combinations most
commonly packed in DAAs and where there is a higher index of known/reported
stability concerns. The stability of drug products meeting these criteria could be
investigated under both experimental and in-use conditions in DAAs to begin to provide
the pharmacist with specific stability data. In one approach, the USP used a “Open
Dish Tests“ (accelerated open-dish testing conditions of 30°C and 75% relative
humidity for 7 days, 30°C and 75% relative humidity for 3 weeks and 25°C and 60%
relative humidity for 6 weeks in an open dish ) to screen solid medicines for changes in
appearance, average mass, and dissolution behaviour (Bempong et al. 1999b;
Bempong et al. 1999c) and products showing changes were then assayed for drug
content (Adkins et al. 1999; Bempong et al. 1999a). Procedures for conducting open
dish studies have been published (Medwick et al. 1998a).
Open dish studies would provide data on the worst case scenario and could have the
potential to be a useful screening method. Open dish studies may, however, produce
many ‘negative’ results that would not reflect stability in the more protected
environment within a DAA, and so unnecessarily complicate the ‘screening’ process.
To facilitate the completion of stability studies to reflect repackaging in DAAs and to
reflect the Australian situation, an easily accessible (widely disseminated) guideline
outlining the methodology for conducting these physicochemical stability studies should
be developed in consultation with the various regulatory authorities, manufacturers of
DAAs and experts in the field. This would assist researchers in the field and provide a
standardised approach to acquiring in-use stability evidence.
In this study, it was possible to identify the most commonly dispensed medicines for
patients using DAAs, thus providing a starting point for prioritised stability studies. In
Phase 2 of this study, 136 community patients who used pharmacy-packed DAAs
consented to the release of their Pharmaceutical Benefits Scheme (PBS) claims data
for the period July 2001 to June 2004 (for most patients). This type of data has also
been used in other study analyses and details of its extraction can be found in 7.1.2.
Quality Medication Care Group, School of Medicine, University of Queensland241
Report of Suspected Physicochemical Instability Observed in
Dose Administration Aids
Patient identifier (e.g. Record no.) _____________________
Visual description of observed/ suspected instability problem: __________________
hygroscopic preparations and solid dose cytotoxic preparations. Consideration
must also be given to the effect of heat sealing the backing on some dosage
forms (e.g. soft gel capsules).” (Pharmaceutical Society of Australia 1999)
This guideline poses a problem for the pharmacist in terms of their ability to assess
incompatibilities and although information is provided on the type of medications
adversely affected by removal from the original packaging, there is a lack of detail
pertaining to the individual drug substances. A more detailed list that can be updated
as more information on drug instability in DAAs becomes available, could be included
as an appendix to the “Good Packing Practices” code.
10.3 POLICY AND PRACTICE IMPLICATIONS FOR
COMMUNITY-BASED RECIPIENTS OF PHARMACY DAA
SERVICES
In Phase 3, additional HIC data and follow-up of community patients at one year gave
additional insights into the characteristics of current community patients being provided
DAAs by community pharmacies and suggested other potential benefits from DAAs
that might arise from closer relationships within the patient care team. An
understanding of the characteristics of current DAA service recipients could inform
future policy and practice.
10.3.1 CURRENT RECIPIENTS
Current community patients who use DAAs represent a subset of the population who
have high care needs across many domains compared to users of medicines in original
packs. In section 8.2, the characteristics of DAA users collected in Phase 2 are
modelled – pharmacy-provided DAA users require more assistance with care, were
older and more likely to have been hospitalised in the preceding 12 months than OP
users. The community patients in this study are self-selected. While DAA use was
initially recommended by the pharmacy for 46% of pharmacy-provided DAA users (with
the doctor made the recommendation for 25% and the family for 11%) (Ientile et al.
2004), in this case, continued use of a DAA can reflect perceived net benefit. Using a
DAA has some negative impact on patients: Patients using a DAA have to give up
control of their medicines, to fit the processes of getting packs and dealing with
medication changes (when the whole pack has to be changed if only one medicine is
altered) into their lifestyle and in many cases, pay for the service. Further, as the
economic analyses in Phase 2 illustrate, pharmacies have few incentives to offer a
DAA service to a community patient that does not have a “need” for the service.
The data obtained from HIC service use data and the follow-up of analysis of
community patients has been used in the economic modelling in Chapter 9 but it also
gives more insight into the characteristics of DAA users and the effect of DAA use on
outcome.
In Chapter 7, HIC service use was examined for the year prior to the day of Phase 2
data collection. Despite having greater care needs and acuity/severity of illness (as
suggested by a higher rate of emergency doctor attendances (Table 7.1)) compared to
Quality Medication Care Group, School of Medicine, University of Queensland250
OP users and non-pharmacy DAA users, community patients using pharmacy supplied
DAAs had the same service use costs (PBS+MBS costs). This may reflect better
control of medication management and service use associated with the enhanced
relationships within the health care team needed to provide a DAA service. Further,
non-pharmacy DAA users had the highest combined PBS+MBS costs yet were more
able and younger than pharmacy DAA users (Ientile et al. 2004). This group also had
the highest number of PBS items, the highest levels of disorganised and memory-
related non-adherence despite but the lowest number of GP MBS items (Table 7.1).
Some non-pharmacy DAA users may have benefited from the closer monitoring
associated with a pharmacy DAA service as they actually had higher PBS costs and
lower GP costs than would be expected after adjusting for covariates (Figure 7.1 and
Figure 7.2).
A small sub-analysis allowed the impact of starting a pharmacy DAA on HIC costs to
be examined. In the post-DAA period, DAA users had higher service use in a number
of areas relative to OP users. In some cases, the pattern was opposite to that
observed in the pre-DAA period (Table 7.2). Continuity of supply (or possible wastage)
increased and/or there was a decline in health between pre- and post-DAA timeframes.
DAA users has significantly more PBS items and incurred greater PBS costs than OP
users with a slight (p=0.103) increase in the number of medicines regularly taken
(using the number of different items supplied more than once and those not
antiinfectives as a proxy) and emergency doctor attendances were higher.
Even though many of the covariates in the HIC data models are proxies for greater
disability and/or poorer health, the pattern of costs and the difference between adjusted
and unadjusted costs for the 1 year cross section (Table 7.1) and the pre-post analysis
(Table 7.2) suggests that some key driver of cost, such as a significant decline in
function, occurred for pharmacy DAA users that did not happen for people capable of
using original packs or non-pharmacy DAAs.
The follow-up of community patients 1 year after Phase 2 further demonstrated that
self-selected users of pharmacy supplied DAA are quite different to either non-
pharmacy packed DAA users or users of medicines in original packs in terms of care
needs, disability and markers of illness burden. Pharmacy DAA users were more likely
to be lost to follow-up due to death, RCF admission and ill health (Figure 6.1). As seen
in Phase 3 (Figure 6.4), compared to OP users, pharmacy-provided DAA users who
remained living in the community also had higher care needs at Phase 2. There are
several findings of interest:
Using a pharmacy provided DAA maintains people with higher care needs in the
community (at Phase 2 and in Phase 3 (Table 6.8 and Figure 6.4));
People using OPs who went to residential care had higher care needs and poorer
health status (e.g. more likely to receive regular community health visit and have a
lower IADL score) compared to DAA users who were admitted to RCFs.
For a person using OP, those living alone had a greater rate of death (9.1%) versus
those OP users who did not alone (1.3%). This compares with 10.9% for DAA users
who lived alone and 9.2% who did not live alone.
OP users who died had a better IADL score, younger age, fewer illnesses and
better self-rated health than DAA users i.e. they died despite having better function,
etc,.
One possible explanation for these patterns (Table 6.13) is that greater communication
with the GP, better rapport (as reported in Phase 2) and the closer monitoring involved
Quality Medication Care Group, School of Medicine, University of Queensland251
with profile updating and prescription management together with closer patient/carer
and pharmacist relationship secondary to the provision of pharmacy DAA services may
promote greater collaboration in the health care team compared to the care team of OP
users. This in turn may lead to earlier recognition of patient health problems or inability
to cope so that appropriate management can be put into place (thereby delaying
transfer to RCFs or death). Pharmacy DAA users who were followed up in Phase 3
(and therefore continued to live in the community) still had greater care needs (as
indicated by support service and other care facility use) compared to OP users (Table
6.9). The value-added effects of a more collaborative health care approach associated
with DAA use warrants further longitudinal study.
10.3.2 POLICY AND PRACTICE IMPLICATIONS
It is difficult to evaluate the key economic findings from this research given the
significant differences between groups, patients lost to follow-up, lack of randomisation
and lack of baseline data to monitor improvements in health associated with using a
DAA. DAA patients in this study were sicker, had higher health service utilisation rates
and higher costs. From a quality use of medicine point, the fact that DAA patients are
utilising health services more frequently than OP patients could suggest that DAA
patients are more proactive in maintaining their health and are therefore more likely to
frequent a health professional or that better monitoring with appropriate action is taken.
The PBS data (Table 7.1) suggest that using a pharmacy-provided DAA has an impact
on continuity of medication supply. As might be expected with improved compliance
(plus the potential for wastage when DAAs have to be changed), pharmacy DAA users
tended to use more PBS items than OP users despite taking a similar number of
regular medicines (using the number of PBS items supplied more than once and
excluding anti-infectives as a proxy).
A finding from Phase 2 of this study was that community pharmacy appear to be
providing a much needed service to a fairly specific needs group largely at a cost to
themselves. The continuation of DAA use by all 80 DAA users contacted after 1 year
and that 20% of OP users has started using some form of DAA during that year (Table
6.1) suggests that patients believe that the benefits of the service outweighs the costs
and inconvenience to them. Further, analysis of HIC data (7.2.1), data from Phase 2
and the follow-up consumer survey (6.2.4) suggests that the cost of health services to
the Australian government is not different for DAA and OP users when the former
group have greater care needs and would be expected to have higher costs. Because
of the vulnerable nature of this group of DAA users, it is important that funding is made
available to ensure that DAA users are afforded the best care possible by the
community pharmacy.
To better understand the economics of providing DAA versus OP in a community
setting it is desirable that a randomised controlled trial is undertaken with appropriate
consideration given to the inclusion of economic data collection instruments. For
comparability purposes, the data collection instruments used in the current study
should be implemented but supplemented with more costing specific questionnaires
including a custom designed health service utilisation survey. It is also imperative that
health related utility instruments (such as the EQ5D) are used and administered at
intermittent data collection intervals to enable change in health to be monitored. A more
robust version of contingent valuation (i.e., the willingness to pay questions) would also
Quality Medication Care Group, School of Medicine, University of Queensland252
provide more reliable and useful indicators of social benefit. Economic appraisals are a
useful policy tool but output is dependent upon quality data input.
10.4 FUTURE PROVISION OF DAAS TO COMMUNITY PATIENTS
DAAs have been provided by pharmacies to community patients in the belief that a
DAA will improve adherence and medication management, reduce care needs and
bring health and/or quality-of-life benefits ((Ientile et al. 2004) p 84). There are,
however, disadvantages and risks for the patient and for some, not using a DAA may
be the best option and safest for the patient. In providing a DAA service, pharmacies
need to target recipients to maximise patient benefits and design the service provided
(e.g. start up training on how to use a specific device correctly or regular patient
education to counteract loss of medication regimen knowledge) to minimise risks to the
patients. Since the status and medication management abilities of community patients
using a DAA may change over time, pharmacies need to re-assess the situation
periodically to ensure that any risks continue to be addressed by the DAA service.
Targeting of DAA use to people most likely to benefit may maximise the effectiveness
of DAAs in improving adherence and medication management, with resulting health
benefits. The effect on adherence is, however, likely to be modest. In a Cochrane
review of randomised controlled trials (RCTs) that measured the adherence and
disease outcome effects of interventions aimed helping patients follow prescribed drug
therapy for medical problems, Haynes et al (Haynes et al. 2002) concluded that
“current methods of improving adherence for chronic health problems are mostly
complex and not very effective”. When improving compliance of older community
patients was the focus of a systematic review (van Eijken et al. 2003), there were two
findings of interest:
That in this group, compliance rates in many studies were over 80% even in the
control subjects.
That multifaceted interventions or those tailored to the problems of an individual
were often more effective at improving compliance.
Interventions that provide a form of social support, directive guidance, have also been
shown to improve adherence, knowledge and outcome (Caplan et al. 1980; Levy
1983). Directive guidance activities include providing information, instruction and
advice and, giving feedback on patient behaviours, thoughts and feelings (in this case
as they relate to health, illness, medication taking and medication management)
(Barrera et al. 1983).
The type of DAA service described in the best practice model (see section 4.5) is
multifaceted (including the device, written information, counselling, and follow-up,
possibly via a home visit; all facets allow directive guidance to be provided) but at the
same time, the needs of the individual are recognised in the assessment process. The
current targeting of community-based DAA recipients is discussed in 10.4.1.1 and the
implications for future on-line eligibility assessment and service evaluation is discussed
in 10.4.2. Targeting of those DAA recipients with the greatest need and those most
likely to benefit from a DAA is particularly important now that a finite pool of funding for
DAA services has been negotiated in the Fourth Community Pharmacy Agreement
between the Commonwealth of Australia and the Pharmacy Guild of Australia
(http://www.guild.org.au/public/cpa/fourthcpa.pdf viewed 10/5/06) signed 16 Nov 2005.
Quality Medication Care Group, School of Medicine, University of Queensland253
10.4.1 TARGETING COMMUNITY PATIENTS FOR DAA SERVICES
To date, the community patients who are receiving a DAA service from their community
pharmacy have had the device recommended by their pharmacy, their GP, family or
other health care providers or in response to a serious health event (e.g.
hospitalisation). The limited pre-post comparison of HIC data supports the argument
that a serious health event or decline in health or function triggers DAA use (see
10.3.1). As mentioned in 10.3.1, continued use of a DAA in these circumstances
suggests perceived net benefit from the service. To identify the characteristics of
people likely to benefit (using continued use as a proxy in the absence of specific
health outcome measures), in chapter 8, the characteristics of Phase 2 community
patients who used either pharmacy-provided DAAs or OPs were examined to see
which characteristics predicted DAA use. The logistic and non-linear modelling
reported in this chapter suggest that it may be possible to target specific community
patients for a DAA but that criteria are probably complex. In this section, we examine
the nature of the DAA predicting characteristics, compare the results of the two
modelling strategies trialled and discuss the limitations of the models.
10.4.1.1 Predicting DAA use
In the literature and current Australian guidelines, there are similarities and differences
in the criteria used to predict who would benefit from a DAA. In the Phase 2 literature
review where the type of patients who might benefit from a DAA was discussed, those
said to receive possible benefit from DAAs were assumed to be the same as those with
adherence problems. These are summarised as follows:
Unintentional non-adherence – forgetful or confused by complex regimen (Bond et
al. 1991).
Decreased manipulative skills or short term memory loss or confusion (Cramer
1998; Pharmacists develop compliance aid for the elderly. 1988).
Psychiatry patients if anxiety/mood affect adherence but not if any cognitive
dysfunction (Bazire 1984).
Living alone or more than 2 drugs or pre-dementia (MMSE score <24) (Barat et al.
2001).
Elderly because polymedicine and complex regimens more are likely rather than
older age itself (Cramer 1998; Lorenc et al. 1993; Pillans et al. 1999).
To assist carer to monitor meds and decrease carer burden (Rivers 1992).
Certain patient subgroups: respiratory pts; cardiovascular, neurology, renal, HIV,
diabetes patients; starting anticoagulants; others with lots of solid medications
(Naughton et al. 1993; Peterson et al. 2003; Simmons et al. 2000).
DAAs are not suitable if frequent medication changes (Sprey 1995).
Note that the literature is somewhat contradictory in that people starting anticoagulants
are more likely to have frequent medication changes as the dose of the anticoagulant is
adjusted.
The variability in the literature describing the characteristics of people who may benefit
from using DAAs is reflected in the PSA guidelines that emphasise the importance of
individual assessment in determining the need for a DAA. Never the less, the following
criteria are included in the guidelines for consideration (Pharmaceutical Society of
Australia 1999):
Patients who take five or more medications;
Patients who have a history of problems managing medication;
Quality Medication Care Group, School of Medicine, University of Queensland254
Patients with complex medication regimens, and;
Patients showing signs of cognitive or physical impairments affecting their
medication management abilities.
Given that pharmacists were the main instigator of DAA use by the community-based
patients in Phase 2 who used a pharmacy-provided DAA, it is also important to
consider the type of patients who pharmacists believe would benefit most from a DAA
Elderly;
Multiple medications;
Confused or with cognitive impairment ;
Unable to coordinate their medications correctly;
Physical barriers to medication taking;
Limited support.
The criteria pharmacists used to decide whether a patient should be offered a DAA
were more varied but matched the type of patient pharmacists believe would more
benefit. Evidence of confusion or cognitive impairment displayed by the patient was the
most common trigger to offer a DAA (Ientile et al. 2004) p86.
The characteristics predicting DAA use based on models of community patient choice
to continue to use a DAA are summarised in Table 10.3. In this study, two modelling
techniques were used to model DAA and OP use based on various initial sets of
characteristics. Its is important to recognise that the variables initially included in the
models can affect the final predictive model. For example, including regular community
health visits and having a carer in a logistic model accounts for all the independent
predictive power of IADL – excluding community health visits and carer means that
IADL becomes an important predictor. There are, however, three variables that are
important predictors in each of the models shown in Table 10.3:
Any hospitalisation in the preceding year as reported by the patient;
Age;
Response to the question “Do you ever forget to take your medications”.
Whenever it was initially included in a model, “regular community health worker visits”
was also an important predictor of DAA use.
Interestingly, certain characteristics were not predictors in any of the models in Table
10.3:
Self rating of health
Comparative health
Gender
Disorganised non-adherence score
Any antidementia medicines – although any antidementia medicines was an
independent predictor in logistic model 4 (Table 8.9) and so a decision on the value
of antidementia medicines as a predictor should be reserved
Any lipid lowering medicines
Any beta blocker medicines
Are you careless about taking your medicines
The patient characteristics predictive of DAA use derived from the modelling
summarised in this chapter generally agree with the literature and guidelines but with
some exceptions:
Quality Medication Care Group, School of Medicine, University of Queensland255
The Meichenbaum question about forgetting to take medicines and the memory-
related adherence score support unintentional adherence where this is
acknowledged as a predictor of DAA use.
The IADL score and antirheumatoid/anti-Parkinson’s medications as predictors
support decreased manipulative skills as a predictor.
Living alone was a predictor in Model 2 and 3 and the tree based on Model 5 but
may well be less important than predictors of higher care need such as IADL,
having regular carer or community health visits, as living alone appears towards the
bottom of two tree branches in a single non-linear model.
The number of solid medicines a person took was not as important a predictor of
DAA use as the number of non-solid medicines. As the number of non-solid
medicines increases, the inconvenience in using a DAA plus a range of other non-
solid packs increases and the added complexity in the medication taking routine
increases risks of non-adherence. In the decision trees, the number of solid
medicines was a moderator only after other care needs characteristics and the
number of non-solid medicines were considered.
The literature suggests that older people were likely to need a DAA because of
underlying factors related to polymedicine and complex regimens but in the models
in this study, older age was an independent predictor of DAA use after adjusting for
higher care needs in logistic regression models and was a node only after higher
care needs and non-adherence were considered. The importance given to age in
these models needs to be interpreted with caution, however, as 46% of DAA users
in Phase 2 had had their DAA recommended by their pharmacist. That age was a
predictor of DAA use may merely reflect the beliefs of pharmacists in Phase 2
where 48% indicated that being elderly was a characteristic of patients who would
benefit most from using a DAA (Ientile et al. 2004) p 85.
Having a regular carer was a predictor of DAA use in logistic models and two trees.
Phase 2 patients with less frequent medication changes were more likely to use a
DAA. Patients with more illnesses, seeing more doctors, higher symptom
frequency and severity and needing multiple medicines affecting Angiotensin II or
additional other antihypertensives may well be more unstable so that the problems
caused by changes outweigh perceived benefit from DAAs.
Contrary to the literature suggestion, respiratory patients were less likely to use a
DAA (possibly because more of these medicines are non-solid).
Apart from other medicines to treat heart failure, use of other cardiovascular drugs
did not predict DAA use. Indeed, for medicines affecting Angiotensin II or additional
other antihypertensives, greater use was associated with a greater probability of
using medicines in original packs. Use of lipid lowering agents and beta blocker
were not predictors. Nitrates as antiangina medicines were not modelled.
Renal and neurology patients with epilepsy were not modelled, nor were people
with human immunodeficiency virus therapy or those on anticoagulants.
Use of oral hypoglycaemic agents was a predictor in trees but not in logistic
regression models, perhaps because other care needs variables were stronger
influences on the decision to use a DAA.
The effects of confusion or cognitive impairment on the decision to use a DAA was not
well addressed by this modelling for three reasons:
It is likely that few cognitively impaired community patients were recruited into
Phase 2. At recruitment, Phase 2 participants had to be capable of answering
Quality Medication Care Group, School of Medicine, University of Queensland256
interview questions and completing a questionnaire (with the help of a carer if
necessary);
There was no explicit measure of cognitive function included in the Phase 2 data;
Only 7 patients were taking specific antidementia medicines, although possible
effects were noted even for this small number of cases.
Table 10.3 Variables initially included in the logistic and decision tree models
Model
2
Tree
2
Model
5
Tree
5
Model
6
Tree
6
Tree +
Prop
Living alone
Having a regular carer
Regular community health worker visits
No. doctors usually seen
Self-rating of health
No. times a GP was seen in last 2 months
No different illnesses reported by patient
Patient reports any hospitalisation in last year
Comparative health
Frequency of illness symptoms in past wk score
Severity of illness symptoms in past wk score
OARS-IADL score
Age
Gender
Total No. of current medicines
Total No. of current solid medicines
Total No. of current non-solid medicines
Memory-related non-adherence score
Deliberate non-adherence score
Disorganised non-adherence score
Any antianxiety medicines
Any antidementia medicines
Any antidepressant medicines
Any antipsychotic medicines
Any antirheumatoid/Parkinson’s disease drugs
Any lipid lowering medicines
Any oral hypoglycaemic medicines
No. respiratory medicines
Any beta blocker medicines
No. medicines acting on angiotensin II
Any other antihypertensive medicines
No. other medicines used to treat heart failure
Days since last medication change
Do you ever forget to take your medications
Are you careless about taking your medicines
When you feel better do you stop taking your
medicines
If you feel worse do you stop taking your
medicines
Included initially but removed; Retained in the final model Prop=propensity score
10.4.1.1.1 Comparison of linear, logistic models and non-linear models
The non-linear modelling technique may better capture the complex decision making
involved in DAA use and certainly better reflects the tenor of the PSA guidelines that
Quality Medication Care Group, School of Medicine, University of Queensland257
recommend an individualised approach to assessment. Table 10.4 compares the
accuracy, sensitivity and false negative rates for the various models used to predict
DAA use or not. Only the results from the initial decision tree model in this table are
directly comparable with the logistic regression results as the others involve additional
techniques8 to increase the robustness of the initial tree. These techniques test how
well the tree would generalise to unseen data (i.e. data not in the training set). Logistic
regression models will almost certainly not generalize as well when applied to unseen
data.
In part, the more sensitive non-linear models may be related to the nature of the mainly
categorical predictor variables used. In general, logistic regression techniques cannot
easily handle categorical variables and they are weak at detecting interactions between
variables. Decision trees handle interactions between variables much more robustly as
cases are partitioned and then analysed in each group separately. Decision trees are
also robust when dealing with missing and noisy data. About 10% of data is missing
within this data set and some variables are highly skewed therefore decision trees will
be suited well for this data. Decision trees also handle both categorical and continuous
variables. The target value must have discrete values which is true in this case (using
original pack medicines or a pharmacy packed DAA) but unlike logistic regression more
than two responses can be used.
Table 10.4 A comparison of specificity and false alarm rates
Logistic Regression
Initial Decision Tree
Decision Tree with Boost 50 and ten-fold Cross Validation
Combined Logistic Regression and Decision Tree
Model2
Accuracy: 78.7% Sensitivity: 78% False Alarm: 21%
Accuracy: 82.3% Sensitivity: 87% False Alarm: 21%
Accuracy: 70.1% Sensitivity: 67% False Alarm:18%
Accuracy: 72.3% Sensitivity: 72% False Alarm: 27%
Model5
Accuracy: 79.0% Sensitivity: 78% False Alarm: 20%
Accuracy: 83.9% Sensitivity: 85% False Alarm: 17%
Accuracy: 63.1% Sensitivity: 65% False Alarm:30%
Model6
Accuracy: 79.5% Sensitivity: 79% False Alarm: 20%
Accuracy: 80.1% Sensitivity: 82% False Alarm: 21%
Accuracy: 62% Sensitivity: 65% False Alarm:30%
The trees do include many of the same predictor variables as the logistic regression
models indicating that they both model the same underlying concept, however, the
trees also highlight the importance of some other variables in making accurate
predictions. For example, variables about the number of medicines, total, solid or non-
solid do not appear in trees 2, 5 and 6 because any predictive ability they have is
already covered by other variables but the trees use more classification nodes based
on specific drug classes compared to logistic models. It is possible that these factors
are less linear in nature than the main factors of the logistic regression. It is also
possible that there is some covariance between these factors to which logistic
regression is not as sensitive to as the decision tree algorithm. In the future some
combination of the two methods may produce the most accurate classifiers.
8 The accuracy of these initial trees could be increased by boosting but this would lead to over-
fitting of the decision tree to the training data. This means that the error rate when predicting
new cases would be higher. In an attempt to estimate the predictive accuracy for new cases,
cross-validation has been used in the decision tree models.
Quality Medication Care Group, School of Medicine, University of Queensland258
10.4.1.1.2 Limitations of models
While the models provided a greater understanding as to the characteristics of
community patients who continue to use a DAA, there are limitations. In interpreting
these models, it is important to recognise that the outcome being modelled is a proxy,
‘choice’ not ‘benefit’ directly. ‘Choice’ may be influenced by factors other than benefit,
such as access and availability. To model ‘benefit’, random assignment to DAA or not
would have been required. This type of experiment was beyond the scope of this study.
As found in this study, the resulting model depends on the training data set and the
following limitations should be considered:
Subjects were also recruited by community pharmacists, and while there was
approximate case matching to users of original packs, systematic bias is possible
Some potential DAA users were under-represented (e.g. people with mental illness,
cognitive impairment and the intellectually disabled).
Some potentially important predictors were:
Not captured e.g. a measure of cognitive function and more objective measures
of burden of illness or severity, a measure of medication regimen complexity.
Excluded from models because the use of a DAA may influence the response
(as these predictors were not measured at baseline but after DAAs were used)
(e.g. intentional non-adherence).
Had low prevalence in the training set (e.g. use of antidementia medications).
The size of the training set was small and there was no validation set for the logistic
regression. For the trees, this may lead to overlearning which, for example, might
account for the inclusion of the number of illnesses in Tree 2.
Care also needs to be taken in interpreting models that include having a carer or
regular community nurse visits as access to these services may be affected by
availability and eligibility, rather than reflecting a need for a DAA.
These models do provide an insight into who has a perceived benefit from using a DAA
in the community (using choice as a proxy) and could be used as a basis for the
development of eligibility criteria for a subsidised DAA service. Further refinements of
the models and more cases are desirable, however, in order to more reliably predict
likely benefit from a DAA in an unseen case, i.e. a person being assessed for eligibility
for a subsidised DAA service.
10.4.2 BUILDING FUTURE EVALUATION AND RESEARCH INTO THE IMPLEMENTATION OF
DAA SERVICES FOR COMMUNITY PATIENTS
The cross-sectional studies carried out in Phase 2 and 3 of the study into the
effectiveness cost-effectiveness of DAA services have not been designed to assess
health outcome effects (i.e. change in adherence and change in health outcome). The
Cochrane review (Haynes et al. 2002) suggests that such effects are modest as
measured by changes in disease markers (although in the Cochrane review only 2
studies included interventions that might be considered as a DAA). It may well be that
the wrong effects or benefits are being measured in the studies included in the review
as the work in Phase 2 describes more humanistic outcomes such as better
satisfaction and confidence in managing medications and reduced carer burden.
A DAA service that is provided over an extended period is a mutifaceted and diffuse
intervention. The service may impact on medication management more generally with
subtle consequences. It is possible that the effects of DAAs in these areas (as reported
Quality Medication Care Group, School of Medicine, University of Queensland259
by community patients in Phase 2) are mediated by mechanisms other than improved
compliance alone, similar to the social and psychosocial aspects suggested to be
responsible for positive effects of pharmaceutical care (Bernsten et al. 2001). More
research is required to establish whether there is a causal link between DAA use and
better outcome as well as providing additional data for economic evaluation of any
future DAA service (as mentioned in 10.3.2). With the introduction of a subsidised DAA
program, there is an opportunity to collect evidence of benefit following the introduction
of the service while increasing the quality and reducing the cost of subsequent
evaluation by collecting data prospectively. While an RCT is a well recognised means
of attributing causality, more recently, longitudinal multivariate analysis of large
datasets (e.g. the clinical practice improvement methodology used by Susan Horn (see
http://www.isisicor.com/Detailed_Overview_of_CPI.html)) has been used to may inform
practice-based research questions. The key elements of the longitudinal data approach
are:
Recording the nature of ‘treatments’, in the case of DAAs, the type of service
provided (e.g. 6 monthly in-home medication counselling) with any additional
services such as HMR, episodes of hospitalisation or aged care admission and
other health service use.
Standardised recording of patient covariates including demographics, illness
severity or burden, function, complexity or psychosocial factors such as cognitive
function. For DAA services, baseline adherence would be included.
Outcomes of interest e.g. health status, quality of life and the humanistic outcomes
such as those measured in Phase 2. Measures used to evaluate the economics of
the service would also be included such as willingness to pay and health utility
measures would also be important.
Data is collected at multiple points in time.
In 10.4.1.1, we suggested that, by and large, pharmacists have been applying best
available evidence in selecting community DAA recipients. The data elements collected
in Phase 2 provide a starting point to create for formalised decision rules to target DAA
services to those in need (and by proxy, those who benefit). Using machine learning,
the decision rules can be improved as more cases are added and, if implemented as
an on-line eligibility check or registration system, the data can become the basis of the
future evaluation of the service. An introduction to machine learning can be found in
Appendix J.
In order that only the most necessary data are collected prospectively, namely that
needed for eligibility checking and a minimum evaluation dataset, a plan for future
evaluation should be developed in parallel with any program implementation plan.
In the community patient best practice model, structured patient assessment is
recommended together with re-assessment or review and concordance checks 6-
monthly. These are opportunities for data collection that informs both the care of an
individual patient and the effects of the program overall. Assessment should include
validated instruments (e.g. OARS-IADL, the BMQ (Svarstad et al. 1999)) and more
objective measures such as those used in this study (this would also allow comparison
with data collected in this study). The adherence measures used in this study were the
4 Meichenbaum questions and seven other questions developed in Australia and used
and validated in this and another study. Other measures omitted in this study could
Quality Medication Care Group, School of Medicine, University of Queensland260
also be included e.g. a measure of cognitive function such as the Mini-Mental State
Exam (MMSE).
In order of evaluate the future DAA program, other measures to include in data
collection from the patient that might be useful to a health economic evaluation would
be:
Confidence in medication management abilities.
A health utility measure, EQ5D.
Other health measures such as perceived health and health transition.
A custom designed health service utilisation tool to capture information on service
use.
A more robust version of contingent valuation (willingness to pay) as a measure of
benefit.
Consent to access health service use data held centrally (e.g. by Medicare
Australia, formerly HIC) could also be sought.
Collection of data on other health outcomes would also inform the evaluation, such as:
Measures of disease severity.
Any admission to an RCF and the date of this admission (allowing survival
analysis).
Whether a person died and the date of death (for survival analysis).
These additional data for evaluation might be collected using a sampling plan.
The evaluation design will need to take into account:
That a comparison group would be required.
That the outcome effects of the intervention are likely to be delayed (e.g. delay in
disease progression, hospitalisation avoided, RCF admission delayed, death
delayed).
That the immediate effects may well be of a small magnitude in a study population
which has many potential covariates for which adjustment will be required.
That there is potential for variation in the level of intervention (i.e. the DAA service
provided).
That any sampling may involve a stratified (one means of adjusting for certain
covariates) and/or clustered design (clustered on the pharmacy providing the DAA).
A comprehensive evaluation plan is needed that addresses data collection and planned
analysis to capitalise on the opportunities for prospective data collection offered by the
implementation of a program of subsidised DAA services for community patients.
10.5 CONCLUSIONS
The preliminary best practice models for the provision of DAA services to RCFs and
patients living in the community developed in this study address the key barriers to the
provision of safe, effective and efficient DAA services. An evaluation of these models
has found that, they are likely to be beneficial in achieving improvements in practice
and generally feasible. Implementation of some recommendations in the models would
require only small changes to existing systems but others would require substantial
change, and particularly in the case of continuity of care guidelines, time. The
evaluation of the models also identified aspects of the models and recommendations
that could be improved but also highlighted areas of disagreement among stakeholders
about what direction changes should take, for example, about standardisation versus
flexibility and individualisation. Agreement among stakeholders on such key principles
Quality Medication Care Group, School of Medicine, University of Queensland261
needs to be reached to provide direction for further revision of the preliminary best
practice models.
While individual practitioners have and could implement aspects of the best practice
models, further development of the best practice models, supporting tools, resources
(such as improved access to information about drug stability in DAAs) and procedures
with the consultation and participation of stakeholders is required before wide spread
implementation. Revision of the model will also be required with the advent of changes
to the health care system such as changes to the PBS and the introduction of e-health
initiatives.
In developing the best practice models, a review of the evidence supporting the stability
of medication in DAAs was undertaken. The stability information available to
pharmacists is extremely limited. A range of strategies to acquire more evidence on
which to make decisions about the likely stability of medications in DAAs have been
suggested.
The modelling of the provision of DAA services to community patients highlighted that
people receiving DAA services from community pharmacies are fundamentally sicker
and have greater care needs that patients using medications in original packs (OP).
Analysis of HIC (now Medicare Australia) service use showed that users of pharmacy-
provided DAAs had the same service use costs (pharmaceutical and medical benefit
scheme costs) despite having greater care needs and illness burden than OP users.
Follow-up of community patients from Phase 2 at one year supported these findings.
As expected for a group with higher care needs, there was a higher rate of death and
RCF admission at one year for DAA users, but OP users who had died at one year had
better function and health in Phase 2 compared to DAA users. This suggests an
additional benefit of DAA use that using a pharmacy-provided DAA maintains people
with higher care needs in the community. This may reflect better control of medication
management and service use associated with the enhanced relationships within the
health care team needed to provide a DAA service.
The characteristics of community patients receiving DAAs largely agrees with current
recommendations and literature. Characteristics of community patients found to predict
perceived benefit as indicated by a choice to continue DAA use included:
Any hospitalisation in the preceding year as reported by the patient;
Age;
Response to the question “Do you ever forget to take your medications”;
Greater care needs as indicated by regular community health worker visits and
impaired instrumental activities of daily living (IADL) score.
The economic modelling indicated that community pharmacies appear to be providing
a much needed service to a fairly specific needs group largely at a cost to themselves.
In Phase 2, the use of DAAs in RCFs was shown to minimised overall costs. For
community patients outcomes such as better satisfaction and confidence in managing
medications and reduced carer burden were described in Phase 2, however, the cost
effectiveness for community patients was less favourable than for RCFs. Follow-up at
one year and analysis of HIC service use data suggests additional benefits but the
additional economic modelling undertaken in this phase did not alter the conclusions
regarding the cost effectiveness of DAA services for community patients derived in
Phase 2. With the negotiation of funding for a subsided DAA service for community
Quality Medication Care Group, School of Medicine, University of Queensland262
patients, there is an opportunity to better understand the cost-effectiveness of the
service by developing an evaluation plan in parallel with the implementation plan for
this service and collecting data prospectively as the new program is rolled out. The
preliminary best practice model for the provision of DAA services to community patients
should be used to inform the implementation plan for this new service.
10.6 RECOMMENDATIONS
The following recommendations relate to both services to community patients and to
RCFs. The recommendations are shown separately for convenience although a
number of recommendations are duplicated.
Recommendations for DAA service provision to community patients
1. Dose Administration Aids (DAAs) should be targeted to community patients with a
need for and likelihood of benefit from the service.
2. Criteria need to be defined to assess need and likelihood of benefit for community
patients. Modelling conducted in this and other studies should be used to inform the
criteria.
3. A structured patient assessment protocol for community patients should be
developed to determine need for a Dose Administration Aid. This assessment
should be repeated at intervals to monitor the effects of the service.
4. The Pharmaceutical Society of Australia (PSA) in conjunction with the Therapeutics
Good Administration (TGA) should develop a “current good packing practice”
document using the code of good manufacturing practice as a frame work. The
good DAA packing practice code should include staff training and competencies.
These could be developed by PSA and the Pharmacy Guild and included in the
Quality Care Pharmacy Program (QCPP) and in dispensary technician training
programs.
5. The strategies identified in this study to better define drug stability in Dose
Administration Aids should be implemented.
6. There needs to be an overarching emphasis on quality in the provision of Dose
Administration Aids services.
7. A Dose Administration Aid service should reflect best practice to optimise the
provision of a safe effective and efficient DAA service.
8. It needs to be recognised that such a model involves the effective collaboration of
the patient care team including the doctor, the pharmacist and community nurses,
patients and carers, and government.
9. The quality assurance measures included in the best practice model should be
seen as a priority for implementation.
10. The issues and strategies included in the preliminary best practice model should be
widely disseminated and the principles integrated into existing QUM and safety and
quality initiatives such as the APAC guidelines (continuity of care and community),
PSA guidelines, QCPP, Aged Care Assessments, Divisions of General Practice
Aged Care, Home and Community Care, Commonwealth Department of Health and
Ageing, Department of Veterans’ Affairs and state health programs.
11. Further consultation among stakeholder peak bodies is needed to reach agreement
on and to define that desired future direction of the best practice models. This
Quality Medication Care Group, School of Medicine, University of Queensland263
should include defining a continuing development plan for the models and an
implementation plan for DAA best practice initiatives.
12. The preliminary best practice model and findings of its evaluation should be used to
inform the development of the implementation plan for a subsidised DAA service for
community patients.
13. The Pharmacy Guild of Australia should convene a working party including
appropriate stakeholder representation (see 5.2.4.2) and participation to develop an
implementation plan for a community DAA service that embraces best practice,
accountability and quality principles.
14. An evaluation plan should be developed in parallel to an implementation plan to
prospectively collect data to inform future program evaluation. Patient assessment
can provide data for both eligibility checking and future evaluation.
15. The implementation working party should seek input from, and convene technical
advisory panels of experts to advise on (1) further development of the knowledge
base about drug stability in DAAs, (2) the evaluation of the health economics of the
subsidised community DAA service, and (3) refinement of eligibility criteria.
16. Any business rules developed for subsidised DAA service to community patients
should embrace the principles of best practice described in the preliminary best
practice model developed in this study or any subsequent revision.
17. It is recommended that an online system be used to register and assess eligibility
for subsidised DAA provision to increase efficiency and to capture data for later
evaluation.
18. The development of documents and protocols to support best practice in the
implementation of a subsidised DAA service should include trialling and use
document/web design expertise.
19. It is recommended that the government support the use of Dose Administration
Aids services in the community where patients meet the appropriate access criteria
and the service provided reflects best practice.
Recommendations for DAA service provision to RCFs
1. Appropriate Dose Administration Aids services should be encouraged in RCFs and
the role of the service in minimising overall medication management costs in RCFs
should be recognised for appropriate funding to the suppliers of the service.
2. The Pharmaceutical Society of Australia (PSA) in conjunction with the Therapeutics
Good Administration (TGA) should develop a “current good packing practice”
document using the code of good manufacturing practice as a frame work. The
good DAA packing practice code should include staff training and competencies.
These could be developed by PSA and the Pharmacy Guild and included in the
Quality Care Pharmacy Program (QCPP) and in dispensary technician training
programs.
3. The strategies identified in this study to better define drug stability in Dose
Administration Aids should be implemented.
4. There needs to be an overarching emphasis on quality in the provision of Dose
Administration Aids services.
Quality Medication Care Group, School of Medicine, University of Queensland264
5. A Dose Administration Aid service should reflect best practice to optimise the
provision of a safe effective and efficient DAA service.
6. It needs to be recognised that such a model involves the effective collaboration of
the patient care team including the doctor, the pharmacist, RCF staff, patients and
carers, and government.
7. Systems whereby a legal order on a medication chart can act as a prescription as
part of a supply claim mechanism should be established to improve the safety,
effectiveness and efficiency of DAA provision to RCFs. This may require changes
to state and commonwealth regulations, and current Medicare payment claim
procedures. Any new system should address accountability, safety and regular
medication review.
8. The quality assurance measures included in the best practice model should be
seen as a priority for implementation.
9. The issues and strategies included in the preliminary best practice model should be
widely disseminated and the principles integrated into existing QUM and safety and
quality initiatives such as the APAC guidelines (residential care and continuity of
care), Aged Care Standards, PSA guidelines, QCPP, Divisions of General Practice
Aged Care, Commonwealth Department of Health and Ageing, Department of
Veterans’ Affairs and state health programs.
10. Further consultation among stakeholder peak bodies is needed to reach agreement
on and to define that desired future direction of the best practice models. This
should include defining a continuing development plan for the models and an
implementation plan for DAA best practice initiatives.
11. The Pharmacy Guild should convene a working party including appropriate
stakeholder representation (see 5.2.4.2) and participation to develop an
implementation plan for an RCF DAA service that embraces best practice,
accountability and quality principles.
12. The requirement for the operation for a best practice DAA service should be
included into existing programs for RCF accreditation, recognising that best
practice is not solely the responsibility of the pharmacy providing the DAA services.
Quality Medication Care Group, School of Medicine, University of Queensland265
11.REFERENCES
Adkins RE, Mirza T, Bempong DK et al. Open-dish study to identify labile preparations (2). Pharmacopeial Forum 1999; 12 (5).
Allen LV, Popovich NG, Ansel HC. Ansel's pharmaceutical dosage forms and drug delivery systems. 8th ed. Philadelphia: Lippincott Williams & Wilkins; 2005. 1-738 p.
Al-Zein H, Riad LE, Abd-Elbary A. Effect of packaging and storage on the stability of carbamazepine tablets. Drug Development and Industrial Pharmacy 1999; 25 (2):223-7.
Aulton ME. Pharmaceutics: The science of dosage form design. 2nd ed. Edinburgh: Churchill Livingston; 2002. 1-679 p.
Australian Institute of Health and Welfare (AIHW). Residential aged care in Australia 2003-04. A statistical overview. Canberra: AIHW.; 2005. (Aged Care Statistics Series).
Australian Pharmaceutical Advisory Council. Guidelines for medication management in residential aged care facilities. 3rd ed: Commonwealth Department of Health and Ageing; 2002. 105 p.
Australian Pharmaceutical Advisory Council. Integrated best practice model for medication management in residential aged care facilities. 2nd ed: Commonwealth Department of Health and Aged Care; 2000.
Australian Pharmaceutical Advisory Council. National guidelines to achieve the continuum of quality use of medicines between hospital and community. Canberra: Australian Government Publishing Service; 1998.
Badawy SI, Gawronski AJ, Alvarez FJ. Application of sorption-desorption moisture transfer modelling to the study of chemical stability of a moisture sensitive drug product in different packaging configurations. Int J Pharm 2001; 223 (1-2):1-13.
Barat I, Andreasen F, Damsgaard EM. Drug therapy in the elderly: what doctors believe and patients actually do. Br J Clin Pharmacol 2001; 51 (6):615-22.
Barrera MJ, Ainlay SL. The structure of social support: a conceptual and empirical analysis. J Community Psychol 1983; 11:133-43.
Bayoumi AM. The measurement of contingent valuation for health economics. Pharmacoeconomics 2004; 22 (11):691-700.
Bazire S. An assessment of the "Dosett" compliance aid in psychiatric patients. Br J Pharm Pract 1984; 6 (Oct):316-18.
Bempong DK, Mirza T, Bradby S et al. Open-dish study to identify labile preparations (1). Pharmacopeial Forum 1999a; 12 (5).
Bempong DK, Mirza T, Grady LT et al. Accelerated screening studies to identify labile preparations (1). Pharmacopeial Forum 1999b; 12 (5).
Bempong DK, Mirza T, Grady LT et al. Accelerated screening studies to identify labile preparations (2). Pharmacopeial Forum 1999c; 12 (5).
Benrimoj S, Langford, J., Berry, G., Collins, D., Lauchlan, R., Stewart, K., Aristides, M., & Dobson, M. Economic impact of increased clinical intervention rates in community pharmacy. Pharmacoeconomics 2000; 18 (5):459-68.
Bernsten C, Bjorkman I, Caramona M et al. Improving the well-being of elderly patients via community pharmacy-based provision of pharmaceutical care: a multicentre study in seven European countries. Drugs Aging 18:63-77, 2001.
Blister-strip warning. Pharm J 1996; 256:85.
Bond WS, Hussar DA. Detection methods and strategies for improving medication compliance. Am J Hosp Pharm 1991; 48 (9):1978-88.
Bowling A. Research methods in health: investigating health and health services. Philadelphia, US: Open University Press; 1997.
Brown LH, Krumperman K, Fullagar CJ. Out-of-hospital medication storage temperatures: a review of the literature and directions for the future. Prehosp Emerg Care 2004; 8 (2):200-6.
Quality Medication Care Group, School of Medicine, University of Queensland266
Caplan RD, VanHarrison R, Wellons RV et al. Social support and patient adherence: experimental and survey findings. Anne Arbor, MI: Institute for Social Research.; 1980.
Chen Y, Li Y. A new model for predicting moisture uptake by packaged solid pharmaceuticals. Int J Pharm 2003; 255 (1-2):217-25.
Church C, Smith J. How stable are medicines moved from original packs into compliance aids? Pharm J 2006; 276 (21 January):75-81.
Classen DC, Pestotnik SL, Evans RS et al. Adverse drug events in hospitalized patients. Excess length of stay, extra costs, and attributable mortality. Jama 1997; 277 (4):301-6.
Corlett AJ. Aids to compliance with medication. Brit Med J 1996; 313 (7062):926-9.
Cramer JA. Enhancing patient compliance in the elderly. Role of packaging aids and monitoring. Drugs Aging 1998; 12 (1):7-15.
Dartnell JG, Anderson RP, Chohan V et al. Hospitalisation for adverse events related to drug therapy: incidence, avoidability and costs. Med J Aust 1996; 164 (11):659-62.
Dean DA. Packaging, package evaluation, stability, and shelf-life. Drugs Pharmaceut Sci 2000; 107:483-513.
Dehejia RH, Wahba S. Propensity score matching methods for non-experimental causal studies.: National Bureau of Economic Research, Inc; 1998. (NBER Working Papers).
Department of Health and Ageing Aged and Community Care Division. Standards for Aged Care Facilities - Standard 2 Health and Personal Care. 2004 [http://www.health.gov.au/internet/wcms/publishing.nsf/Content/ageing-standard-facility-std2.htm cited 10/5/2006].
Devlin N, Hansen P, Herbison P. Variations in self-reported health status: results from a New Zealand survey. N Z Med J 2000; 113 (1123):517-20.
Drummond M, O'Brian B, Stoddart G et al. Methods for the economic evaluation of health care programmes. 2nd edition. ed. New York: Oxford University Press Inc; 1997.
Florence AT, Attwood D. Physicochemical principles of pharmacy. 3rd ed. Hampshire: Palgrave; 1998. 1-564 p.
George B, Silcock J. Economic evaluation of pharmacy services - fact or fiction? Pharm World Sci 1999; 21 (4):147-51.
Glass BD, Novák C, Brown ME. The thermal and photostability of solid pharmaceuticals. Journal of Thermal Analysis and Calorimetry 2004; 77 (3):1013-36.
Green CF, McCloskey S. UK Survey of the provision of multicompartment compliance aids and medicines reminder charts on discharge from hospital. Int J Pharm Prac 2005; 13 (1):85-9.
Haynes RB, McDonald H, Garg AX et al. Interventions for helping patients to follow prescriptions for medications (Cochrane Review). Cochrane Database Syst Rev 2002; (2):Cd000011.
Ientile C, Lewis G, Stokes J et al. Effectiveness and cost-effectiveness of dose administration aids (DAAs) Final Report. Brisbane: Quality Medication Care Pty Ltd in conjunction with the Therapeutics Research Unit, University of Queensland.; 2004. 450 p.
Jeffcoate J, Chappell C, Feindt S. Best practice of SME adoption of e-commerce. Benchmarking: An international journal. 2002; 9 (2):122-32.
Kommanaboyina B, Rhodes CT. Trends in Stability Testing, with Emphasis on Stability During Distribution and Storage. Drug Dev Ind Pharm 1999; 25 (7):857 - 68.
Kononenko I. Machine Learning for medical diagnosis: history, state of the art and perspective. Artif Intell Med 2001; 23:89-109.
Larkin C, Murray R. Assisting Aboriginal patients with medication management. Aust Prescr 2005; 28 (5):123-5.
Lavrac N. Selected techniques for data mining in medicine. Artif Intell Med 1999; 16:3-23.
Levy RL. Social support and compliance: a selective review and critique of treatment integrity and outcome measurement. Soc Sci Med 1983; 17:1329-38.
Quality Medication Care Group, School of Medicine, University of Queensland267
Liu Z. Expected length of stay in nursing homes and hostels over a lifetime in Australia. Canberra: Australian Institue of Health and Welfare Report No.: Welfare Division Working Paper No. 23.
Lorenc L, Branthwaite A. Are older adults less compliant with prescribed medication than younger adults? Br J Clin Psychol 1993; 32 (Pt 4):485-92.
MacLaughlin EJ, Raehl CL, Treadway AK et al. Assessing medication adherence in the elderly - Which tools to use in clinical practice? Drugs & Aging 2005; 22 (3):231-55.
McColl E, Jacoby A, Thomas L et al. Design and use of questionnaires: a review of best practice applicable to surveys of health service staff and patients. Health Technology Assessment 2001; 5 (31).
Medwick T, Bailey LC. A procedure for conducting open-dish studies. Pharmacopeial Forum 1998a; 24 (3):6315-16.
Medwick T, Okeke C, Grady LT. A scientific basis for beyond-use dating. Pharmacopeial Forum 1998b; 24 (1):5643-44.
Mitchell JL. CQI: changing systems to improve quality. American Pharmacy 1992; 32 (Sep):47-53.
Morrison A, Wertheimer AI. Compilation of quantitative overviews of studies of adherence. Drug Information J 2004; 38 (2):197-210.
Naranjo CA, Busto U, Sellers EM et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther 1981; 30 (2):239-45.
Naughton MJ, Wiklund I. A critical review of dimension-specific measures of health-related quality of life in cross-cultural research. Qual Life Res 1993; 2 (6):397-432.
Overman ES, Boyd KJ. Best practice research and postbureaucratic reform. J PART 1994; 4:67-83.
Peterson G, Naughton M, Fitzmaurice K. The drugs could be OK, but do not forget problems with dosage forms. Aust Pharm 2003; 22 (3):212-15.
Pharmaceutical Society of Australia. Australian Pharmaceutical Formulary and Handbook. 19th ed. Canberra: Pharmaceutical Society of Australia; 2004. 1-509 p.
Pharmaceutical Society of Australia. Dose Administration Aids Guidelines. Canberra: Pharmaceutical Society of Australia, July 1999.
Pharmaceutical Society of Australia. Professional Practice Standards [version 2]. Canberra: Pharmaceutical Society of Australia; 2002. 136 p.
Pharmaceutical Society of Australia. Professional Practice Standards [version 3]. Canberra: Pharmaceutical Society of Australia; 2006. 147 p.
Pharmacists develop compliance aid for the elderly. Pharm J 1988; 241 (Oct 8):474.
Pillans PI, Roberts MS. Overprescribing: have we made any progress? Aust N Z J Med 1999; 29 (4):485-6.
Reamer JT, Grady LT. Moisture permeation of newer unit dose repackaging materials. Am J Hosp Pharm 1978; 35 (Jul):787-93.
Report of the moisture permeability testing of monitored dosage systems. Pharm J 1994; 252:18-9.
Richards G, Rayward-Smith VJ, Sonksen PH et al. Data Mining for indicators of early mortality in a database of clinical records. Artif Intell Med 2001; 22:215-31.
Ryan M. A comparison of stated preference methods for estimating monetary values. Health Economics 2004; 13:291-6.
Saville DJ. Influence of storage on in-vitro release of ibuprofen from sugar coated tablets. Int J Pharm 2001; 224 (1-2):39-49.
School of Pharmacy and Medical Sciences at University of South Australia, Australian Association of Consultant Pharmacy. Dose Administration Aids. Adelaide: School of Pharmacy and Medical Sciences at University of South Australia, 31 October 1997.
Quality Medication Care Group, School of Medicine, University of Queensland268
Schumock G, Butler, m., Meek, P., Vermeulen, L., Arondekar, M., & Bauman, J. Evidence of the economic benefits of clinical pharmacy services: 1996-2000. Pharmacotherapy 2003; 23 (1):113-32.
Shiell AG, L. If the price is right: vagueness and values clarification in contingent valuation. Health Economics 2003; 12:909-19.
Simmons D, Upjohn M, Gamble GD. Can medication packaging improve glycaemic control and blood pressure in Type 2 diabetes? Diabetes Care 2000; 23:153-56.
Smith B, O'Malley S, Lawson J. The costs and experiences of caring for sick and disabled geriatric patients-Australian observations. Aust J Public Health 1993; 17 (2):131-4.
Sorensen L, Stokes, J., Purdie, D., Woodward, M., Elliott, R., & Roberts, M. Medication reviews in the community: results of a randomized, controlled effectiveness trial. British Journal of Clinical Pharmacology 2004; 58 (6):648-64.
Sturgess I, McElnay, J., Hughes, C., & Crealey, G. Community pharmacy based provision of pharmaceutical care to older patients. Pharm World Sci 2003; 25 (5):218-26.
Svarstad B, Chewning B, Sleath B et al. The brief medication questionnaire: A tool for screening patient adherence and barriers to adherence. Patient Educ Couns 1999; 37:113-24.
Therapeutic Goods Administration. Australian Code of Good Manufacturing Practice for Medicinal Products (16 August 2002). Canberra: Therapeutic Goods Administration.; 2002.
Therapeutic Goods Administration. Australian Regulatory Guidelines for Prescription Medicines. Australian Government Department of Health and Ageing, June [http://www.tga.gov.au/pmeds/argpm.htm].
Tønnesen HH. Formulation and stability testing of photolabile drugs. Int J Pharm 2001; 225 (1-2):1-14.
United States Pharmacopeial Convention. USP 28 - NF 23 supplement 2. Supplement to 28th rev. of the U.S. Pharmacopeia & 23rd ed. of The National Formulary. Rockville, Maryland: United States Pharmacopeial Convention, 2005; 2005.
USP <661>. US Pharmacopeia. National Formulary (USP-28/NF-23). <661> Containers. Rockville, MD: United States Pharmacopeial Convention; 2005.
USP <671>. US Pharmacopeia. National Formulary (USP-28/NF-23). <671> Containers - Permeation. Rockville, MD: United States Pharmacopeial Convention; 2005.
USP <1146>. US Pharmacopeia. National Formulary (USP-28/NF-23). <1146> Packaging Practice - Repackaging a single solid oral drug product into a unit-dose container. Rockville, MD: United States Pharmacopeial Convention; 2005.
USP <1150>. US Pharmacopeia. National Formulary (USP-28/NF-23). <1150> Pharmaceutical Stability. Rockville, MD: United States Pharmacopeial Convention; 2005.
USP <1178>. US Pharmacopeia. National Formulary (USP-28/NF-23) Second Supplement. <1178> Good Repackaging Practices. Rockville, MD: United States Pharmacopeial Convention; 2005.
USP <1191>. US Pharmacopeia. National Formulary (USP-28/NF-23). <1191> Stability considerations in dispensing practice. Rockville, MD: United States Pharmacopeial Convention; 2005.
van Eijken M, Tsang S, Wensing M et al. Interventions to improve medication compliance in older patients living in the community. A systematic review of the literature. Drugs Aging 2003; 20 (3):229-40.
Walker R. Stability of medicinal products in compliance devices. Pharm J 1992; 25:124-6.
White TJ, Arakelian, A., & Rho, J. Counting the costs of drug-related adverse events. Pharmacoeconomics 1999; 15 (5):445-58.
Whynes DK, Wolstenholme, J.L. & Frew, E. Evidence of range bias in contingent valuation payment scales. Health Economics 2004; 13:183-90.
Zinn C. Australian Government subsidises long term care by up to 22,000 pounds a year. Br Med J 2002; 324:1543.
Quality Medication Care Group, School of Medicine, University of Queensland269
APPENDIX A: ETHICAL APPROVALS
PAH Ethical Approval
University of Queensland Ethical Approval
Quality Medication Care Group, School of Medicine, University of Queensland270
Quality Medication Care Group, School of Medicine, University of Queensland271
Quality Medication Care Group, School of Medicine, University of Queensland272
Quality Medication Care Group, School of Medicine, University of Queensland273
Quality Medication Care Group, School of Medicine, University of Queensland274
APPENDIX B: DAA EXPERT PANEL MEETING
Consultation and Feedback Meeting Agenda
Consultation and Feedback Meeting materials
Consultation and Feedback Meeting Minutes
CONSULTATION & FEEDBACK ON THE DEVELOPMENT OF
BEST PRACTICE FOR DAAS
Meeting to be held on 4th March, 2005, 11.30am – 12.45pm
VENUE: Meeting Room 4, First Floor
Gold Coast Convention and Exhibition Centre (Map attached)
Lunch will be provided
Proposed Agenda
11.30 am Welcome, Introductions & Objectives
11.40 am Results of Phase 1 & 2 Rationale for Phase 3
11.50 am Preliminary Drafts of
Best practice in a Community setting
Best practice in an RCF setting
12.10 pm Our research plan
12.15 pm Feedback & Discussion
12.40 pm Meeting close
Quality Medication Care Group, School of Medicine, University of Queensland275
Meeting Materials - Consultation & Feedback on the development of Best
Practice for DAA
The results of Phases 1 and 2 indicate that DAAs are effective in helping community patients
and RCFs to manage medications, and that the stakeholders value DAAs, as evidenced by their
high levels of satisfaction. While the economic evaluations indicated that DAAs were a more
cost efficient system for medication provision for RCFs compared with original packs, the use of
DAAs in the community setting was cost-ineffective. However, there were a number of
opportunities for further research identified with respect to the economic evaluation
methodology and in the development of best practice guidelines to maximise the efficiency and
effectiveness of DAA provision.
We see this Phase as an important opportunity to develop:
(3) Best practice guidelines and tools to facilitate improvements in the way DAAs are used
in the RCF and community settings.
(4) A more sophisticated methodology to re-examine the cost-effectiveness of DAAs in the
community setting by measuring and valuing the benefits to the health care system utilising
additional data on health service use.
Best Practice for DAA provision
Aims and objectives
In Phase 3, we propose to develop best practice guidelines for DAA use in the community and
RCF settings by expanding on the recommendations in Phase 2, and identifying quality
improvement methods in consultation with key stakeholders.
Rationale
Presently, while there are a range of guidelines and standards that relate to DAA use, there are
no specific best practice guidelines for DAA provision. Best practice is defined as a technique or
methodology that, through experience and research, has been proven to reliably lead to a
desired result. The development of best practice guidelines, through a synthesis of Phase 1
and 2 findings and the experience of key stakeholders, will provide strategies and tools to
maximise the efficiency of DAA provision and the probability of positive patient outcomes. The
results of Phase 1 and 2 indicate that there are a number of issues in current practice that could
be improved upon. The quality of DAA services currently provided by pharmacies is
inconsistent and does not conform to all Pharmaceutical Society of Australia (PSA) guidelines
and PSA/Quality Care Pharmacy (QCP) standards. In addition, there were a number of
problems identified in the literature and through data collection that are not adequately
addressed by existing guidelines.
The underlying issues with DAA provision and use are essentially the same across the two
settings (community and RCF) but there are differences related to who the key stakeholders are
and what the priorities are. Issues common for both settings include:
DAA users are not adequately informed about what is involved or trained in how to
maximise the benefits of DAA use.
The frequency and quality of communication between stakeholders (pharmacy, patients,
RCF staff and GPs) regarding medication changes and errors is inadequate.
DAA packing sessions did not optimally utilise the skills and experience of pharmacists.
The processes involved in packing and checking were not clearly identified in the pharmacy.
During packing sessions, pharmacists and pharmacy staff had limited resources to which to
refer regarding the effect of DAAs on the stability of medications.
Community pharmacy bears the burden of medication management and professional
responsibility without adequate remuneration.
Specific issues with respect to DAA use in RCFs include:
Quality Medication Care Group, School of Medicine, University of Queensland276
The observed rate of errors in DAAs.
The accuracy of medication administration and the professional de-skilling of registered
nurses (RNs).
The lack of standard operating procedures for effective handling of medication
administration when DAAs are used in RCFs.
Specific issues for DAA use in the community are:
Community patient loss of ownership with respect to medication management, as
evidenced by the lower levels of medication knowledge and high levels of dependence on
the pharmacy.
DAAs are highly relied upon in the RCF and community settings, and are effective provided that
the limitations of supplying and using DAAs are addressed. The development of best practice
guidelines will resolve some of the effectiveness and practice issues surrounding DAA use in
the community and RCF settings.
Methods
In Phase 3 we propose to use a variety of qualitative and quantitative techniques to develop and
evaluate best practice guidelines for the RCF and community settings with respect to the areas
identified above. We intend to develop on the preliminary best practice guidelines identified in
the Phase 1 and 2 using the methodology presented in Figure 1. Best practice requires
collaboration between stakeholders, and the synthesis of research and experience. In the
development of guidelines, we will use the Consensus Development Panel technique, as
described by Bowling (Bowling 1997), populating a series of panels with key stakeholders from
our target populations.
The following stakeholders are the proposed participants:
Doctors that prescribe medications for DAA patients in RCFs and the community.
Pharmacists that supply DAAs to RCFs and community patients.
Nursing staff from RCFs using DAAs.
DAA manufacturers and suppliers.
Community nurses with patients who use DAAs.
Community patients and carers who use DAAs.
Therapeutic Goods Administration (TGA).
It is well recognised that the TGA has a pre-eminent role in the approval of manufacturer goods
and packing and have had a long track record of improving outcomes in terms of the quality use
of medicines mission of the Australian government. Discussions have been held with officers of
TGA (senior officer Dr John McEwen) and enthusiastic support has been offered for further
consultations, should the tender be successful.
To assess the feasibility of implementing the best practice guidelines devised by the expert
panel, we propose to develop survey tools designed to evaluate the anticipated impact of these
guidelines on the following areas:
Impact of adhering to best practice guidelines with respect to changes in current practice.
Changes in workloads and associated costs.
Perceived outcomes with respect to changes in the quality of DAA service provision.
Surveys will be circulated among samples of doctors, pharmacists, RCF staff, community
nurses, community patients and carers.
Quality Medication Care Group, School of Medicine, University of Queensland277
RESEARCH EXPERIENCE+
Identification of issues with the supply and use of DAAs
Phases 1 & 2
Exploration of issues and identification of practice limitations
and possible solutions
Focus groups
Development of preliminary best practice guidelines
Synthesis of findings
Refining best practice guidelines
Consensus panel techniques with
stakeholders
Development of tools to facilitate the implementation of best
practice
i.e. packing templates, contracts
Evaluate the feasibility and impact of implementing best practice
guidelines
Survey stakeholders
Analyse and report on the outcomes and processes
Appendix Figure 1 Proposed methodology for the development of best practice.
Economic evaluation of the benefits of DAAs use in the community
Aims and objectives
In Phase 3, we propose to re-examine the cost-effectiveness of DAAs in the community setting
by measuring and valuing the benefits to the health care system utilising additional data on
health service use.
Rationale
The results of Phase 1 and 2 indicate that DAAs are effective in helping community patients to
manage medications and that community patients value DAAs, as evidenced by their high
levels of satisfaction and willingness to pay for the service. Despite these findings, the economic
evaluations indicated that DAAs are not cost-effective in the community setting as the costs of
DAA provision exceeded the benefits of DAAs. This is largely because the provision of DAAs
by pharmacy is a labour-intensive and costly exercise. Sensitivity analysis, however, suggests
that variations in the magnitude of the benefits and/or the efficiency of the service provision may
provide an alternative view on cost-effectiveness.
In Phase 2, the measurement of costs of DAA provision was based on a detailed content
analysis of work flow observations from 83 pharmacies providing DAAs, and interviews with 353
community patient to ascertain the probability and quantity of the DAA related services provided
by the pharmacy. Due to time and resource limitations, it was not possible to collect the same
level of data to measure and evaluate the benefits of DAAs with respect to healthcare savings.
Instead, the benefits of DAAs with respect to healthcare resource savings were identified and
measured using literature values imputed into a decision analytic model, rather than actual
service use. The literature values indicated the probability of service use and the cost of service
use resulting from adverse drug reaction (ADR) when DAAs or OPs were used by community
patients. The rate of adverse drug reactions was based on the community patients’ reports.
Quality Medication Care Group, School of Medicine, University of Queensland278
While this technique was valuable for exploring the potential savings to the healthcare system
from using DAAs, further research is required to establish the true magnitude of these savings.
The technique utilised in Phase 2 may underestimate the magnitude of the savings to the
healthcare system in the following ways:
(1) the literature does not include probabilities or values for the full range of services used (i.e.
does not include the purchase of additional medications, specialists appointments etc).
(2) DAAs may actually have a wider impact than just reducing rates of ADRs (i.e. may reduce
service costs by preventing unnecessary alterations to therapy due to non-compliance).
(3) Benefits of DAAs may not be observed in a one year timeframe.
A great deal of uncertainty remains regarding the effect of DAA use on the costs of medicines
and the cost of health service resource use. The cost of medications was not included in the
economic evaluation in Phase 2 due to uncertainty regarding the impact of DAAs on medication
use. It was found that DAA patients reported being less likely to run out of medications and less
likely to forget to take their medications, than non-DAA patients. In effect, this means that over
the course of a year a DAA patient may require more prescriptions filled and cost more to the
government through PBS. On the other hand, the evidence suggests that DAA patients are
likely to visit their GP less often than non-DAA patients (possibly due to the pharmacy
requesting repeat prescriptions on the patients behalf instead of the patient visiting the doctor).
This would result in DAA patients placing less of a burden on the healthcare system. From an
economics viewpoint, it is beneficial to take a stochastic9 approach to assess the overall effects
of DAA on health service resource use.
The inclusion of the full range of health service use over a longer timeframe may alter the cost-
benefit ratio such that the use of DAAs in the community proves to be cost-effective. We
propose to re-evaluate the cost-effectiveness of DAAs in the community through the collection
of retrospective service use data from the community patients that participated in Phase 2.
Methods
Data will be collected through interviews with community patients and by requesting four years
of health service expenditure for those patients. The Health Insurance Commission (HIC) is a
potential source of additional data on health services use (subsidised under the Medical
Benefits Scheme (MBS)) and medications dispensed (subsidised under the Pharmaceutical
Benefits Scheme (PBS)) for the sample of community patients recruited for Phase 2. As a
condition of participation in Phase 2, community patients consented to the release of their
records: of services provided between 1 January 2000 and 30 June 2004. We intend to request
these records from HIC and match treatment data with DAA use status and demographics. This
data will be aggregated to produce total costs of service use and compared for community
patients using DAAs and original packs after adjusting for potential covariates of health service
use (i.e. number of medicines used, age and health status).
Additional service use including community nursing, respite and long term care will be collected
through telephone interviews with the sample community patients and/or carers that participated
in Phase 2. This is also an opportunity to follow-up participants and to collect patient –specific
data on costs and outcomes for economic analysis.
Figure 2 shows how the HIC health service data and the patient-specific outcome data will be
combined to allow for comparison between community patients using DAAs and community
patients not using DAAs. The collection of this data allows for stochastic analysis where both
the costs and effects are determined from data sampled from the same patients. In addition,
9 Stochastic data: where both costs and effects are determined from data sampled from the
same patients in a study.
Quality Medication Care Group, School of Medicine, University of Queensland279
formal statistical testing can be performed on any observed differences in costs and risk
adjustment strategies such as cohort selection can be utilised to control for covariates of
healthcare costs.
Cost-benefits analysis will be conducted from a societal perspective to evaluate the full range of
costs associated with DAA use. Cohorts of community patients who benefit most from DAA (in
terms of savings in healthcare costs) may be identified and sensitivity analysis and threshold
analysis will be conducted to assess the precision around the costs and effects of DAA use.
No DAA
Respite care
Respite care
RCF admittance
RCF admittance
GP services
GP services (MBS)
Pathology (MBS)
RPBS drug costs
Pathology (MBS)
PBS drug costs
Death
%
%
Total cost
%
%
%
%
%
%
%
%
%
% Community Nursing
% Hospitalisation
Total cost
% Death Number
% Hospitalisation
Community Nursing%
DAA
Matched sample of community
patients
Number
Appendix Figure 2 Evaluation plan for the comparison of community patients using
DAAs or original packs on a combination of health service resource
use collected through HIC and through interview with community
patients.
PRELIMINARY BEST PRACTICE
Best practice in the community setting
Current tasks and activities performed in the provision of a DAA service to community patients
are shown in the following table. The initial set-up of a DAA service requires the involvement of
all stakeholders. In contrast, once the DAA service is established, the majority of tasks can be
conducted by the pharmacist with minimal input from these sources. However, if pharmacies
are no longer having regular contact with GPs and community nurses, this could potentially be
detrimental to the quality of the DAA service and the patients’ therapeutic outcomes.
Quality Medication Care Group, School of Medicine, University of Queensland280
Appendix Table 1 Tasks and activities involved in the provision of DAAs in the
community
Who is involved? Tasks and activities When Pharm Dr Pt Carer
Initial Set-up for new community patients
Identify potential DAA users Any time Is the patient willing to use a DAA? Any time Explain what is involved for the patient? Prior to
starting DAA Arrange for the patients’ medicines and prescriptions to be stored at the pharmacy.
Day 1 of DAA service
Determine which medicines should be packed and how they should be packed (develop medication profile).
Day 1
Pack the DAA (generally 1 week supply of medicines)
Day 1
Attach labels to the DAA with the patients, pharmacy name, days of the week, dose time and list of packed and unpacked medicines.
Day 1
Check that the pack is correct Day 1 Arrange to deliver the pack to the patient or have the pack collected
Day 1
Advise the patient to inform the pharmacy of any medication changes
Day 1
Administer medicines from DAA following the Day & Time prompts
Day 1 - Day 7
Managing continuity of supply
If no changes to medication regimen, pack another weeks supply as per previous week
Day 6
If medication regimen changes, change medication profile to reflect this.
As it happens
Where original pack supply is insufficient to fill next weeks supply, dispense new supply as per normal.
Day 6
Where there is no repeat prescription avail-able to dispense required medicines, request a repeat prescription from the doctor or advise the patient to obtain the repeat prescription
Obtainrepeat prior to Day 6
Arrange to deliver the pack to the patient or have the pack collected before previous supply runs out
Day 6
Collect previous DAA when the new DAA is delivered or collected.
Day 6
DAA packing and individual prescriptions charged to the patients account
Day 6
Key: Pharm = Pharmacists, Dr = Doctor, Pt = Patient, Carer = Carer, patients’ family and/or community
nurses (if involved).
Many of these tasks are covered by existing standards or guidelines whoever, the previous
results regarding evidence of compliance with standards, indicate the need to develop best
practice guidelines for the implementation of DAA provision to new patients in the community
setting. A synthesis of the above findings was used to develop the following implementation
strategy with the goal of resolving some of the effectiveness and practice issues surrounding
DAAs. There are three key aspects to this model. First, a tripartisan agreement between the
pharmacy, patient/carer and GPs/specialists with the goal of formalising the obligations and
expectations of the parties. Second, the model requires the development of a template by the
pharmacist, in conjunction with other stakeholders, for the packing and checking of patients
DAAs. This template will reflect the patients’ optimal medication schedule and aim to maximise
Quality Medication Care Group, School of Medicine, University of Queensland281
therapeutic benefits and minimise risks. The template would be approved by all stakeholders
and subject to regular update and six-monthly review. Third, the patient or carer is given the
responsibility of ensuring the template reflects their current medication status. This strategy
encourages the patient to take ownership of their medication management and is a valuable
resource for enhancing patient medication knowledge. Where the patient is unable to perform
this role a carer or family member will act as a proxy. Copies of the templates will also be held
by the patient’s pharmacist and doctor(s). The Best Practice implementation model is
presented in Figure 3.
New patient to start DAA(need already assessed)
Is it 6 months since last
formalcheck?
Yes
No
Pa
tien
t-h
eld
te
mp
late
kep
t u
p-t
o-d
ate
by
pa
tien
t, c
are
r &
GP
Inclu
din
g m
edic
ations c
hanges n
ot needin
g p
rescription
DAA packed by staff using template
Pharmacist checks DAA using template
DAA sent to patient
Pharmacist check/update template before next packing
Patient-held template/ medication record
Tripartisan Agreement*• Patient/carer• GP (+/- specialists)• Pharmacy
Formalise service to be delivered, expectations & obligations
*see break-out box for content
Full review/ renewal of template
• Patient, GP & pharmacy
Template for medication packing• Shows medication regimen & preferred
time of day for each dose
Reflect• Patient habits/preferences• GP preferences• Current medication & optimal schedule• Type of pack & packing interval• Check medications suitable for packing• Other constraints e.g vision impaired
Pre
scriptions
dis
pensed
Template approved by all
Template copy to GPNew patient to start DAA(need already assessed)
Is it 6 months since last
formalcheck?
Is it 6 months since last
formalcheck?
Yes
No
Pa
tien
t-h
eld
te
mp
late
kep
t u
p-t
o-d
ate
by
pa
tien
t, c
are
r &
GP
Inclu
din
g m
edic
ations c
hanges n
ot needin
g p
rescription
DAA packed by staff using template
Pharmacist checks DAA using template
DAA sent to patient
Pharmacist check/update template before next packing
Patient-held template/ medication record
Tripartisan Agreement*• Patient/carer• GP (+/- specialists)• Pharmacy
Formalise service to be delivered, expectations & obligations
*see break-out box for content
Full review/ renewal of template
• Patient, GP & pharmacy
Template for medication packing• Shows medication regimen & preferred
time of day for each dose
Reflect• Patient habits/preferences• GP preferences• Current medication & optimal schedule• Type of pack & packing interval• Check medications suitable for packing• Other constraints e.g vision impaired
Pre
scriptions
dis
pensed
Template approved by all
Template copy to GP
Break out box
Issues to address in agreement Obligations of parties - examples In simple language
Explanation of how service expected to work
Patient consent to service and necessary information
sharing between GP & pharmacy
Patient, GP and pharmacy agreement to service
obligations
Agreed cost of service including any GP costs as
negotiated
Billing/account aspects
Where prescriptions and original packs physically
stored
Is pack to be collected or delivered (address
timeliness & exception procedures)
How medication regimen changes to be handled
Duration of agreement & an understanding that the
situation to be reviewed every 6 months
Pharmacy to prepare packs at agreed interval
(frequency & period in advance of distribution)
Any additional support services e.g. education, Home
Medicines Review the pharmacy agrees to provide
GPs responsibilities for owing prescriptions &
prescription continuity (e.g. writing repeat
prescriptions without patient consultation) including
timeliness of prescription receipt
Pharmacy processes negotiated to fit in with GP
practice
Address GP expectations for payment & who will pay
if any cost
Patient & GP to give timely notification of pharmacy
of any medication changes including those that do
not generate a prescription e.g. ceasing a medication
Patient/carer & GP to maintain patient held template
Appendix Figure 3 Best Practice model for the operation of a DAA service for
community patients
Quality Medication Care Group, School of Medicine, University of Queensland282
The benefit of this Best Practice model is that it addresses the problems reported by
stakeholders with existing processes for DAA provision. This model proposes the following
solutions to problems reported with existing processes:
1. The tripartisan agreement addresses the issue of poor provision of information and training
to DAA stakeholders. The advantage of this formal agreement is that it ensures the
patient, GP and pharmacist are all aware of their obligations as part of the DAA provision
process (see breakout box, Appendix Figure 3). These obligations may include the
provision of information to new patients and ensuring that patients receive adequate
training in the use of their DAA.
2. The patient medication template addresses a variety of issues, these include: the accuracy
and adequacy of patients’ current medication profiles; the suitability of patients’
medications for packing in DAAs; and the roles of all parties in the DAA provision process.
With respect to the accuracy and adequacy of patients’ current medication profiles, the
template ensures that the pharmacist has a current and complete list of medications from
which to pack and check the patient’s DAA, including the suitability of each medication for
packing in the DAA. The template acts to alleviate confusion with regards to patients’
dosing times and directions. It also provides a central record for all communication
regarding patients’ current medication status and any medication changes. The process
by which doctors approve the medication template ensures that the patients’ current
regimen is reviewed with the aim of optimising therapeutic benefits and minimising
medication risks.
3. The strategy of having patient held medication templates has a twofold effect. First, this
strategy encourages the patient to take ownership of their medication management by
increasing their responsibility in that process and improving their knowledge of
medications. Second, this ensures that any medication changes, especially those not
requiring a prescription (i.e. ceased medications or changes to medication dosages), are
communicated in a timely and accurate manner, to the pharmacist.
This model is based on the current situation whereby no subsidies are available for the
provision of DAAs. Were funding to become available in the future for DAA provision, the
patient template and/or tripartisan agreement could be incorporated as part of the system of
check and balances that would be necessary to ensure accountability with respect to the quality
and quantity of DAA services offered. An additional benefit is that these documents could also
be used to form part of a program evaluation of any future DAA implementation.
Best practice in the RCF setting
The provision of DAA services to RCF residents requires the collaboration of the following
stakeholders: the RCF staff, the community pharmacy, the residents’ doctors (GP and
specialists), the resident and their family or carer. The minimum set of activities that are
required for DAAs to be provided by pharmacists to RCF residents are shown in the following
table. While the pharmacist is involved in all these tasks, the RCF staff have a fairly limited
involvement and the resident and their families are often overlooked by the current processes.
In the RCF setting, the use of DAAs relies on the collaboration of RCFs and pharmacies.
However, both parties are governed by different and sometimes conflicting regulatory and
business limitations. Best practice guidelines should again maximise effectiveness and
efficiency while addressing the regulatory and business issues. A best practice model should
address the following issues:
The need to redefine professional responsibilities for pharmacy and RCF staff with the goal
of ensuring that the use of DAAs results in a minimisation of packing and medication
administration errors.
The need to implement strategies to ensure that RCF staff maintain and develop their
knowledge of medications when using DAAs; and
Quality Medication Care Group, School of Medicine, University of Queensland283
The need to develop standard operating procedures for situations that arise when DAAs are
used instead of original packs (i.e. when a tablet is lost and no additional supply is
available).
Appendix Table 2 Tasks and activities involved in the provision of DAAs in the RCF
setting
Who is involved? Tasks and activities When Pharm. Dr RCF Res &
Family
Initial Set-up for a new resident
Arrange for the residents’ medicines and prescriptions to be stored at the pharmacy.
Day 1 of DAA service
Determine which medicines should be packed and how they should be packed based on the residents medication chart (develop a medication profile).
Day 1
Who is involved? Tasks and activities When Pharm. Dr RCF Res &
Family
Initial Set-up for a new resident
Pack the DAA (generally one week supply of medicines)
Day 1
Attach labels to the DAA as per RCF standards.
Day 1
Check that the pack is correct Day 1 Deliver the pack to the RCF Day 1
Administer medicines from DAA following the day and time prompts
Day 1 - Day 7
Managing continuity of supply
If no changes to medication regimen, pack another weeks supply as per previous week
Day 6
If medication regimen changes, RCF faxes changes to the pharmacy. Pharmacy changes the medication profile and packs to reflect these changes.
As it happens
Where original pack supply is insufficient to fill next weeks supply, dispense new supply as per normal.
Day 6
Where there is no repeat prescription available to dispense required medicines, request a repeat prescription from the doctor or advise the RCF to obtain the repeat prescription from the doctor
Obtain repeat prior to Day 6
Deliver the pack to the RCF before previous supply runs out
Day 6
Collect previous DAA when the new DAA is delivered.
Day 6
DAA packing and individual prescriptions charged to the residents account (DAA may also be charged to facility)
Day 6
Key: Pharm = Pharmacists, Dr = Doctor, RCF = Residential care facility staff, Res. & family =
Resident and their family.
The standard operating procedures should be based on a best practice model that should
address the following issues:
The need to redefine professional responsibilities for pharmacy and RCF staff with the goal
of ensuring that the use of DAAs results in a minimisation of packing and medication
administration errors.
Ensuring that RCF staff receive adequate training in the use of DAAs.
Quality Medication Care Group, School of Medicine, University of Queensland284
Ensuring that the frequency and quality of communication between pharmacies, patients,
RCFs and GPs regarding medication changes is adequate.
The packing of PRN (as needed) medication for RCF residents.
The need to implement strategies to ensure that RCF staff maintain and develop their
knowledge of medications when using DAAs; and
The need to develop standard operating procedures for situations that arise when DAAs are
used instead of original packs (i.e. when a tablet is lost and no additional supply is
available).
The need to ensure that DAA provision by pharmacy is efficient by developing and
promoting strategies to improve efficiency. These strategies may include:
Reminder notices sent to GPs about owing prescriptions,
Set ordering times/procedures,
Routine delivery procedures, and
Optimal utilisation of the skills and experiences of pharmacists and the appropriate
utilisation of less qualified staff in the DAA packing process.
The following figure contains some preliminary ideas on the best practice model by considering
the flow of information. To increase efficiency, these steps need to be optimised. For
effectiveness, timely and correct medication supply also depend on information flows and the
provision of training and support.
Medication Information flows in RCF
GP
Pharmacy* supplying RCF
New resident
RCF
*assumes main supplying pharmacy but resident has choice to choose separate one
Current drug regimen
Order toadminister drug
Medication chart
Advise of medication
supply system,
obligations &
expectations
Order to supply
Medication chart
Authorising
payment for
supply
PBSprescription
Reminder on
continuity of
supply
RCFmedicatio
n system
including DAAre
fillschedule
System of supply – obligations & expectations
including responsibility for script management,
who notifies pharmacy of changes
Mutual obligations &
expectations
Quality Medication Care Group, School of Medicine, University of Queensland285
[Draft] Minutes of meeting at APP 4 March 2005, 11.30am
Consultation & Feedback on the development of Best Practice for DAAs
Present: Gary Lambrides and Simon James (APHS), Klaus Pertulis (Persocare), Gerard
Stevens and Paul Hannan (Webstercare), John Proper (MPS), Karalyn Huxhagen (Community
Others invitated but not attending: Dianne Grant (Douglas), Lance Emerson (Guild
Secretariat), Kos Sclavos (Guild, Qld Branch)
Meeting commenced 11.45am. The following minutes incorporate the presentation images and summarise the comments from the panel.
Effectiveness & Cost Effectiveness of DAAs
in RCF & Community settings – Phase 3
Consultation & Feedback on the development of
best practice for DAAs
Agenda
Welcome, Introductions & Objectives
Phase 1 & 2 results to inform Phase 3
Preliminary models
Research Plan
Feedback & Discussion
The Panel members were welcomed and introduced to each other. The plan for the meeting was outlined.
This meeting is aimed at seeking the panel’s comments and feedback that will be used to achieve a best practice model.
What is Phase 3 about?
Improving the effectiveness & efficiency of provision of DAAs by pharmacy to RCFs & community patients through best practice
Best practice: a technique/methodology that, through experience & research, has been proven to reliably lead to a desired result
synthesis of Phase 1 and 2 findings
experience of key stakeholders
strategies & tools to maximise DAA provision efficiency & probability of +ve patient outcomes
Phase 3 is about finding a way to improve efficiency and effectiveness of DAA service provision through a best practice approach (see definition). In phases 1 (a literature review) and 2 (study capturing data on effectiveness and cost effectiveness of current services), a number of issues were identified that affected both sides of the cost-effectiveness equation. A best practice model could be developed that is feasible to use in practice, and that gives strategies and tools that can be used to maximise outcomes and minimise cost.
The approach in Phase 3 builds on research such as the findings from the literature review, observations of packing and checking, observations at RCFs, interviews/questionnaires completed with RCF staff, pharmacy staff, patients at home, GPs and community nurses. There have been a number of limitations mentioned. Gerard Stevens identified many of these in his presentation at the AACP workshop on 3 March. Best practice is about finding solutions to these limitations. The model needs to have enough definition so that there is not too much uncertainty in rolling out the model.
Quality Medication Care Group, School of Medicine, University of Queensland286
Phase 3 approachRESEARCH EXPERIENCE+
Identification of issues with the supply and use of DAAs
Phases 1 & 2
Exploration of issues and identification of practice limitations
and possible solutions
Focus groups
Development of preliminary best practice guidelines
Synthesis of findings
Refining best practice guidelines
Consensus panel techniques with
stakeholders
Development of tools to facilitate the implementation of best
practice
i.e. packing templates, contracts
Evaluate the feasibility and impact of implementing best practice
guidelines
Survey stakeholders
Analyse and report on the outcomes and processes
See later
A preliminary best practice model has been developed for community patients. A very early model has been developed for RCFs focusing on the information flows as poor communication was identified as a cause of many problems with systems. The RCF systems are more complicated than community patient systems. An iterative process will be used as well as the development of potential tools e.g. communicat -ion log, tools to monitor packing time or quality. Then drafts will be circulated more widely to the various states as legislation and practices vary from state to state. Plan is to survey peak bodies, RCFs and pharmacies more widely
Any best practice model needs to keep costs down and not be a huge paperwork burden; it needs to be usable, user-friendly, and not cost more to implement. People will want to implement the model because it will have an immediate positive effect. It was noted that there was much variation in what people (pharmacies & RCFs etc) did. The aim of this best practice model would not be to prescribe what had to be done but to give a range of strategies that do reliably lead to the best outcome. Phase 2 showed that in some areas, people (pharmacists) did have enough things to refer to and in others, there were too many. The model has to be flexible enough to account for some variation, such as the variation needed for the different device types.
Evidence to date is that people did not follow guidelines and the team needed to look at conduits to help people use guidelines. It was suggested that collaboration with the PSA team working on new DAA guidelines was desirable. PSA guidelines would then feed into QCPP guidelines as a way of increasing awareness and a roll-out methodology. The researchers felt that if the best practice model was successful in leading to a better, more reliable outcome that pharmacies should be interested in adopting core parts of the model. However, engendering people (pharmacists) to reflect on their practice was most challenging. People must see benefit for their business if they are to adopt the models. The phase 2 data on just what it costs to provide a DAA service may be a motivator for implementation.
The panel felt that the model needs to be in the format of an implementation model and not a guideline. People don’t read guidelines when they are setting up a service but would use an implementation template model (cookbook approach). This should be the goal of the model and from a QCPP viewpoint, it needs to be about implementation.
A question was asked how the proposed guideline would fit into existing guidelines already out there. Guidelines are usually under the auspices of organisations such as APAC. There was no mission or charter for the work but the proposal was in response to a need and there was an obvious opportunity. The model would be expected to be built into community pharmacy procedures like QCPP as the goal of the model was the same as QCPP i.e. to improve the quality of service provision in both the efficiency and effectiveness of the DAA service provided. There was coverage in the various APAC guidelines to give a “head of power” to be developing local professional guidelines that fit with APAC guidelines yet develop concepts further, spelling them out in more detail. There have been guidelines before but “Best Practice” is new. Pharmacists would accept guidelines, guidance/mentoring but “best practice” was another level again. The problem with the existing guidelines is that there are a plethora of them, they are not read and if so, people find it difficult to work out which one to comply with or what they need to do. There needs to be more specific details for implementation; a road map with tools and resources to aid implementation and operation. PSA are currently reviewing DAA guidelines and it has been difficult to find evidence for each recommendation. Phases 1 and 2 (and other DAA studies) could provide some evidence and this phase 3 could marry with the PSA standards which could then be a reference point for pharmacies through QCPP and reference to several APAC guidelines. There is a need for more research/evidence for example about the stability of drugs in the packs. Where did the 6 week statement come from? The issue of stability information is part of Phase 3 best practice. [Non-minute note: Karalyn has provided a contact with the PSA team].
Quality Medication Care Group, School of Medicine, University of Queensland287
It is envisaged that Phase 3 will produce a best practice model with a systems focus rather than guidelines.
The panel went on to discuss whether the goal was “best” or “better” practice since best practice implies the ultimate model but best practice changes over time. It was decided that “best practice” should be retained as it encapsulated the approach as defines in Slide 2.
Objectives of today’s session –
our perspective
You are our DAA ‘experts’
Range of backgrounds & perspectives
Gain a shared understanding of findings & issues to date
What we are asking of you:
Support for project by contributing ideas, information & contacts
Feedback on model development
Aid getting support of other people we should talk to
The objectives of the meeting were outlined. The researchers sought input on others who should be contacted. A pharmacist DAA expert in Western Australia was identified for consultation.
The findings on cost-effectiveness were described together with planned refinements in the community model to be part of Phase 3.
Phase 2 Cost effectiveness in community
Cost-benefit ratio components & results based on 30 community customers using pharmacy supplied DAAs
Cost to pharmacy
Providing DAAs to 30
community patients/ year
=Cost DAA-Cost OP
=-$27,487
-
Cost-savings to Government
DAAs preventing 15% of ADRs
= Healthcare OP-Healthcare DAA
=$15,316
+Benefits to customers
Measured by willingness to pay (WTP)
=$5.25 x 30 x 52
=$8,190
= -$3,982
Phase 2 Cost effectiveness in RCFs
Cost-minimisation results: overall savings for DAAs and a shift in costs between the pharmacy and RCF
RCF Costs
$236,724
Pharmacy Costs
$175,704
43% of total cost
RCF Costs
$388,609
Pharmacy Costs
$109,127
22% of total cost
Total $497,736 Total $412,428
Cost Shift
OPs DAAs
Among staff of RCFs, there was an attitude about paying pharmacies for DAAs that was at odds with the value placed on the service. There was a perception that pharmacists make enough income from the prescriptions. Preliminary modelling suggests that this is not the case and that many pharmacies provide the service at a loss. The panel felt that this information might be valuable in changing behaviour of pharmacies (with respect to charging) and the attitudes of RCF staff. More work and consultation is required on the break-even point.
RCFs do depend on the pharmacy service and often abdicated medication management responsibility to the pharmacy. This seems to be a wide spread view even among assessorswho indicated (through preliminary discussions) that they did not feel the need for more guidelines and that provided the RCF “had a really good pharmacy looking after you” there was no problem meeting the medication management standards.
Assessors needed facilities to demonstrate that their system, whatever it is, is safe, and that current standards were general to allow facilities to choose their own system. Part of the best practice project is to raise awareness among RCFs.
The issues to be addressed in the best practice model (see following 4 slides) were discussed. DAAs are a service not just a device. Some community patients got DAAs for convenience but did not pay, blurring the cost-effectiveness of the service. Existing PSA guidelines refer to many things that are part of the service but there is a need for explicit implementation models e.g. helping pharmacies deal with issues such as deliveries, what to do when the pharmacy delivers or starting a new person (the initial set-up as an opportunity for review) and continuity of supply. There are risks to community DAA users of poor information –
Quality Medication Care Group, School of Medicine, University of Queensland288
25% never see the pharmacy and the risk that if the pharmacy does prescription management, the patient might not see their GP. The panel mentioned the link to home medicines review (HMR) where a recommendation to start a DAA was often an outcome. It will be important that the best practice model links to other programs like HMR (which might also make both programs more financially viable from a pharmacy viewpoint). Conducting an HMR prior to starting a DAA would also provide an opportunity for the patient to be assessed and trained on the appropriate use of the device and how the service would operate, so increasing the potential effectiveness of the service.
Patient assessment for DAAs should also include assessment by their GP or the community nurses. The initial draft community model does not cover assessment but starts once the patient is identified. Assessment could be included. A trial of teaching pharmacists to target patients for HMR and whether a DAA was required was described as a suggested input into the model.
Phase 2: Issues to be addressed
Targeting of DAA community recipients & service components
DAA users are not adequately informed about what is involved or trained in how to maximise the benefits of DAA use
CPs not trained & expectations not clear
Observed rate of errors in DAAs
Accuracy of medication administration
Lack of standard operating procedures for effective medication administration when DAAs used in RCFs
Issues from Phase 2
Deskilling
Loss of medication knowledge (CPs)
RCF staff deskilled & reliance on pack not chart
Pack issues wrt identifying medicines
Frequency & quality of communication between stakeholders (pharmacy, patients, RCF staff and GPs) about medication changes & errors is inadequate
Regulatory, business pressures not appreciated
Address relationship issues (GPs, RCF)
Phase 2 Issues
DAA packing sessions did not optimally utilise skills & experience of pharmacists
Processes involved in packing & checking were not clearly identified in pharmacy
Need systems for monitoring error trends
During packing sessions, pharmacists & staff had limited resources about medication stability in DAAs
Issues Phase 2
Community pharmacy bears burden of medication management & professional responsibility without adequate remuneration
Procedures to increase efficiency
• Handling changes, continuity of supply, ordering, delivery etc
The nature of errors and the error detection methodology of phase 2 was discussed (comparison of chart and pack). Errors were a mixture of pharmacy and nurse error in RCFs – most did not reach the patient. The error rate among community patients (either DAA or original pack users) was not measured as observers did not know what the
patient should have been taking, although qualitative information about poor pack use was available. The baseline error rate by community patients should be the comparator for errors associated with DAAs. A key performance indicator of 0% errors for DAA users was unrealistic.
The panel felt that it was important to recognise that in some cases, especially in hostels or independent living areas, that the DAA was correct and the chart was wrong. Strategies to minimise errors needed to be included in the model. These strategies are also needed for the administration point; there are no standard operating procedures (SOP) for staff, for example, who drop a dose – what do you do? Use tomorrow’s dose? Procedures are needed for what to do when things go wrong, so that effectiveness and efficiency can be maintained. One panel member mentioned recent NSW Pharmacy Board bulletin about checking DAAs including checking the label on the original pack dispensed. A list of things that should be checked is not explicit in standards.
The panel then went on to discuss system factors that were not the pharmacy’s responsibility but reflected on the pharmacy e.g. maintenance of current resident photos. One RCF organisation
Quality Medication Care Group, School of Medicine, University of Queensland289
expected the pharmacy to organise the Therapeutic Drug Committee meetings (required under APAC guidelines).
Many problems with the system were due to poor communication especially of expectations and obligations, e.g. the doctor’s responsibility, expectations of responsiveness to changes. Two draft models were presented and feedback and comment was invited from panel members (see following 3 slides). The models need to address establishing the nature of the service and agreeing on expectations and obligations up front. Strategies were required to maintaincurrency of the medication profile (especially where a change was not associated with a new prescription e.g. stopping a medication). Safety issues such as patient or carer deskilling with respect to medication knowledge need to be addressed. The panel felt that it was important to measure this knowledge at baseline, before commencing DAAs as a means of monitoring knowledge loss, as many people started DAAs just because they had poor medication knowledge.
The panel raised other issues to be addressed: Situations where 2 pharmacies are used each providing DAAs or where one pharmacy
provides a DAA and another fills the occasional prescription and so that item is not packed. Patients getting medicines from either hospitals (e.g. outpatients for specialist items) or
mental health clinics – not from the pharmacy packing DAAs. The pharmacy needs to maintain the profile but decisions on best practice of packing these non-pharmacy sourced medications are needed
What happens with non-packed medications (also an issue for RCFs)
Preliminary Community model draftNew patient to start DAA(need already assessed)
Is it 6 months since last
formalcheck?
Is it 6 months since last
formalcheck?
Yes
No
Patien
t-he
ld t
em
pla
te k
ept
up-t
o-d
ate
by
pa
tient,
ca
rer
& G
PIn
clu
din
g m
ed
ica
tions c
ha
nge
s n
ot
nee
din
g p
rescrip
tio
n
DAA packed by staff using template
Pharmacist checks DAA using template
DAA sent to patient
Pharmacist check/update template before next packing
Patient-held template/ medication record
Tripartisan Agreement*• Patient/carer• GP (+/- specialists)• Pharmacy
Formalise service to be delivered, expectations & obligations
*see break-out box for content
Full review/ renewal of template
• Patient, GP & pharmacy
Template for medication packing• Shows medication regimen & preferred
time of day for each dose
Reflect• Patient habits/preferences• GP preferences• Current medication & optimal schedule• Type of pack & packing interval• Check medications suitable for packing• Other constraints e.g vision impaired
Pre
scrip
tio
ns
dis
pen
se
d
Template approved by all
Template copy to GP
Issues & obligations for community DAA use
Break out box
Issues to address in agreement Obligations of parties - examples
In simple language
Explanation of how service expected to work
Patient consent to service and necessary information
sharing between GP & pharmacy
Patient, GP and pharmacy agreement to service
obligations
Agreed cost of service including any GP costs as
negotiated
Billing/account aspects
Where prescriptions and original packs physically
stored
Is pack to be collected or delivered (address
timeliness & exception procedures)
How medication regimen changes to be handled
Duration of agreement & an understanding that the
situation to be reviewed every 6 months
Pharmacy to prepare packs at agreed interval
(frequency & period in advance of distribution)
Any additional support services e.g. education, Home
Medicines Review the pharmacy agrees to provide
GPs responsibilities for owing prescriptions &
prescription continuity (e.g. writing repeat
prescriptions without patient consultation) including
timeliness of prescription receipt
Pharmacy processes negotiated to fit in with GP
practice
Address GP expectations for payment & who will pay
if any cost
Patient & GP to give timely notification of pharmacy
of any medication changes including those that do
not generate a prescription e.g. ceasing a medication
Patient/carer & GP to maintain patient held template
Preliminary RCF model draft
Medication Information flows in RCF
GP
Pharmacy* supplying RCF
New resident
RCF
*assumes main supplying pharmacy but resident has
Current drug regimen
Order to
administer drug
Medication chart
Advise of medication
supply system,
obligations &
expectations
Order to supply
Medication chart
Authorising
payment for
supply
PBS prescription
Reminder on
continuity of
supply
RCFmedicatio
n system
including DAAre
fillschedule
System of supply – obligations & expectations
including responsibility for script management,
who notifies pharmacy of changes
Mutual obligations &
expectations
The RCF model was based on information flows and was less well developed because the system was more complex. Note that carers in some RCFs administering medications don’t have to be able to read the chart and that there is only so much a pharmacist can do in this situation.
More input on the RCF model was sought from the panel. It was suggested that it was hard to put all components in one model and that several models may be required.
The RCF model needs to be explicit in what is involved in service provision to address unrealistic expectations about responsiveness and the paperwork trail. The RCFs and GPs need to understand what it takes for the pharmacy to provide the service, what information is needed and how things can go wrong. One of the “best practice” tools could be a promotional brochure or similar to explain these aspects.
Quality Medication Care Group, School of Medicine, University of Queensland290
DAA Best practice consultation framework
RCF Personnel
Nursing Management
Staff
Other Aged Care Stakeholders
NIMAC (peak body)
Residents & Family
Assessors (ACS)
Community patients
Consumer representative organisations
DAA users
Carers/family
GPs perspective
GPs
AMAQ, RACGP (peak bodies)
GP Divisions
DAA Expert Panel
Manufacturers
Pharmacy practitioners
Pharmacy peak bodies (PGA, PSA, SHPA, AACP, QCPP)
Other Stakeholders
TGA (stability)
Health Depts (state, federal)
DVA
Pharmacy Personnel
Pharmacists
Dispensary Assistants
Pharmacy Assistants
Community nurses
Iterative development of models
The consultation plan was presented. The panel indicated two groups missing from the stakeholder consultation plan: Support people like the pharmacy IT
dispensing program companies, Healthconnect, other IT players e.g. Palm, Medical Director; and medication chart people (e.g. Compact Business Systems)
Hospital pharmacists working with DAAs
Initial development will be done on the model then wider consultation.
General methodology for consultation
Initial development
Local consultation – primary stakeholders
• Group techniques, structured interviews
• Explore issues, how things could be done better
• Examine feasibility, usability & potential impact of preliminary model
Second draft
Wider consultation of 1o stakeholders; consult 2o stakeholders
• Structured interview, national surveys
Consultation with peak bodies, consumers, carers and families, GPs and community nurses will test the acceptance of and interest in the models, what sort of problems might they see with the models and possible solutions. The next draft of models will be circulated more widely to each state as the legislation differs from state to state. Example questions might be: what would this model do to your costs, what is the impact on your practice.
The research team plan to feedback to the panel and use the panel as a sounding board.
Tools will be developed in parallel and input is welcome.
Meeting closed at 12.50pm
Quality Medication Care Group, School of Medicine, University of Queensland291
APPENDIX C: FOCUS GROUP AND INTERVIEW
MATERIALS AND REPORTS
RCF FOCUS GROUPS
RCF MANAGEMENT FOCUS GROUP MATERIALS
SESSION: AAPT Hosted telephone conference on the XXXXX 2005
TIME: XXXXX The session is expected to take 1 ½ hours and participants
will be remunerated $100 for their time on receipt of an invoice.
DISCUSSION QUESTIONS:
1. Have there been situations where the use of DAAs has resulted in inefficiencies or
unsafe practices? Please describe.
2. What practices or solutions have been implemented to overcome these problems?
3. We have attached a model of medication supply when using DAAs, please refer to
this model on Page 2. From your perspective what are the missing steps?
1. Medication order on chart
2. Develop or update pharmacymedication profile
0. GP orders & prescribes medication
2a. Support activities• Prescription management*
• Accounts• Pack-chart audit†
3. Dispense prescription
4. Pack medication
4a. Pack medication
5. Check packed medication
4b. Check medication
4c. Deliver to pharmacy
3a. Send profile +/-medications to external packer
6. Deliver medication
7. Medication receipt & storage
7a. Counsel self-medicating residents
8. Administer medication
* If chart for administering and prescription for payment
† Could be done independent of pharmacy
Non-packed medications
9. Medication returns to pharmacy
Different procedures needed for:• New resident• Medication change• Respite (in & out)• Going to & returning from hospital
• Death• Other permanent separation• Other temporary absences
In RCF In PharmacyKEY: External packer
4. What can go wrong at each step that may affect the safety and efficiency of
medication supplied in DAAs? What are some possible solutions?
5. How much education and training is currently provided to facility staff with respect
to administering medication from DAAs and managing the continuity of supply of
medications? How much training should be provided?
6. Are there any issues or barriers to efficient DAA use related to existing standards
and legalities (i.e. Poisons Board, Nursing regulations, Aged Care standards)?
Quality Medication Care Group, School of Medicine, University of Queensland292
7. What impact would the following hypothetical changes have on your current
practices (in terms of quality of care, time, staffing, costs etc):
a. Doctors orders on a medication chart alone are sufficient for the pharmacy to
dispense medication (i.e. prescriptions not required)?
b. The initial supply of DAAs to new residents is integrated with the provision of an
Medication Management Review (MMR)?
c. The residents medication profile is updated via an electronic interface that can
be accessed by the facility, pharmacy and doctor?
RCF STAFF FOCUS GROUP MATERIALS
SESSIONS: April 2005
TIME: XXXXX The session is expected to take 1 hour and participants will be
remunerated $60 for their time on receipt of an invoice.
DISCUSSION QUESTIONS:
In responding to the following questions, please draw upon your day to day
experiences administering medications using dose administration aids.
In your responses, please think about all steps in your medication administration
system as shown in the diagram on the next page.
1. What are the problems you have experienced with the dose administration aids
system used in your facility? How have these problems been dealt with?
Consider steps 0 to 9 in the diagram.
2. Are there any problems that occur particularly when a new resident enters the
facility? What strategies have helped reduce these problems?
3. Are there any other specific problems related to medication administration in the
following situations and how have these been addressed?:
A change in medication orders
Telephone orders
Residents going to and from hospital or other temporary absences
Respite care
Particular medications such as non-packed medications, PRN medications or
Schedule 8 drugs
Death or other permanent separation
Other specific situations
4. Does the medication administration policy at your facility help you in your job? Are
there any aspects of the policy that make it harder to administer medications?
What would you change if you could?
5. Why would a staff member find it difficult to follow all the procedures and guidelines
related to medication management in RCFs?
6. What procedures are in place to ensure that the facility’s policies are complyd to by
staff? How effective are these procedures in minimising the risk of errors or other
problems?
7. Do agency staff have problems following your procedures? What strategies do you
have in place to deal with this situation?
8. Have the attitudes or behaviours of residents and/or their family caused difficulties
with the medication system? How were these problems addressed?
9. Among the GPs who attend residents in your facility, what kind of attitudes or
behaviours do you find helpful in working with your facility’s medication system?
Quality Medication Care Group, School of Medicine, University of Queensland293
What kind of GP attitudes or behaviours do you think make this part of the job
harder? What kind of strategies have been used to address problems?
10. What impact would the following hypothetical changes have on your current
practices (in terms of quality of care, time, staffing, etc):
a) Doctor’s orders on a medication chart alone are sufficient for the pharmacy to
dispense medication (i.e. prescriptions not required)?
b) The initial supply of DAAs to new residents is integrated with the provision of a
Medication Management Review (MMR)?
c) The residents’ medication profile is updated via an electronic interface that can
be accessed by the facility, pharmacy and doctor?
1. Medication order on chart
0. GP orders medication on chart + writes prescription
Inefficiencies and unsafe practices resulting from DAA use
Participants in the RCFs focus groups noted several causes of unsafe and inefficient
practice arising from DAA use. The most commonly cited of these were potentially
unsafe situations due to difficulties experienced by RCFs in accessing medications
after hours (10 responses) – either when emergency supplies were needed or when
medication changes took place. Inability to access medications after hours has the
potential to adversely impact patient health, either through patients not receiving the
necessary medications or through receiving an incorrect dose of medication. This
problem was reported by many participants and was independent of the type of DAA
used. One DON stated that when medication changes occurred after hours, it was
often impossible for the RCF to arrange for a supply of the new medication until the
next day - potentially leading to nurses giving patients incorrect medication dosages. In
terms of system specific practices, one RCF using 5 week Webster multi-dose DAAs,
stated that because medications were packed so far in advance, changes were not
only problematic but also led to substantial medication wastage.
Quality Medication Care Group, School of Medicine, University of Queensland294
Another DAA related problem was the occurrence of packing errors by the pharmacy (7
responses). Where errors occur but are undetected, the patient may again be at risk of
receiving the incorrect type of dosage of medication. Interestingly, during the focus
group it emerged that while some DONs considered packing errors to be of paramount
importance, others were barely concerned at all. Many DONs reasoned that packing
errors would be noted by nurses upon administration and as such there was no need to
check for errors packs upon receipt at the facility. However, as one DON commented,
“if errors aren’t detected first up, they tend to be perpetuated until the pack runs out”.
Detection of errors was also a major concern in low care facilities where Personal Care
Assistants (PCAs) administer medications. Three participants, both DONs and RCFs
staff, noted that checking DAAs for errors was a time consuming and inefficient
practice, especially in cases where DAAs were checked multiple times: once, upon
receipt and once before each medication administration. Another issue DONs reported,
was instances where nurses using DAAs had cut corners to save time (2 responses).
One DON recounted the story of an RN she described as a ‘cowboy’, who had been
administering all medications during the round as per usual, but had been waiting until
after the round finished to sign off on the medications for each patient.
Nursing staff also noted that temporary nurses (from agencies) were typically lax when
it came to signing for medication administration. An interesting adjunct to the
occurrence of packing errors in DAAs was the perception by some nursing staff that
they actually bore the brunt of responsibility for errors made by the pharmacy. One RN
stated that, “once the pills are popped out (of the DAA) there’s no way for us to prove
that they were wrong. It’s demoralising”.
Solutions to inefficient and unsafe DAA practices
All but three of the participating RCFs had implemented solutions to overcome the
unsafe and inefficient practices identified. The most commonly used approach, was to
ensure that staff were thoroughly trained to check DAAs for errors (11 responses).
Most DONs stressed to staff the importance of being aware that pharmacies do make
packing errors and that thorough checking is required. As one DON stated, “we have a
zero tolerance policy for medication administration errors, any errors in administration
are sufficient to trigger extra education or training for the staff member”’. Similarly, one
RN reported that the facility she worked at administered nursing staff with yearly
medication competency testing. As well as training, many RCFs also had auditing and
incident report systems in place for when errors were detected (9 responses).
Generally, RCF protocol was to inform the pharmacy of the error immediately and to
make a notation in the facility register regarding the nature of the error.
To overcome the issue of after-hours medication supply, most RCFs had pre-existing
arrangements with their pharmacy or kept emergency supplies of medication on hand
(6 responses). These facilities reported that while after-hours medication changes
could be inconvenient, they posed no threat to patient welfare, as medications were
generally available in time. There was one RCF that reported great difficulty in dealing
with their pharmacy, especially regarding after-hours medication supply, and noted
several instances where the pharmacy had simply refused to supply medications after-
hours. This also related to the assertion from some DONs and nursing staff that good
communication with the pharmacy could prevent the need for after-hours supply of
medication (2 responses). Indeed, one RCF had never required medication supply after
Quality Medication Care Group, School of Medicine, University of Queensland295
hours – something the DON attributed to good communication with the supplying
pharmacy.
Model of medication supply
Delivery of medication
All participants in the focus groups agreed that existing procedures for the delivery and
receipt of medication are sufficient and that no changes need be made to these. The
only problem encountered was by one RCFs that had been unable to negotiate with
their pharmacy regarding the time of medication delivery, as a result they often
received medications at a highly inconvenient time for them – during the middle of a
medication round.
Medication receipt and storage
RCF respondents were almost evenly split as to whether or not DAAs were checked
upon delivery (5 checked and 5 not checked). Of those who did check, most felt that it
was important to remain vigilant about the possibility of errors and to do as much as
possible to prevent these reaching the patient. Of those who did not check, most
believed that it was the pharmacists’ role to ensure that DAAs were packed correctly
and that and that if any errors had occurred, nurses were likely to pick these up during
the process of administering the medications to the patients. As mentioned previously,
some nurses felt that checking DAAs twice was unnecessary and an inefficient use of
time. Regarding storage of DAAs, none of the participants had received any
information from their pharmacy regarding specific storage conditions, and as far as
they were aware, standard medication storage conditions were adequate. Note
however, that in three instances, DONs and RCF staff from rural areas observed that
on occasions medications would stick to each other, or to the DAA itself, during
transport.
Medication administration
The vast majority of DONs (12/14) reported that DAAs were preferable to Original
Packs (OPs). Five respondents noted not problems with medication administration.
Interestingly this sentiment was not shared by the nursing staff actually administering
medications from DAAs, with one RN stating “DAAs are not a time saving modality, we
are more certain of the medication dispensed if we do it ourselves from the original
packs”. Among the problems that DONs and staff reported with DAAs were: unclear
and confusing medication labeling (9 responses); staff experiencing repetative strain
injury (RSI) from popping DAA blisters (3 responses); and uncertainty about the
suitability of medications for crushing (3 responses). With respect to labeling, problems
with DAAs can be categorised into four groups. First, nurses reported that DAAs could
be difficulty to identify and expressed a preference for DAAs where patient photos were
included (as with some Webster packs). Second, “generic” brand medications were
often not labeled adequately and nurses preferred packs labeled with both the
“generic” brand name supplied and prescribed brand name.
This issue was particularly salient to the nursing staff, with almost all commenting that
DAAs packed and labeled with generic medicines were more time consuming to check
and administer from, as often the medication chart would contain the name of the
originator brand, while the DAA contained only the generic name. Third, patient and
drug information can occasionally be obscured by the frame of the DAA, making it
more challenging for nurses to identify medications. Fourth, DAAs do not typically
Quality Medication Care Group, School of Medicine, University of Queensland296
contain information regarding the suitability of medication for crushing. In the case of
sachets where medications are often crushed in the DAA itself, it was reported that
packing of crushable and non-crushable medication in the same pack created
inefficiencies for nurses who had to identify and remove those medications unsuitable
for crushing. DONs also reported that when patients refused medications, or when
medications were lost or dropped by nurses, DAA use typically meant that there was
not a ready replacement available (2 responses). In these instances nurses would
often remove the lost medication from the last blister on the patients pack, then arrange
for the pharmacy to repack the DAA – resulting in increased medication wastage and
cost. Indeed, one RCF was actually considering changing from DAAs back to OPs
because of the problems caused by advance packing and medication changes.
However, this RCF was using the five week Webster multi-dose pack and was not
representative other participating RCFs.
Return of medications to the pharmacy
All participants reported that pharmacies provided a service for the return of unused
medication. Typically unused medications were collected by the pharmacy when
delivering a new supply. All participants reported satisfaction with this service.
Current and optimal levels of staff training for DAA use
There was a substantial amount of variation between facilities in the type and amount
of training provided to staff (Appendix Table 3). At the lower end of the scale, DONs
reported that no specific DAA training was necessary and that training in DAA use was
incorporated into the nurses’ orientation at the RCF. At the upper end, one DON
reported that 2.5 days worth of training was provided to PCAs who were administering
medications from DAAs. Generally training was done in house - only three of the
participating RCFs had pharmacy input into the training program. Another important
issue regarding DAA use is the potential ‘deskilling’ of RCF staff. This refers to the
widely held view that using DAAs removes the necessity for RCF staff to develop
further skills and knowledge regarding medications, and in fact that using DAAs may
reduce the skills and knowledge of nurses. Interestingly, the majority of participants
(both DONs and nursing staff) felt that this was not an issue, as resources such as
MIMS are readily available and nurses have the ability to contact the pharmacy in
situations of uncertainty. However, one DON strongly disagreed commenting that
DAAs “add to the concept RNs have that administering medications isn’t important, it
takes away their professional responsibility”. This particular facility had been proactive
in addressing the issue of deskilling and had arranged for the pharmacist to send
fortnightly drug information sheets.
Appendix Table 3 Levels of training provided to RCF staff
Theme Frequency
Training is provided by the RCF without pharmacist input 7 Training provided to new staff and initially DAAs were implemented 6 Yearly training, refresher courses or information sheets are provided 5 Training is provided to staff on an ‘as needed’ basis 3 Training is provided with pharmacist input 3 Temporary staff not properly trained 3 Staff are trained through a mentoring system 2
These sheets were then distributed to each nursing staff member to read and sign off
on, to affirm their knowledge. Nurses who had signed the drug information sheet then
Quality Medication Care Group, School of Medicine, University of Queensland297
agreed to be tested about that information at any time. A related issue to that of
training, and one that was posed only to nurses, was the extent to which temporary
nursing staff followed medication administration procedures at the facility. Nurses
reported that temporary staff often did not follow medication administration errors, with
some facilities ceasing to use them as a result. At those facilities that continued to use
temporary nursing staff, there was a real sense of frustration among the permanent
staff that procedures weren’t being followed, as one nurse stated “they have the same
problems as us but they don’t actually live them – they don’t have the same standards”.
Issues and barriers to DAA use resulting from existing standards and legislation
DONs and nursing staff raised several issues and barriers to DAA use that they felt
were due to current legislative restrictions (Appendix Table 4). Most commonly, it was
cited that requiring doctors to write prescriptions, despite already having placed an
order on the patient’s medication chart, was an inefficient and potentially unsafe
practice. Several DONs also noted that the current poisons act, which is from 1968, is
outdated and needs updating to allow for the electronic transfer of records. An issue of
importance to nursing staff, but one that was not mentioned by any of the participating
DONs, was the inability of nurses to make notations on patient charts. Several nurses
commented that they would like to be able to write alternative drug names on
medication charts (e.g. the “generic” brand equivalent of the prescribed brand name),
but they are unable to do so due to legislative restrictions. Another important issue,
primarily for DONs, was the level of inconsistency between states regarding
medications that RCF staff can and cannot administer. DONs expressed a preference
for a national standard of medication administration. It should also be noted, several of
the participants were satisfied with current legislation and did not believe that any
changes were strictly necessary – although as some conceded, they would be nice.
Appendix Table 4 Legislative issues and barriers to DAA use
Theme Frequency Medication charts should be allowed to act as a prescription 8 No changes required 4 Poisons act is outdated 4 Nursing staff unable to make notations on medication charts 3 Confusion as to which medications in what pack type can be administered by which staff (RN, EN, PCA)
3
Differing requirements across state nursing boards 3
Nursing staff also noted problems obtaining a legal and current medication chart in a
timely manner when residents were discharged back to the facility from hospital.
Discharge procedures varied between hospitals (as to information flow and whether
discharge medicines were supplied) and the timing of discharges, GP ability or
willingness to write a new medication chart and prescriptions for the pharmacy to
dispense new medications all complicated the timely administration of the correct
medicines to a recently returned resident. These problems occurred irrespective of
whether medicines were used in original packs or DAAs but repacking medications into
a DAA was an added complication.
Impact of hypothetical changes
As part of the focus group discussions, the facilitator suggested to participants a
number of hypothetical scenarios that arose of out the Phases 1 and 2 report ((Ientile
et al. 2004)).
Quality Medication Care Group, School of Medicine, University of Queensland298
Hypothetical change 1: Doctors orders on a medication chart are sufficient for
the pharmacy to dispense medication
DONs and nursing staff were asked about the potential impact on efficiency and safety
if patients’ medication charts were allowed to act as a prescription. The overwhelming
response from ten focus group participants was that changes to allow medication
charts to act as prescriptions would result in greatly increased efficiency of medication
supply. As one DON stated “it would be great, it would really save us a lot of time and
reduce the amount of paperwork needed”. Only two participants felt that this change
would make no difference to the efficiency of medication supply.
Hypothetical change 2: The initial supply of DAAs to new residents is integrated
with the provision of an Medication Management Review (MMR)
Participants were also asked about the potential impact of incorporating a MMR into
the initial supply of a DAA to patients. Reactions to this suggestion were mixed. A
response typical of most nurses (6) was that while it was a good idea in theory, in
practice it would be, as one DON stated, “a logistical nightmare”. Overall, all
participants agreed that current standards regarding MMRs and DAAs are suitable and
as long as MMRs are conducted within a reasonable time frame (most DONs stated
between one and three months), there is no need for such a change.
Hypothetical change 3: The residents medication profile is updated via an
electronic interface that can be accessed by the facility, pharmacy and doctor
Participants were asked about the potential impact of an electronically based patient
medication profile, that can be accessed by all necessary parties (RCFs, pharmacists
and doctors). Overall, the majority of DONs and nurses were in favour of electronic
medication profiles. As one nurse commented “that would have lots of advantages”.
However, there were several caveats identified: First, participants (particularly DONs)
expressed concern about who would pay for the implementation and upkeep of the
system (9 responses). As one DON noted “it’s a great idea, but that sort of thing always
requires upkeep and support and the facilities just can’t afford it”. Second, there was a
high level of concern about the level computer literacy required by RCF staff,
pharmacists and doctors, to operate such a system work (7 responses). One DON
commented that staff at her facility were currently receiving computer training (for a
separate reason) and that training had proven difficult, as many nurses were not
computer literate. Similar concerns were raised about the computer literacy of doctors,
many of whom both nurses and DONs felt, might not have the requisite skills to use
such a system. Third, several nurses noted that the proposed system would need
safeguards to ensure the accuracy and security of patient information (3 responses).
PHARMACY MANAGEMENT FOCUS GROUPS
COMMUNITY PHARMACIST FOCUS GROUP MATERIALS
SESSIONS: AAPT Hosted telephone conference on the XXXXX April 2005
TIME: XXXXX The session is expected to take 1 hour and participants will be
remunerated $140 for their time on receipt of an invoice.
DISCUSSION QUESTIONS:
1. What are some of the problems or issues that the residential care facilities (or
community patients) have come to you with? What practices or solutions have been
implemented to overcome these problems?
Quality Medication Care Group, School of Medicine, University of Queensland299
2. What problems or issues related to DAA provision and use have arisen within your
business and how have you addressed these?
3. We have attached a model of medication supply when using DAAs; please refer to
this model below. From your perspective what are the missing steps?
4. What can go wrong at each step that may affect the safety and efficiency of
medication supplied in DAAs? What are some possible solutions?
5. Are there any issues or barriers to efficient DAA use related to existing standards
and legalities (i.e. Poisons Board, PSA guidelines, Aged Care standards)?
1. Medication order on chart
2. Develop or update pharmacymedication profile
0. GP orders & prescribes medication
2a. Support activities• Prescription management*
• Accounts• Pack-chart audit†
3. Dispense prescription
4. Pack medication
4a. Pack medication
5. Check packed medication
4b. Check medication
4c. Deliver to pharmacy
3a. Send profile +/-medications to external packer
6. Deliver medication
7. Medication receipt & storage
7a. Counsel self-medicating residents
8. Administer medication
* If chart for administering and prescription for payment
† Could be done independent of pharmacy
Non-packed medications
9. Medication returns to pharmacy
Different procedures needed for:• New resident• Medication change• Respite (in & out)• Going to & returning from hospital
• Death• Other permanent separation• Other temporary absences
In RCF In PharmacyKEY: External packer
6. The results of phase 2 suggest that pharmacies and facilities are not always
complying to the existing standards and guidelines. Why do you think this may be
happening? How can this be fixed?
7. Describe the templates, tools or materials (i.e. PSA templates, DAA supplier forms,
IT solutions) you use to facilitate DAA supply to RCFs? What are the problems with
the existing tools and what tools would be useful to you?
8. What impact would the following hypothetical changes have on your current
practices (in terms of quality of care, time, staffing, costs etc):
a) Doctor’s orders on a medication chart alone are sufficient for the pharmacy to
dispense medication (i.e. prescriptions not required)?
b) The initial supply of DAAs to new residents is integrated with the provision of a
Medication Management Review (MMR)?
c) The residents’ medication profile is updated via an electronic interface that can
be accessed by the facility, pharmacy and doctor?
Quality Medication Care Group, School of Medicine, University of Queensland300
COMMUNITY PHARMACIST FOCUS GROUP REPORT
Problems and issues related to DAA supply
Pharmacists reported a number of DAA supply issues that RCFs had raised with them
and that they had encountered while supplying DAAs to both RCFs and community
patients (CPs) (Appendix Table 5). The most frequently cited and important issue that
pharmacists faced with regard to DAA supply was ensuring that patient medication
charts were unambiguous and up-to-date. As one pharmacist noted “communication in
forms or charts faxed [from RCFs/GPs] are often illegible…some are really shocking”.
The issue of up-to-date medication charts was particularly important. Current charts are
necessary to ensure that the pharmacist provides the patient with the correct
medications, in the correct dosages, as well as allowing pharmacists to identify
potentially dangerous interactions between medications.
Another issue of concern for pharmacists was the amount of time required to supply
DAAs, due to constantly switching between the pharmacy’s dispensing program and
the DAA manufacturer’s DAA supply program. This particular problem occured
because pharmacists must conduct the activities of DAA supply, such as updating the
patients medication information, including dosage time and strength and printing the
DAA label in the DAA manufacturers software, as well as updating much of this same
information, plus additional dispensing and payment information, in the pharmacy’s
own dispensing software. As a result the efficiency of DAA supply was greatly reduced
and a significant amount of pharmacists time was wasted duplicating their activities.
For this reason, several pharmacists voiced the need for an integrated dispensing and
DAA system.
Other frequently cited issues included supply of DAAs to RCFs, where the RCF, CP or
GP didn’t supply prescriptions on time, and inefficiencies in supply caused by frequent
medication changes. In both of these cases DAA supply became inefficient due to
factors that were largely outside the pharmacy’s control. For instance, where the RCF
or GP had failed to supply a prescription prior to packing, the pharmacy must then
spend time following up these sources to ensure that the pharmacist has a prescription
to legally dispense. This problem contributed to the situation known as “owing scripts”
where some pharmacists dispensed a regular medication needed at a particular time to
fill a DAA, before the prescription for the item had arrived from the doctor or RCF.
Community patients not informing the pharmacy of changes or not bringing new
prescriptions to the pharmacy in time also contributed to these inefficiencies.
When medication changes occur during the DAA supply cycle (between DAA packing
and delivery), pharmacists will often have to dispense extra medicine or discard certain
medicines and then repack and reseal the patient’s DAA – resulting in a great deal of
time spent conducting activities which would be unnecessary had the pharmacist
received all of the pertinent information prior to packing the DAA.
Several pharmacists also commented on problems associated with hospital admission
for both RCF residents and community patients. When patients went to hospital, the
pharmacy was not notified and so packed a DAA when it was not needed. When
patients were discharged, establishing the current drug regimen so that a new DAA
could be prepared in a timely manner for a recently discharged patient was difficult, as
was obtaining prescriptions in time for medications that had to be dispensed before
Quality Medication Care Group, School of Medicine, University of Queensland301
packing. The amount and types of medications supplied by hospitals varied as did the
pharmacist’s decision as to whether to repack medications dispensed by the hospital.
One issue which was not explicitly mentioned by many pharmacists, but that was
implicated in many of the problems that were identified (such as frequent medication
changes), was poor communication. Poor communication between RCFs and
pharmacies also resulted in costly inefficiencies when of un-needed DAAs were
supplied. Supply of un-needed DAAs occurs when the pharmacy packs a DAA for a
patient that no longer requires that DAA (either because the patient no longer has need
for sufficient medicines to warrant DAA use or because the patient has moved on from
the RCF). The supply of un-needed DAAs arises because pharmacies will often pack
DAAs in a cycle, and if the RCF has failed to inform the pharmacy that a patient no
longer requires a DAA, but the pharmacy packs on the assumption that they do. This
leads to increased medication wastage and inefficient use of pharmacy staff time
packing DAA unnecessarily.
Pharmacists also reported various other issues that customers had raised with them, or
that they themselves had encountered. These included: frequency of medication audits
in RCFs (pharmacists noted that RCFs conducted audits on an infrequent basis); and
complaints from RCFs and CPs about the costs of DAAs (pharmacies reported that
despite the cost they incurred supplying DAAs, in terms of both labour and materials,
many RCFs and CPs felt that pharmacies were charging too much for DAAs and
should either reduce or eliminate their fees).
Problems internal to the pharmacy were also raised with limited room for packing DAAs
and potential for interruption increasing the potential for packing errors.
Appendix Table 5 Problems and issues related to DAA supply
Problem/issue with DAA supply Frequency
Ambiguous and out of date medication charts 9 Inefficiencies caused by current IT process 5 Ongoing supply of medications without prescriptions to RCFs 4 Inefficiencies caused by frequent medication changes 4 Inefficiencies caused by hospital admission and discharge 4 Community patients not informing pharmacy in time of changes 3 Insufficient room in the pharmacy to adequately pack DAAs 2
Solutions to problems and issues of DAA supply
Pharmacists had a diverse range of solutions to the problems they encountered. In
response to the most common issue of ambiguous and out of date medication charts:
seven pharmacies had introduced a new drug chart system to RCFs and doctors, while
two had worked with doctors to ensure that chart instructions were clearly noted and
kept up to date. Similarly, to overcome issues such as drug changes during the packing
cycle, supply of un-needed DAAs and supply of medications without prescription, six
pharmacies implemented improved and structured communication systems. Improved
communication procedures was a solution for continuity of care problems when
patients were admitted and discharged from hospital. This included changing RCF
practices so that the pharmacy was notified. In several cases, improved communication
procedures at hospital discharge were actually initiated by the hospital pharmacy.
In one pharmacy, an automated prescription reminder system was implemented, to
remind both RCF staff and doctors when medication refills and repeat prescriptions
Quality Medication Care Group, School of Medicine, University of Queensland302
were due. As a result, the owing prescriptions rate fell from 40%-50% to 2%-3%. This
pharmacist also took the novel approach of dispensing medications privately when
prescriptions were not received on time, and charging the RCF accordingly – he noted
“that really gets them into gear”. Two pharmacists improved internal organisation and
use of staff time.
Review of the best practice model of DAA supply
Pharmacists were asked to review the flow diagram of DAA supply developed from the
findings of Phase 2 study (sent to the participants in advance) and to provide
information on any additional steps that should be included to the model, or on aspects
of the model they felt required improvement.
Five of the pharmacists stated that they were satisfied with the model and felt that it
accurately reflected the steps involved in DAA supply. Others suggested
improvements. One area that many pharmacists felt should be included in the DAA
best practice model, was a component allowing for an assessment of the suitability of a
given medication for packing in DAAs (4 responses). In terms of suitability, pharmacists
were interested in information about the stability of medications once removed from
their original packs - primarily any possible adverse effects on medications through
close contact with other medications or potentially increased exposure to a
warm/humid environment. Three participants felt that for model to function as intended,
it should include the provision of up to date, current drug regimen information to
pharmacies by the doctor. Other issues which pharmacists raised included: a provision
in the model for medication changes during the packing process (2 responses),
notation of where accounts should be sent to (i.e. the patient or the RCF); and inclusion
of a step between the dispensing of medication and packing of the DAA, for the
pharmacist to check that the correct medications have actually been dispensed.
Events that may affect the safety and efficiency of DAA supply
Pharmacists identified numerous issues that can impact on the safety and efficiency of
DAA supply. Appendix Table 6 contains a list of the most common barriers to safe and
efficient DAA supply and the solutions pharmacists use to overcome these. The
majority of pharmacists had developed solutions to offset one or more of these
problems, however, such solutions were not seen as foolproof and there still exists the
very real problem of the unsafe medication supply. Pharmacists primarily cited
problems with poor communication, especially related to medication changes, with only
two pharmacies offering solutions to this issue – in the form of detailed records of
communication regarding medication changes and changes to patients’ DAA status.
This solution was designed to ensure that accurate records were kept regarding any
changes involving patient DAAs and that these records would be accessible to all staff
involved in the supply of DAAs.
Another problem related to medication changes, was the inefficiencies that occurred
due to GPs being unaware of the effect that medication changes can have on the
workflow of the pharmacy. One pharmacist noted that most GPs do not know that once
the DAA packing has begun, changes can often involve increased demands on the
time of DAA packing staff and increased wastage of medications. Solutions to this
problem essentially involved educating GPs about the impact that medication changes
have on pharmacy workflow and setting guidelines regarding making medication
changes. These guidelines focused on raising GPs awareness of the DAA packing
Quality Medication Care Group, School of Medicine, University of Queensland303
cycles for their patients and attempting to schedule patient appointments so that any
likely medication changes would occur just prior to the pharmacy beginning a new
packing cycle. GPs were also encouraged to consider the urgency of any proposed
medication change. However, as one pharmacist emphatically remarked “it doesn’t
always work”.
Appendix Table 6 Problems/issues with DAA supply and solutions implemented by
pharmacies
Problem/issue Frequency Solution(s) to problem Frequency
Poor communication, especially in relation to medication changes
6 Keeping detailed record of changes and patient DAA status
2
5 Development of medication change guidelines for GPs
2
Having separate staff packing and checking DAAs
7
Training of DAA staff 4 Keeping clear and accurate medication charts
2
Lack of understanding from GPs about the effect of med changes Packing errors 4
Dedicated area and minimise interruptions
2
Finally, pharmacists identified the potential for unsafe medication supply due to DAA
packing errors. The majority of pharmacies attempted to minimise errors in packs
leaving the pharmacy by having different staff involved in the packing and checking
roles, by training staff regarding the procedures to follow when packing and checking
DAAs and by ensuring that clear and accurate medication charts are available for staff,
from which to pack DAAs. Some pharmacies also set up a dedicated DAA area in
which the packing and checking of DAAs could be conducted with minimal risk of
interruption.
Issues and barriers to efficient DAA use related to current standards and
legislation
Pharmacists commented upon a variety of barriers to effective DAA supply that they
felt were due to the impact of legislation and pharmacy and nursing body standards.
The most common among these, and a problem that was identified by almost every
participant across all focus groups, was the issue of owing prescriptions. As one
pharmacist summarised this issue, “when the prescriptions aren’t available you have to
dispense the medicine as an owing script, which often isn’t legal, but the patient needs
their medication to go on with and if you have their best interests in mind then you must
break the guidelines in order to continue doing your job”.
With respect to the impact of differing RCF guidelines on the use of DAAs, a number of
pharmacists expressed real frustration with this issue (Appendix Table 7). Some felt
that nurses only selectively interpreted the guidelines, others felt that occasionally
nurses avoided taking professional responsibility for medication administration, while
one pharmacist commented that ultimately many of the present pharmacy and nursing
guidelines concerning DAA use are simply not consistent with the real world situation.
Other reported barriers to efficient DAA supply were: keeping profiles up to date
following patients’ release from hospital back into the community; and concerns for
pharmacists regarding carers who administer from DAAs and whether the pharmacist
is then legally responsible for the occurrence of errors. It should also be noted that
Quality Medication Care Group, School of Medicine, University of Queensland304
three of the participating pharmacists were satisfied with the current guidelines and
legislations, while others had only minor issues.
Appendix Table 7 Legislation and standards as barriers to efficient DAA supply
Barriers to efficient DAA supply Frequency
Problems with nurses interpretation of guidelines, professional responsibility and consistency
4
Transition of patients from hospital back to the community – ensuring profiles are kept up to date
4
Current guidelines have unrealistic expectations of what pharmacists should be doing
3
Reasons for pharmacies not complying to current guidelines
The reasons pharmacists might have difficult complying to legislation, standards or
guidelines were explored (Appendix Table 8). The most commonly cited reason for
pharmacists not complying to guidelines and standards was that, as one pharmacist
put it, “sometimes you have to bend the rules to make it work”. Other pharmacists
noted that if the business was not dedicated to supplying DAAs on a large scale,
complying to the standards while still attempting to perform all the tasks of a regular
pharmacist was almost impossible. Pharmacists also cited confusion brought about by
different interpretations by different regulatory bodies and a lack of standards regarding
communication procedures between RCFs and pharmacies. One pharmacist also
observed that as pharmacies are not paid for the work, there is little incentive to
perform it at the highest possible standard.
Pharmacists did make a number of recommendations that might make compliance with
standards easier, including: making payment to cover cost of DAA provision a
requirement; hiring staff dedicated solely to DAA packing; and greater flexibility with
regard to the standards pharmacists are required to meet. Overall, participants felt that
provided guidelines were followed and interpreted sensibly, the problems of DAA
supply caused by these guidelines and standards were not insurmountable. Ultimately,
however, the issue of safe and efficient DAA supply was viewed as one of professional
responsibility more than a rigid compliance to guidelines and standards, as one
pharmacist stated “it really all comes down to the standards of each individual
pharmacist”.
Appendix Table 8 Reasons cited by pharmacists for not complying with existing
standards and guidelines
Reasons for contravention of standards Frequency
Pharmacies must bend the rules to make DAA supply possible 4 Time constraints make it difficult to incorporate DAA supply into pharmacies 3 Confusion caused by different interpretations of guidelines and standards by different bodies
2
No standards of communication between RCFs & pharmacies leads to confusion 2
Materials presently used to supply DAAs and improvements in materials that
may facilitate greater efficiency in supply
Pharmacists were asked what materials or tools they presently used in the supply of
DAAs and what, if any, improvements could be made to these to make supply more
efficient. The tools or materials mainly used were those provided by the DAA
Quality Medication Care Group, School of Medicine, University of Queensland305
manufacturer (9 responses). The other tools identified by 3 participants were the
Quality Care Pharmacy Program delivery log and the PSA packing/checking log.
In terms of improvements, some suggestions were:
A portable (laptop based) IT solution that could be used by pharmacists when
visiting RCFs to check charts and profiles, that could also store images of
medications to aid their identification; and
An IT solution to assist in keeping all charts, packs and signing sheets up to date
and consistent across the board.
Additionally, one pharmacy had already devised their own IT solution to facilitate
communication between the pharmacy and RCFs. This pharmacy had developed the
IT solution as they supplied multiple DAA types and needed the ability to separate
customers by DAA types during the ordering and packing processes.
Impact of hypothetical changes to current DAA practices
Pharmacies were presented with four scenarios in which they were asked to assess
the impact of hypothetical changes to DAA practices in terms of quality of care, time,
staffing costs and general impact on the supply of DAAs.
Hypothetical change 1: Doctors orders on medication charts are sufficient for
pharmacies to dispense medications
Pharmacists were asked to consider the hypothetical impact on DAA supply of changes
that would make doctors orders on medication charts sufficient for the pharmacy to
dispense medications i.e. separate prescriptions would not be required. Theoretically,
this change would eliminate the need for pharmacists to spend time following up owing
prescriptions and should result in greater efficiency of DAA supply. Responses to this
suggestion were mixed (Appendix Table 9). Many pharmacists believed that it would
greatly streamline the supply of DAAs and dramatically reduce the burden on the
pharmacy of owing prescriptions. Conversely, a number of pharmacists expressed
concern about this proposition, with one citing the number of errors that occur on
medication charts as a major problem and also noting that prescriptions were more
“legally satisfying” than simply dispensing from the chart. One pharmacist also felt that
that this scenario should be extended to all chronic care situations.
Appendix Table 9 Hypothetical impact of changes to pharmacy dispensing
requirements
Hypothetical impact of change Frequency
Proposed change would save time and money and reduce owing prescriptions 8 Proposed change is positive however should be approached with caution 3 Proposed change is not appropriate due to the potential for errors 3
Hypothetical change 2: Integration of MMR/DMR with the provision of DAAs to
new customers
Most pharmacists agreed that while the idea of integrating medication review with DAA
initiation was sound in theory, it was unlikely to work in the real world (Appendix Table
10). As one pharmacist stated, “an initial review will improve quality of care but the
practicality will be the main barrier”. Other common responses by pharmacists were
that: provided the MMR/DMR was conducted within a reasonable timeframe (most
nominated one month or less), there was no need to integrate it into DAA supply; and,
the idea might work well in the community setting, but was unlikely to be of benefit in
Quality Medication Care Group, School of Medicine, University of Queensland306
the RCF environment where there were a large number of patients and those patients
often needed medications supplied in DAAs immediately.
Appendix Table 10 Hypothetical impact of changes to integrate MMR/DMR into DAA
supply
Hypothetical impact of change Frequency
Theoretically sound idea but not practical for real world application 6 Unnecessary provided DMR/MMR done within a reasonable timeframe 4 Sound idea in the community setting but not in the RCF environment 2
Hypothetical change 3: Development of a medication profile that can be updated
electronically by RCFs, doctors and pharmacists
Responses were overwhelmingly positive to the idea of having an online medication
profile for each patient, that could be electronically updated by pharmacists, doctors
and RCFs. Most pharmacists (8 responses) commented that it would improve
communication between the relevant parties and/or facilitate the efficient supply of
DAAs. All respondents were in favour of this proposal and felt that it would be of great
benefit in the DAA supply process. Two respondents indicated that it would lead to
improved timing and quality of supply. Some caveats that were mentioned were: the
need to maintain patient privacy; a system to alert all parties as soon as medication
changes occur; and ensuring that all intended users of the system possess the
necessary technical proficiency.
Hypothetical change 4: Implementation of a formal tripartisan agreement for DAA
supply between doctors, customers/RCFs and pharmacies
Pharmacists were asked to consider the impact of implementing a formal tripartisan
agreement pertaining to the supply of DAAs between the patient/RCFs, any doctor(s)
seeing the patient and the supplying pharmacy. It was suggested to pharmacists that
the tripartisan agreement could cover areas such as the type of service to be supplied
as well as the expectations and obligations of each party. Overall pharmacists were in
favour of the idea (Appendix Table 11), with most agreeing that it would assist in
formalizing the responsibilities of each party and clarify the obligations of each within
the DAA supply process, as one pharmacist stated “we don’t tend to use agreements
for a lot of things that we should do, but certainly it is easy for someone to comply if
they know what their responsibilities are in the first place”. Other pharmacists
supported the idea but stated that they would prefer for the agreement not to be legally
binding, as it raised the possibility that pharmacists could be held legally responsible
for the misuse of DAAs by patients or RCFs. Still others noted the need to exercise
caution when entering into agreements with community patients as they may not fully
understand the nature of the agreement or may feel pressure to enter into such
agreements.
Appendix Table 11 Hypothetical impact of implementing a formal tripartisan agreement
for DAA supply
Hypothetical impact of change Frequency
Good idea, would help to formalise responsibilities of each party 7 Sound idea but prefer agreement not to be legally binding 2 Caution must be exercised when entering into agreements with patients 2
Quality Medication Care Group, School of Medicine, University of Queensland307
HOSPITAL PHARMACISTS FOCUS GROUPS
HOSPITAL PHARMACIST FOCUS GROUP MATERIALS
SESSIONS: AAPT hosted telephone conference
The session is expected to take 1 hour and participants will be remunerated $140 for
their time on receipt of an invoice.
DISCUSSION QUESTIONS:
1. What are some of the problems or issues that arise in discharge planning when a
patient is returning to a residential care facility where they use a DAA? What
practices or solutions have been implemented to overcome these problems?
2. Are specific guidelines required for patients admitted from and discharged to
residential care?
3. What are some of the problems or issues that arise in discharge planning when a
community based patient has been using a DAA service? What practices or
solutions have been implemented to overcome these problems?
4. What problems or issues arise in compiling a current medication profile for a new
admission and consulting with the patient’s general practitioner? What strategies
have been used to facilitate this process? Describe the templates, tools or materials
you use (i.e. RACGP health summary form, IT solutions).
5. When a patient is admitted from a RCF, what medication information is provided to
the hospital and who provides this information? Is this adequate?
6. What information is provided to the RCF when the patient is discharged? Who else
is provided with information (i.e. patient, doctors, community pharmacists). How
would you improve this communication process?
7. Do you recommend and/or provide DAAs as part of a discharge plan? If yes, how
often and in what circumstances? What are the procedures involved in
arranging/providing this service and could they be improved?
8. What is your hospitals’ policy relating to the use of the patient’s own medication
and does this differ when a patient brings in their medication in a DAA?
9. The APAC guidelines state that prior to discharge, adequate/sufficient medication
should be dispensed? How do you know what is adequate and does a patients’ use
of a DAA affect this judgment?
10. How can medication wastage be minimized when a patient is admitted to hospital
or discharged from hospital?
11. Are there any issues or barriers to efficient DAA provision related to existing
standards and legalities (i.e. Poisons Board, APAC guidelines, hospital standards,
patient health charters and Medicare agreements)?
12. The results of phase 2 suggest that pharmacies and facilities are not always
complying to the existing standards and guidelines. Why do you think this may be
happening? How might compliance to guidelines be enhanced?
HOSPITAL PHARMACIST FOCUS GROUP REPORT
Problems and issues with discharge planning for patients returning to RCFs and
solutions implemented by hospital pharmacies
Hospital pharmacists were asked to consider the types of issues that arose during
discharge planning for patients returning to RCFs where DAAs were used. Essentially,
Quality Medication Care Group, School of Medicine, University of Queensland308
pharmacists noted that the core problem encountered was one of communication,
especially with respect to the RCF informing the hospital of what medications the
patient was on originally, how many remain, and in turn, the hospital pharmacy letting
the RCF know what medications have changed. This was particularly important if
patients had been prescribed new medications while in hospital and would be
continuing to use these following discharge. In terms of practices and solutions used by
hospital pharmacies to overcome this issue, the two most commonly used were: 1.
Supply of a discharge sheet for patients returning to RCFs (four responses); and 2.
Faxing RCFs a copy of any new prescriptions and providing verbal and/or written
updates of any new information about patients’ medications (three responses). Other
strategies used included: supplying prescriptions only for new medications, not all
medications the patient received while in hospital; and developing a hospital database
to match patients and RCFs with their pharmacies, to ensure continuity of care and
ongoing supply of medications.
Proposed requirement for specific admission and discharge procedures for RCF
patients
Participants were asked whether or not they felt that specific patient guidelines should
be developed for patients regarding admission from, and discharge to, RCFs.
Theoretically, such guidelines should would provide standardised procedures for
patient admission and discharge, and ensure a uniformly high standard of patient care.
Overall, hospital pharmacists were divided on this issue, with some believing that
specific procedures would help to reduce inconsistencies in the admission and
discharge of patients from different RCFs, while others felt that standardised
procedures were unnecessary, provided common sense was allowed to prevail (see
Appendix Table 12). Of those pharmacists that favoured the development of formalised
guidelines, many expressed concern about the likelihood that their implementation
would place extra demands on the time of all involved, but felt that the benefits
outweighed the costs. There was also a large group of pharmacists who were
ambivalent about the prospect of introducing formal guidelines, because, as one
Appendix Table 12 Proposed requirement for specific guidelines for patients requiring
admission from, or discharge to, RCFs
Requirement for specific guidelines Frequency
Yes – inconsistencies across hospitals make guidelines necessary 5 Ambivalent - formal guidelines are unnecessary 3 No - a common sense approach is sufficient 2
Problems and issues in discharge planning for community patients and
practices and solutions to overcome these
The problems experienced at discharge planning for community patients were diverse.
The only problem cited by more than one pharmacist was: patients not being able to
answer questions about their DAA and not bringing their DAA to hospital with them
(two responses). This was problematic for pharmacists for two reasons: (1) as they
could not be sure of the medications that patients were taking and meant that,
potentially, patients may be at increased risk of adverse drug events, and (2) if the
hospital pharmacist did not know that a patient usually used a DAA, appropriate plans
for discharge supplies were not made. Other problems pharmacists had encountered
included: patients using old DAAs instead of their latest one; planning ongoing
Quality Medication Care Group, School of Medicine, University of Queensland309
medication supply for patients living in remote areas – especially indigenous
communities; and ensuring the availability of prescriptions for ongoing supply.
While the range of problems was diverse, the range of solutions was not.
Overwhelmingly, the most common solution utilised by pharmacists was, simply,
improved communication (five respondents). Commonly employed strategies to
improve communication were: liaison with the patient’s pharmacy to ensure a DAA was
packed ready for discharge (with any new medications if necessary); and ensuring a
consistent flow of information between the hospital, patient and pharmacy regarding
medication changes. One hospital pharmacist placed specially made stickers on
patients DAAs upon discharge, which read “not to be used until checked by local
pharmacy”. This served the dual purpose of providing the patient with a reminder and
reminding the community pharmacist that they should check the patient’s DAA and
current prescriptions (including any added while in hospital) before allowing the patient
to continue use of that DAA.
Problems and issues for compiling medication profiles for new patients and
practices and solutions used to overcome these
Incomplete, missing or unreliable patient information was the main problem faced by
hospital pharmacists when compiling medication profiles for new patients; it was the
only problem raised in the focus groups (six respondents). Complete medication
information for patients is necessary, as it allows pharmacists to ensure all appropriate
medications are written on the hospital drug chart, to detect potentially risky drug-drug
interactions, as well as helping them plan for patient discharge.
The solutions used to address this issue varied. To do this, some pharmacists had
junior doctors at the hospital chase missing information from the patient’s community
pharmacist, others developed medication reconciliation forms, while some used the
patients’ DAAs as a source of information (see Appendix Table 13). In any case, the
only satisfactory solution was for pharmacists to acquire the missing information from
whatever sources available.
Appendix Table 13 Practices and solutions for overcoming missing information when
compiling medication profiles for new patients
Practice/solution Frequency
Pharmacist delegates work to other hospital staff 6 Development of a medication reconciliation form 2 Use of patient DAAs as a source of information 2
Information, and adequacy of information, provided to hospital following patient
admission from RCF
When asked about the type and adequacy of information hospital pharmacies received
following patient admission from an RCF, responses were generally positive. Six of the
pharmacists felt the information they received was adequate (e.g. a patient transfer
form and a fax of the patient’s medication profile, from the RCF). Two pharmacists
reported that the information they received was inadequate. Additional information they
felt should be supplied were: the name of the patient’s GP and supplying pharmacy;
and, having the patient bring their DAA with them upon admission.
Quality Medication Care Group, School of Medicine, University of Queensland310
Information provided to RCFs and other sources upon patient discharge and
potential improvements to the communication process
Hospital pharmacists were also asked to report on the type of information they provided
to RCFs, doctors and community pharmacists following patient discharge together with
any improvements they felt could be made to the communication process. Appendix
Table 14 contains a list of the most common types of information provided by hospital
pharmacists to RCFs and other sources. The majority of pharmacists provide RCFs
with a discharge summary but do not routinely provide any information to the patient’s
community pharmacist. While it was desirable, to include the community pharmacy in
the communication loop, it was not generally actively initiated by hospital pharmacists.
As one pharmacist stated “when we get involved with RCF patients, the charts go to
the nursing home but we don’t contact the community pharmacy. If the community
pharmacy didn’t receive a copy it’s a bit of work for us, so we would like to know
whether that communication is happening [between the RCF and community
pharmacy]”.
There were a number of hospital pharmacists who did provide discharge summaries to
both the RCF and the community pharmacy. Several of these pharmacists noted that it
was essential to ensure that the patient’s community pharmacist was kept up-to-date
on all medication changes. Finally, there were some hospital pharmacists that simply
provided the patient with a medication profile and requested that the patient pass it
onto the relevant parties as reflected by the comment: “the patient is also given a folio
which contains what the hospital pharmacy believe to be a complete and accurate
record, which we assume is then passed on to the RCF or carer”. There was some
discussion that in this situation, there was a reasonable possibility that the intended
party(s) might not receive important patient information.
Appendix Table 14 Type of information provided on patient discharge and source
information is provided to
Information provided and source provided to Frequency
Doctor’s letter/ discharge summary – provided to RCFs, no information provided to community pharmacy
6
Doctor’s letter/ discharge summary – provided to RCF and community pharmacy 4 Medication profile– provided to patient to pass on to RCF/carer 4
Recommendation/provision of DAAs as part of patient discharge and
circumstances and procedures involved in any such recommendation
The issue of whether hospital pharmacists recommended or provided DAAs for
patients upon discharge was explored, and if so, under what circumstances this might
occur and the procedures followed in those cases. As Appendix Table 15 shows, the
majority of hospital pharmacies did not have an official policy regarding DAA provision,
but in some cases provided patients with a DAA if both the patient and the hospital
pharmacist agree it upon it: “we have no set policy regarding DAA supply so it’s just up
to the ward pharmacist and whether the patient agrees to using a DAA”. A greater
number of hospital pharmacies did not actually provide DAAs themselves, but were
willing to recommend patients to pharmacies that would. Reasons for not providing
DAAs through the hospital pharmacy were, characteristically: “yes we will recommend
DAAs if appropriate – but we [hospital pharmacy] don’t fill them ourselves as it is too
time consuming and its also confusing for the patient if they have to change the type of
DAA later on ”.
Quality Medication Care Group, School of Medicine, University of Queensland311
Appendix Table 15 Provision of DAAs upon discharge and circumstances surrounding
provision or recommendation of DAA use
Recommendation and circumstances for DAA provision Frequency
Hospital has no formal policy regarding provision or recommendation of DAAs 6 Hospital will not provide DAAs but will recommend a pharmacy that does 6 Hospital will provide DAAs if agreed by both the hospital pharmacist and the patient
3
Hospital policy regarding patients own medication and own DAAs
A critical factor in the wastage of medications is what happens to those medicines that
patients bring with them to hospital, either in original packs (OPs) or in their DAAs.
Responses to this issue were polarised, with six pharmacists stating that patient
medications were not used when bought to hospital (regardless of whether in OP or
DAA), while four pharmacists stated that patient medications could be administered
from DAAs providing certain criteria were met (those criteria and frequency of response
are contained in Appendix Table 16). There were no standard criteria for use of
medications in DAAs among those hospitals that allowed it; however, in all cases, use
of the DAA was dependant upon inspection and approval by a hospital pharmacist. Of
those hospitals that did not permit the use of medications from DAAs, the most
frequently cited reason was that once medications were packed in DAAs, they all
looked similar and could be confused by nurses during administration.
Appendix Table 16 Criteria for use of patient DAAs following hospital admission
Criteria for DAA use Frequency
DAA use not permitted in hospital 6 DAA can be used if quality judged as good by a hospital pharmacist 2 DAA can be used if a hospital pharmacist has individually removed and identified each medication
1
DAA can be used by nurses with the approval of a hospital pharmacist until the hospital’s supply of the medication is available
1
Judgment of adequate/sufficient medication supply upon discharge and relation
to DAA use
According to APAC guidelines (Australian Pharmaceutical Advisory Council 1998),
hospital pharmacists are required to provide patients with ‘adequate/sufficient’
medications upon discharge from hospital without the meaning of “adequate or
sufficient” being defined. Pharmacists were asked what they considered to be an
adequate/sufficient supply and whether or not this definition changed for patients using
DAAs. The definition of adequate/sufficient adopted by the different hospital
pharmacies could be categorised into three groups (Appendix Table 17). Hospitals
were split almost equally between the three length of supply groups, with each of the
pharmacists stating that the timeframe of supply defined as being adequate was
determined as a matter of hospital policy without input from the dispensing pharmacist.
No pharmacist reported differences in the length of medication supply for DAA versus
non-DAA patients.
Appendix Table 17 Length of time for which supply of discharge medication is deemed
adequate/sufficient
Adequate/sufficient supply Frequency
Supply of 5 days or less 4 Supply of 7 days 3 Supply of standard PBS dispensed medication quantity 3
Quality Medication Care Group, School of Medicine, University of Queensland312
Methods for minimising medication wastage during patient admission to, and
discharge from, hospital
In addressing the issue of medication wastage, hospital pharmacists advocated a
variety of approaches, many of which, subject to hospital policy, could most likely be
implemented immediately. The most commonly cited means of minimising medication
wastage, was returning to patient’s the medications they brought with them on
admission to hospital (3 responses). However, one obstacle to this approach was
actually locating a given patient’s medication – as one pharmacist said “we try to
minimise wastage by returning the patient’s medications to them, but we can’t always
find them”. Another suggested strategy to minimise wastage was to make it compulsory
for community pharmacies providing a post-discharge DAA to use patients’ discharge
medications rather than disposing of these and dispensing new prescriptions (2
responses). One pharmacist stated that this was her ”biggest gripe”; she believed that
community pharmacies were using the dispensing fee as a means of subsidising the
cost of DAA supply, instead of trying to minimise medication wastage. One final
suggestion for reducing wastage was to have hospitals use the medications patients
bring with them on admission (a practice already occurring in some locations) (2
responses).
Issues and barriers to DAA provision based on current standards and legislation
Hospital pharmacists were asked to consider whether any of the current standards and
legislation regarding DAA provision, acted as barriers to the efficient supply of DAAs. A
number of pharmacists felt that there were no legislative barriers to efficient DAA
provision (3 responses), while others felt that stricter guidelines regarding who could fill
DAAs (2 responses). The guidelines or legislation about who can fill DAAs were
inconsistent and ambiguous, and there was disagreement among pharmacists. One
pharmacist who stated, “sometimes nurses fill them instead of pharmacists, but it’s
better to have strict guidelines as to who can fill”, alternately there was a pharmacist
who said “we investigated DAA provision and found that a pharmacist has to label and
dispense them, but I see no reason why someone else can’t do it”. In both of these
cases the pharmacists were advocating more defined guidelines for the packing of
DAAs, but approached this issue from different angles.
Increased resources and money for DAA provision, was also thought to help make
DAA supply more efficient (2 responses). Hospital pharmacists noted that most often
DAAs were filled under sub-optimal conditions, by over-worked pharmacists and
pharmacy staff, for little or no monetary reward. Hospital pharmacists expressed the
opinion that increased remuneration for DAAs would improve all DAA related activities,
including streamlining the process of patient admission and discharge from hospital.
Reasons why pharmacies are not complying to current guidelines and ways
compliance may be improved
Finally, hospital pharmacists were asked to consider the reasons why pharmacists may
not always comply to the current guidelines for DAA provision, and ways in which
compliance to these guidelines might be enhanced. Pharmacists voiced a variety of
opinions, the most common being that there is simply insufficient time and resources
available to supply DAAs and follow all necessary guidelines and standards (three
respondents). One pharmacist noted, “time is the key feature – there’s a lot of things to
be done in a short period – it is one big problem”. Another reason stated for non-
Quality Medication Care Group, School of Medicine, University of Queensland313
compliance to guidelines in the hospital setting was that all of the discharge protocols
when DAAs were involved were “not practical to follow” and that often the guidelines
were unclear. Only one pharmacist actually suggested a solution to enhance
pharmacists’ compliance to DAA provision guidelines, and this was, quite simply, “to
make the guidelines more clear”.
PHARMACY DISPENSARY ASSISTANTS/TECHNICIANS
INTERVIEWS
DISPENSARY TECHNICIAN INTERVIEW MATERIALS
1. We have attached a model of medication supply when using DAAs; please refer to
this model below. From your perspective, are there any missing steps? What
activities are you involved in?
2. What can go wrong at each step that may affect the safety and efficiency of
medication supplied in DAAs? What are some possible solutions?
3. What kind of training did you receive to help you perform the tasks you need to do
to pack DAAs (how much, what kind). Could the training have been better? In
what way could it be improved?
1. Medication order• Prescription from GP• Medication chart from RCF
2. Develop or update pharmacymedication profile
0. GP orders & prescribes medication
2a. Support activities
• Prescription management*
• Accounts
3. Dispense prescription
7. Pack medication according to profile
8. Check packed medication
10. Deliver / collect medication
11. Medication receipt &/or storage
12. Administer medication
* If chart for administering and prescription for payment
Non-packed
medications
13. Medication returns to pharmacy
Different procedures needed for:• New resident• Medication change• Other situations eg. going to &
returning from hospital
External Internal to PharmacyKEY:
4. Check dispensed prescription
6. Prepare for packing (labels, assembling medications etc)
5. Store dispensed medications not yet packed
9. Store packed medication
4. How much detail was there in the instructions you were given or the procedure you
were asked to follow? Is there flexibility for performing tasks differently?
5. On top of your initial training, what other skills and knowledge have you gained to
help you in your role of providing DAAs i.e. things you learned while doing the job?
6. What is your involvement in starting people on DAAs? Do you hand out DAAs or
deliver them? Are these processes working well? How could they be improved?
Quality Medication Care Group, School of Medicine, University of Queensland314
7. What guidelines, standards or regulations that relate directly to DAA services are
you familiar with?
8. Are there any issues or barriers to efficient DAA use related to existing standards
and legalities (i.e. Poisons regulations, PSA guidelines, Aged Care standards)?
9. The results of phase 2 suggest that pharmacies and facilities are not always
complying to the existing standards and guidelines. Why do you think this may be
happening? How can this be fixed?
10. Describe the templates, tools or materials (i.e. PSA templates, DAA supplier forms,
IT solutions) you use to facilitate DAA supply? What are the problems with the
existing tools and what tools would be useful to you?
Quality Medication Care Group, School of Medicine, University of Queensland315
APPENDIX D: LEGAL OPINION
HYPOTHETICAL DAA SITUATIONS FOR CONSIDERATION OF
LIABILITY
The following information and hypothetical situations were provided to Guild Legal to
highlight issues related to best practice and to provide a framework for comment
Background: The steps involved in arranging for medications to be packed into dose
administration aids (DAAs) including the administration of medications is shown in the following
figure.
1a. GP writes medication order• On medication chart in RCF as order to administer• Records full current drug regimen for community patients
1b. GP writes prescription as order to dispense
3. Develop or update pharmacy-held medication profile• Review regimen, prepare dosing schedule (what to take when)• Develop packing plan (identify what is packed & not packed)
2. Transmit drug regimen information to pharmacy
3a. Support activities• Manage prescriptions
(e.g notify when new script due)
• Accounts, adequate stock on hand
4. Dispense prescription
8. Pack medication according to profile
9. Check packed medication
5. Check dispensed prescription
7. Prepare for packing (labels, assembling medications etc)
6. Store dispensed medications not yet packed
10. Filled packs stored in pharmacy
11. Deliver / collect medication
12. Medication receipt
13a. RCF stores medications (packed & non-packed)
13b. RCF self-medicator residents given medications
13c. Counsel on medications (community patients & RCF self-medicators)
14. Administer medication (packed & non-packed) 15a. Check medications taken by RCF self-medicators
15. Record administered medications on RCF chart
16. Return unused medication to pharmacy• Pharmacy records & monitors missed doses,
taking action as needed; deals with medications
Non
-pa
cke
d m
ed
ica
tio
ns
In Residential Care Facilities, medications can be administered by Registered Nurses (who
have their own duty of care including checking medications against a medication chart that is
the doctor’s authorisation to administer a medication), Medication-endorsed Enrolled nurses
(not as much training as RNs but certified to administer medications). In some RCFs, a
Personal Care Assistant (PCA) or carer (a facility staff member unregistered and unqualified to
administer medications under the regulations of the various states) assists a resident to take the
Quality Medication Care Group, School of Medicine, University of Queensland316
contents of a dose aid blister, or certain residents self-medicate. Pack types used can be blister
unit dose or multi-dose packs, or unit or multi-dose sachets (untrained staff and self-medicators
will only use a multi-dose pack). Dosett type boxes are not used in RCFs.
For people in the community using DAAs, they take the content of the blister in the pack without
reference to any medication chart. In some cases, a carer (often a spouse) assists them to take
medications. In this case, the carer is generally not trained. Multi-dose devices of all 3 types
might be used at home.
In developing a best practice model, we are focusing on maximising safety (prevention of
errors), effectiveness (in improving compliance and management of medications) and efficiency
(e.g. in packing and minimising rework, for example, when changes are made).
Situation 1: A pharmacy supplies a residential care facility with a DAA that is erroneous and a
Personal Care Assistant (PCA) or carer assists a resident to take the contents of a dose aid
blister with the wrong medicine or wrong dose, and the resident has a negative outcome. The
error is due to discordance between the pharmacy profile and the facility chart either due to the
failure of the facility to adequately communicate changes to the resident’s regimen to the
pharmacy or due to the failure on behalf of the pharmacy to update the resident’s profile and
revise any packed medication. How would the following procedures or practices impact on the
relative liability of the pharmacy and/or facility?
The pharmacy and facility have an agreement in place that clearly defines that it is the facility’s responsibility to ensure that changes to medications are communicated to the pharmacy in time for changes to be made to packs versus no formal understanding.
The pharmacy maintains written records of all communications from the facility about medication changes but has no records relating to this particular medication change versus the pharmacy keeps no documentation of communication.
The facility maintains a record of all communication with the pharmacy about patients medication changes and has a record of communication relating to this incident versus the facility keeps no documentation.
The medication change is a substitution of one drug for another (one stopped and another started) but both drugs are packed and administered (like a double dose), because the facility notified the pharmacy of the new drug only versus communicating the whole revised drug regimen to the pharmacy.
The pharmacy has a procedure and documentation as evidence of regularly reviewing medication charts to ensure the pharmacy profile is accurate versus the pharmacy has no policy or documentation of reviewing charts.
The facility has a policy of having an RN check packs prior to administration by unqualified staff versus the facility has no procedure or documentation of checking packs that will be administered by unqualified staff.
Situation 2: A pharmacy supplies a facility with a DAA that is erroneous and the error is
administered to the resident by a RN and the resident has a negative outcome. The error is due
to human factors; mistake made by packer in the pharmacy (i.e. wrong drug/dose is packed)
and the error is administered as the RN didn’t detect the error because the RN failed to check
the chart, but followed the pack label or just administered the contents of the blister due to be
administered. How would the following procedures or practices impact on the relative liability of
the pharmacy and/or facility?
The pharmacy maintains records of checking the packs before delivering to the facility versus the pharmacy checks the pack but maintains no records of checking packs.
The facility has documentation to indicate that the RN did check the residents chart (i.e. chart is signed by RN) versus the facility has no documentation to support the RN checking the pack against the chart.
The facility has internal monitoring/auditing procedures for medication administration by RNs
Situation 3: A pharmacy supplies a community patient with a DAA that is erroneous and the
community patient takes the medication according to the pack and has a negative outcome.
Quality Medication Care Group, School of Medicine, University of Queensland317
The error is due to the fact that the patient’s profile (from which the pharmacy prepares the
packs) does not reflect the patients current regimen either due to the failure of the patient/doctor
to adequately communicate medication changes to the pharmacy or due to the failure on behalf
of the pharmacy to update the patient’s profile and reflect these changes in the packs. How
would the following procedures or practices impact on the comparative liability of the pharmacy?
The pharmacy maintains written records of all communications from the patient or doctor about medication changes but has no records relating to this particular medication change versus the pharmacy keeps no documentation of communication. The patient, doctor and pharmacy have a formal agreement stipulating whose responsibility it is to communicate changes to the pharmacy versus there is no formal agreement. The medication change is a substitution of one drug for another (one stopped and another started) but both drugs are packed and administered (like a double dose), because the change was made by providing the pharmacy with a prescription for the new medication only versus communicating the whole revised drug regimen to the pharmacy.
REPORT FROM GUILD LEGAL LTD: DOSE ADMINISTRATION
AIDS (DAAS) - SOME LIABILITY CONSIDERATIONS
The following summary of liability issues surrounding the provision of DAA services
was prepared by Elizabeth McDowell, Principal Solicitor, Guild Legal Limited (28 June
2005)
Introduction
In this paper, we consider:
1. The areas of professional responsibility requiring particular focus in arranging for
medications to be packed into DAAs.
2. By reference to a number of scenario’s involving the packaging of medications in
DAAs, to review:
a. Liability issues which may arise in arranging for medications to be packed in to
“DAAs” by a pharmacist to be supplied either to a patient within a residential
care facility or to a community patient.
b. Some strategies which could be implemented by pharmacists in relation to
ensuring the minimisation of claims against pharmacists in supplying
medication packed into DAAs.
Pharmacist’s Professional Obligations- some matters requiring particular
attention
The provision of medications packed into DAAs is a service provided by pharmacists to
the community benefiting in particular the elderly, the infirm and those who are
responsible for their care. The provision of this service is at a cost to the pharmacist
which is often significant (Refer Effectiveness and Cost Effectiveness of Dose
Administration Aids – Final report 5/11/2004 – Project conducted by Quality Medication
Care Pty Limited in conjunction with the Therapeutics Research Unit, University of
Queensland, Princess Alexandra Hospital) but the costs pressures placed on the
pharmacist do not diminish in any way the professional obligations of the pharmacist in
relation to the dispensing of medication. At all times appropriate protocols and
procedures must be observed and in this regard, reference is made to published
standards in relation to appropriate protocols to be maintained in the pharmacy
Some of the professional consideration that arise in particular in relation to the
packaging of medications in DAAs include:
Quality Medication Care Group, School of Medicine, University of Queensland318
1. The requirement at all times that appropriately trained and supervised persons are
involved in the packaging of medications into DAAs.
The pharmacist at all times remains responsible in relation to the medications so
packed and appropriate systems (including the checking of medication charts, the
reviewing of original packaging of medications, the keeping of appropriate records,
ensuring the integrity of packed medications) must be maintained and recorded.
The Pharmacy Board of NSW has published in Bulletin VIII some timely reminders
in relation to those matters necessary to ensure sound professional practice.
2. Whilst the practical considerations in relation to counselling may differ in the
context of supply of medication to patients within a residential care facility, the
professional obligations in relation to medications dispensed which are packaged in
the DAAs by the pharmacist remain. It is important that the pharmacist ensures that
proper information and counselling is provided to residents within the residential
care facility and their carers to ensure that medication regimes are fully understood
so as to be safe and effective. In relation to patients resident within a residential
care facility, the facility must have in place appropriate procedures to ensure that
medication is appropriately administered and this is a separate liability which rests
with the facility. It is important that the pharmacist who is involved in the provision of
medication packaged within DAAs to patients within a residential care facility
ensures that, as a separate obligation, effective communications are maintained
with appropriately qualified staff in the facility and that the facility at all times has
access to accurate and up-to-date information in relation to medications. It is
important that the fact of the provision of this information be recorded and records
retained within the pharmacy.
Proper and effective counselling is also critical in relation to DAAs used by
community patients and the pharmacist must at all times ensure that the patient, or
their carer, knows how to use the medication correctly and is provided with
accurate and up to date information in relation to the medication. Particular
attention is required in the context of patients who return to their own homes after
being in a residential care facility (and who no longer have their medication
packaged in a DAA) and in those patients who change from having their medication
packaged in a DAA to collecting the medication in original packaging. When
medications belonging to a patient which have not been included in DAAs are
collected by the patient, particular care is required in relation to checking the
medication provided to the patient and to ensuring that the patient or their carer
understands how to use the medication correctly. In this situation it would be
important to keep a record in the pharmacy of the medication which was provided
and the counselling which was given. In addition to maintaining copies of
dispensed prescriptions it would be useful to record on the patient’s records the
medication which was collected by the patient and some details of the counselling
provided.
3. An issue which unfortunately arises with some frequency in the context of
medication packaged in DAAs is the recycling of medications previously dispensed
in a DAA which is returned to the pharmacy, normally by a residential care facility.
The re-use of previously dispensed medication (dispensed for one patient and then
subsequently “re-dispensed” for another patient) constitutes not only a fraud
against the Health Insurance Commission in terms of a double claim being made
for pharmaceutical benefits in respect for the same medication, but also a breach of
the pharmacist’s professional and ethical obligations to each of the patients of the
Quality Medication Care Group, School of Medicine, University of Queensland319
pharmacy. A pharmacist simply can not be properly satisfied that, in the case of
returned medication, the integrity of that medication has not been compromised by
incorrect storage or tampering. Likewise, using one patient’s (unwanted) medication
which is stored for that patient within the pharmacy (and for which payment has
been made by that patient) for another patient is not acceptable, even though the
original medication may not have been taken outside the pharmacy. If medication is
no longer required by a patient, it must be disposed of in the appropriate container
for unwanted medication. It is critical that a pharmacist have well documented
procedures in relation to the proper disposal of returned medication and that the
RUM system is properly utilised.
Some liability issues involving the use of DAAs and strategies for consideration
The following are presented as examples of matters which occur in professional
practice involving the packaging of medications in DAAs highlighting some of the
liability issues which may arise and some strategies for consolidation.
Situation 1
A pharmacy supplies a residential care facility with a DAA that is incorrect. The error is
due to a difference between the pharmacy profile maintained in relation to the patient
and the facility chart maintained by the residential care facility. Such an error would
arise as a result of either the facility failing to adequately communicate to the pharmacy
changes to the resident’s medication regime or due to a failure by the pharmacist to
update the patient’s profile and revise the packed medication. A personal care
assistant or carer, assists the resident to take the contents of the DAA, the wrong
medication (type or dose) and the patient has a negative outcome.
Some strategies which would minimise the risks of such an error occurring and which
would assist in clarifying the professional responsibilities of the pharmacist vis-a-viz the
residential care facility in the event of such an error occurring would include the
following:
1. The Contract for provision of pharmacy services to the residential care facility
should clearly set out that it is the residential care facility’s responsibility to ensure
that changes to medication are communicated to the pharmacy in a clear and
unequivocal way and in a timely manner to enable any changes to be made to
medication packs. This would reinforce the obligation of the residential care facility
in its dealings with the pharmacist and it would be of assistance in the relationship
between the pharmacist and the facility that this be set out in the contract.
2. In relation to the contract with the residential care facility, it would obviously be of
assistance to all parties if the agreement stated each party’s professional
responsibilities and obligations. It is important that proper consideration be given to
the way in which these responsibilities are described so that additional burdens are
not placed on the parties. Some pharmacists may have a concern that entering
into such agreements may place on them an additional liability over and above their
own professional responsibilities – for example, that they may in some way be
liable for misuse of DAAs by patients or residential care facilities. A contract which
properly sets out the obligations of the parties in relation to the supply of DAAs
would not in itself create such additional burdens and indeed would reinforce the
separate obligations of the residential care facility in relation to DAAs.
3. The pharmacist should have in place appropriate protocols to ensure maintenance
of written records of all communications from the facility about medication changes.
Quality Medication Care Group, School of Medicine, University of Queensland320
This would be a useful risk management tool for the pharmacy in the event that the
facility asserts that a medication change was communicated which is disputed by
the pharmacy.
4. The residential care facility maintains a record of all communications with the
pharmacy about medication changes. Again, this would be a useful risk
management tool for the residential care facility in the event that receipt of a
medication change is disputed by the pharmacist.
5. Where the medication change involves the cessation of one medication and the
commencement of another medication, this must be clearly and unequivocally
stated in the direction to the pharmacist. The pharmacist should, as part of their
sound professional practice, have careful regard to any medication changes and, if
in doubt, verify the medication requirements of the patient with the qualified person
who has authorised the medication. If the prescriber has ordered that a medication
is to be ceased, such remaining medication should be removed from the medication
currently stored for that patient within the pharmacy and either appropriately
disposed of or separately stored. It is appreciated that, where the prescriber is
altering a patient’s medication regime on a regular basis, an appropriate protocol
may include the separate storage of previously dispensed medication for a patient
and, provided that appropriate protocols are maintained (including that close regard
is had to the expiry dates of medication) it is felt that this is a reasonable step for
the pharmacist to take. Consideration shall be given to the patient assigning its
rights to the pharmacy to undertake the disposal of such medication in the
circumstances. The removal of the medication (either by disposal or separate
storage from currently dispensed medication) will assist in ensuring that errors do
not occur.
6. The pharmacy should have in place a procedure for regularly reviewing medication
charts to ensure the pharmacy profile is accurate. This should be in conjunction
with the residential care facility in the event of a patient who is resident within such
facility or, in the case of a community patient, in consultation with the patient’s
general practitioner.
Situation 2
A pharmacy supplies a residential care facility with a DAA which is incorrect. The
wrong medication is administered to the resident by a registered nurse and the resident
has a negative outcome. The error in the medication dispensed in the DAA is due to a
mistake within the pharmacy in that the wrong medication (type or dosage) is
dispensed in the DAA. The error is compounded as the registered nurse failed to
check the medication chart, following instead the pack label or simply administered the
contents of the DAA.
Some practices and protocols which would limit the possibility of the error initially
occurring, and in turn being compounded with the residential care facility include that:
1. In relation to the packaging of the DAA, if the DAA is not packed by a pharmacist,
the packer of the DAA be an accredited pharmacy technician properly supervised
by a pharmacist. The use of unqualified staff who have not been properly trained is
not appropriate professional practice. It is important that the training of the staff
and the policies and procedures in relation to the dispensary using DAAs be
properly documented and that these policies include appropriate checking and
supervision and that the communications of these policies be properly documented
Quality Medication Care Group, School of Medicine, University of Queensland321
(for example, on induction of new staff, the recording of the holding of regular staff
meetings and so). It is also important that records in relation to checking and
supervision in the dispensary be maintained so that, if necessary, the pharmacist is
able to provide objective contemporaneous proof in relation to these matters. This
can be critical not only in the context of a civil claim arising out of the error, but also
in respect of any disciplinary proceedings.
2. The pharmacist must check the prepared DAA by sighting the original packaging
from which the medication was taken. The check must be a thorough check to
ensure that the proper medication is being supplied in accordance with that which
has been prescribed (both in terms of medication type and dosage). It is again
important that the checking procedures be properly set out in the pharmacists
written policies and protocols and that there are records maintained of the regular
communication and enforcement of these policies and protocols in the pharmacy.
The communication of these matters to staff should also be recorded (for example,
minutes of regular staff meetings new employees acknowledging receiving and
reading the pharmacy’s policies and protocols, and so on).
3. Particular care needs to be given in relation to medication which is to be taken
weekly rather than daily. Dispensing errors involving DAAs not infrequently involve
the inclusion in the DAA of medications which are not to be taken daily but which
are wrongly included as part of daily medication in the DAA. The computer systems
used in packing the DAAs need to be correctly configured in relation to the way that
weekly medication is supplied. More importantly, however, no system can be an
effective substitute for a proper checking being undertaken by the pharmacist to
ensure that such errors do not occur. The importance of this cannot be overstated
in relation to dispensing medications of a narrow therapeutic index, where the
consequences of an error can be fatal.
4. The pharmacist having checked the DAA for delivery to the facility keeps a signed
record of the DAA having been checked by the pharmacist.
5. To prevent the error being compounded by the registered nurse, appropriate
protocols and procedures must be in place within the residential care facility to
ensure that the registered nurse checks the medication chart and records that the
medication chart has been checked.
Situation 3
A pharmacy supplies a community patient with a DAA that contains an error. The
community patient takes the medication according to the pack and has a negative
outcome. The error occurs because the patient’s profile (from which the pharmacy
prepares the pack) does not reflect the patient’s current medication regime. This is
either due to the failure of the patient/doctor to adequately communicate medication
changes to the pharmacy or the failure of the pharmacy to update the patient’s profile
and reflect any communicated changes in the DAA.
Some protocols and procedures which could prevent the error occurring would include:
1. The patient, doctor and pharmacist have a written agreement setting out clearly the
responsibility in respect of communicating changes in the patient’s medication to
the pharmacy and how those changes are to be communicated. In this regard, it is
important that any such agreement be properly explained to the patient and the
patient not feel pressured into entering into any agreement. It would be useful to
Quality Medication Care Group, School of Medicine, University of Queensland322
include the patient’s carer or a close relative in the discussions about the
agreement so that matters are fully understood.
2. The pharmacy maintain proper written records of all communications from the
patient/doctor about medication changes. Most pharmacy software packages are
capable of having such entries separately recorded in relation to patients and a
central storage system of all data in relation to a patient is encouraged. This would
be an effective protection to the pharmacy in the event about there being a dispute
in respect of the communication of a medication change.
3. Where the medication change involves the substitution of one drug for another, this
should be unequivocally stated in a direction from the patient and the doctor. If
there is a written agreement in place between the doctor and the pharmacist, it
would be important to include this in the agreement and how such medication
changes are to be communicated. If the pharmacist is in any doubt as to the
medication regime, appropriate enquiries should be made of the prescriber of the
medication. If medication previously supplied is to be ceased, all supplies of that
medication held on behalf of the patient should be removed and either separately
stored for the patient (again keeping a careful note of expiry dates) or appropriately
disposed of to prevent an error occurring. Again, in the situation where the
prescriber is altering a patient’s medication regime on a regular basis, separate
storage from the regularly supplied medication may be more appropriate than
disposal but it is crucial that appropriate protocols are maintained and recorded. It
would be critical, in the case where a patient ceases to use DAAs for the packaging
of their medication and arrangements are made to collect their medication from the
pharmacy (either because they move to a different residential facility or into the
community) that the previously supplied medication is not given to them (or is only
given after appropriate counselling and advice) to ensure that errors in the taking of
the medications do not occur.
4. It is important in the supply of all medications to community patients that proper and
effective counselling be given to ensure that the patient understands how the
medication is properly to be taken. Often, the counselling process will alert the
pharmacist to an error in the medication being dispensed.
The examples above include elements of matters which have arisen in practice in
relation to the use of DAAs. They reflect the most common issues which arise in a
pharmacy context. However, we include the following as some less typical example of
the types of issues which can arise in a pharmacy involving the supply of medications
packaged in a DAA.10
Situation 4
An elderly patient receives medication packaged in a DAA whilst a resident of a
residential care facility. Upon discharge from that facility to home, the patient attends
the pharmacy which had supplied the DAA and requests that she be supplied with the
balance of her medication which had not been packaged within the DAA. All
medication is the subject of a valid prescription. The medication is given to the patient
who is counselled by the pharmacist in relation to the appropriate use of medication.
The patient’s carer is also present in the pharmacy and the pharmacist forms the view
that the patient understands the directions for use. Shortly thereafter the patient takes
10
We appreciate the assistance of Guild Insurance Limited in allowing us to extract information from claims received in
the preparation of this paper.
Quality Medication Care Group, School of Medicine, University of Queensland323
their own life by an overdose using the medication. In the investigations which follow
there is no adverse finding in relation to the conduct of the pharmacist but the matter,
and its subsequent investigation, is obviously a matter of grave concern for the
pharmacist.
This matter is illustrative of the need for counselling at the time of supply of all
medication and particularly when a patient is discharged from a residential care facility
to their own home and no longer has medication supplied in a DAA.
Situation 5
A matter arose in respect of a complaint involving the supply of previously dispensed
medications to nursing homes, such medications being included within DAAs. The
matter arose out of a routine inspection of the pharmacy by State authorities who
raised concerns about the quantities of previously dispensed and returned medication
in the dispensary (for example, medications in original packs dispensed by other
pharmacies, medicines dispensed to patients who had either died or ceased using
DAAs) The allegations were disputed by the pharmacist who was able to defend the
charges brought against the pharmacy by reference to strictly enforced and
documented protocols and procedures within the pharmacy in relation to compliance
with Commonwealth and State requirements prohibiting the recycling of returned
medications and evidence being able to be brought of the communication of those
procedures to staff in the pharmacy. As a result of these rigorously enforced and
documented procedures, the complaint was dismissed. The matter highlights however
the importance of ensuring appropriate protocols and procedures within a pharmacy,
and documentation of the communication of these policies and procedures to all
employees within the pharmacy.
Situation 6
A matter arose in relation to the labelling of medications which were used in a DAA.
This incident involved the packing of a DAA with medication for patient A. Additional
medication was needed to be dispensed. The pharmacist had however commenced
preparing a DAA for patient B and incorrectly labelled the medication which he was
dispensing for patient A with patient B’s name. The DAAs for both patient A and patient
B were correct and no dosage errors occurred. However, when patient B was
transferred to a different residential care facility, patient B’s medication was supplied to
the new nursing home and it was then discovered that a container of patient A’s
medication, labelled as having been dispensed to patient B, was wrongly included in
that package of medication. Fortunately, there was no incorrect medication ingested
by either patient as the protocols and procedures within the pharmacy in relation to
current treatment sheets and the packaging of DAAs were well established and
followed. This example indicates the ease with which a mistake can occur when DAAs
are being packaged, particularly when a pharmacy is under time pressures in relation
to the supply of the DAA. This matter also raises the need to ensure, when
medications are being transferred by direction either to the patient or to another
residential care facility, that the details of each medication so supplied are carefully
checked.
Quality Medication Care Group, School of Medicine, University of Queensland324
APPENDIX E: THEORETICAL CONSIDERATIONS
IN REPACKING DRUG PRODUCTS INTO A DAA
This appendix provides supplementary information on the various theoretical aspects
pertaining to drug stability in addition to that provided in section 4.3.2.
UNSUITABLE SOLID DOSAGE FORMS
Sublingual and buccal tablets are solid dosage forms used for drug release in the
mouth (under the tongue and in the side of the cheek, respectively) followed by
systemic uptake of the drug. They are often small and porous, the latter facilitating fast
disintegration and drug release (Aulton 2002). Chewable tablets are softer than
conventional tablets as they are designed to be chewed and thus mechanically
disintegrate in the mouth (Allen et al. 2005). Rapidly dissolving tablets (RDT’s) or
Instant-release tablets or wafers are characterised by dissolving in the mouth within
one minute. These tablets are prepared using highly water-soluble excipients designed
to absorb water into the tablet for rapid disintegration or dissolution. They are also quite
friable (Allen et al. 2005). Lozenges are used for local medication in the mouth and
throat (Aulton 2002). The excipients utilised to form the sweetened confectionery-type
base are hygroscopic and the lozenge is likely to become sticky upon exposure to
moisture. Effervescent tablets are designed to undergo rapid dissolution when in
contact with water due to internal liberation of carbon dioxide. The effervescent carbon
dioxide is created by a reaction in water between a carbonate or bicarbonate excipient
and a weak acid excipient such as citric or tartaric acid (Aulton 2002). Dispersible
tablets are designed to disperse (disintegrate and/or dissolve) rapidly in water. These
solid dosage forms by design are softer, more friable and porous in nature and/or
contain excipients that are designed to absorb water and swell resulting in tablet
disintegration and/or contain excipients that are highly hygroscopic. For these reasons,
any exposure to moisture during storage and handling will result in deterioration in
physical integrity of the solid dosage form, including possible changes in appearance
and potential microbial contamination.
MOISTURE-SENSITIVE SOLID DOSAGE FORMS
Ambient relative humidity (RH) (0% desert, 55% temperate and 87% tropics) can vary
widely and continually depending on the weather and air temperature, and these cyclic
changes lead to constant variations in the moisture content of unprotected drug
products. The constant sinusoidal change in day and night temperatures is a major
contributing factor. For this reason pharmaceutical air conditioning is usually set below
50% RH, and very hygroscopic products are manufactured below 40% RH (Aulton
2002). Temperature cycling can also lead to increased condensation in DAAs with
suboptimal seals/ closures, resulting in increased physicochemical stability issues and
also an increased potential for microbial contamination. In use, DAAs may be subjected
to a reasonable degree of handling, during which accidental rupture of nearby blister
seals may occur (Saville 2001), allowing further exposure to humidity.
Water-soluble drugs present in a solid dosage form will dissolve in any moisture which
has been adsorbed on the solid surface and may then be subject to hydrolysis – the
most common form of chemical degradation of drugs in aqueous solution (Allen et al.
Quality Medication Care Group, School of Medicine, University of Queensland325
2005; Aulton 2002; Florence & Attwood 1998). Drugs that are a derivative of carboxylic
acid or contain functional groups based on this moiety (e.g. ester, amide, lactone,
lactam, imide, carbamate) are liable to hydrolytic degradation. Common examples
include: chloramphenicol, ergometrine, benzylpenicillin sodium, nitrazepam and
chlordiazepoxide (Florence & Attwood 1998). In addition to causing chemical
degradation in susceptible actives or excipients, moisture can cause tablets to harden
(or soften) with subsequent effect on disintegration and dissolution behaviour. Certain
drugs and excipients are unstable when exposed to moisture (elevated humidity). For
example, the anhydrous form of carbamazepine (CBZ) has been shown to transform to
a more thermodynamically stable, less soluble, dihydrate form when in contact with
moisture. Studies have shown that CBZ tablets, on exposure to humidity, harden and
may dissolve poorly and may lose as much as one-third of their potency (Al-Zein et al.
1999).
It is essential that the packaging material and closure of the DAA offers sufficient
moisture protection for moisture-sensitive solid dosage forms. The USP describes the
various packaging materials commonly utilised in DAAs and refers to them broadly as
having nominal, medium, high, and extreme barrier properties (USP <1146> 2005).
The USP also describes how to determine and classify moisture permeation rates and
states that if the manufacturer’s labelling includes “Protect from Moisture”, the
repackager shall utilise a high barrier film (USP <671> 2005). A repackager is detailed
in the USP (USP <1146> 2005) as “an establishment that repackages drugs and sends
them to a second location in anticipation of need. Repackaging firms repackage
preparations for distribution (e.g., for resale to distributors, hospitals or other
pharmacies), a function that is beyond the regular practice of a pharmacy. Distribution
is not patient specific in that there are no prescriptions. Unlike dispensers
(pharmacies), repackaging firms are required to register with the FDA and to comply
with the current Good Manufacturing Practice regulations in 21 CFR 210 and 211”,
where repackaging is “the act of removing a preparation from its original primary
container and placing it into another primary container, usually of smaller size” (USP
<1146> 2005) and package “is the term synonymous with that of the term ‘container’”
(USP <1146> 2005).
SOLID DOSAGE FORMS SENSITIVE TO AIR (OXIDATION)
Drug substances and/or excipients that are subject to oxidative degradation may
undergo a greater degree of degradation when packaged in plastic (gas-permeable) as
opposed to glass (Allen et al. 2005). Oxidation involves the removal of an
electropositive atom, radical or electron, or addition of an electronegative atom or
radical. The molecular structures most likely to oxidise are those with a hydroxyl group
directly bonded to an aromatic ring (e.g. phenol derivatives such as catecholamines
and morphine), conjugated dienes (e.g. vitamin A and unsaturated free fatty acids),
heterocyclic aromatic rings, nitroso and nitrite derivatives and aldehydes (e.g.
Thank you for your continued participation in the Pharmacy Guild of Australia Dose
Administration Aids (DAAs) study run by the University of Queensland/ Princess
Alexandra Hospital. We are pleased to inform you that this is the last stage of this
study which has been a great success thanks to your time and effort.
We are asking you to:
read the summary of the recommendations to improve DAA services,
to complete the enclosed survey, and
return it in the reply paid envelope to the study office.
We expect that the survey will take approximately 15 minutes. The purpose of this
survey is to find out what you think and how you would be affected by our
recommendations to improve the way DAAs are provided. To show our appreciation for
you continued support we will send you a $20 Coles Myer gift card if you return the
completed survey by the 10th of August if at all possible. If you have any questions
please call the study office (free of charge) on 1800 555 803.
Thanks again for your assistance and we look forward to receiving your comments.
James Leslie
Project Officer
COMMUNITY PATIENT BEST PRACTICE ISSUES SUMMARY
AND SURVEY
Recommendations to improve the provision and use of Dose
Administration Aids (DAAs) in the community
Introduction
Dose administration aids (DAAs) are devices packed by your pharmacy to help
organise your medicines. The provision and use of DAAs is a complex process
involving a number of steps and the input and collaboration of different parties (the
pharmacy, doctors, patients, carers and sometimes community support services such
as community nurses).
334
Through consultation with pharmacists, doctors and patients using DAAs we have
come-up with a number of recommendations to provide a range of strategies and tools
to ensure that DAAs are:
safe (DAA is packed without errors),
effective (help patients manage their medicines at home) and
efficient (minimise the costs to pharmacy and to the patient).
Recommendations
The key recommendations that may impact on you as a community patient include:
1. That all community patients be formally assessed to ensure that they are able
to use a DAA and that they would benefit from this service before it is
provided by the pharmacy.
This assessment would involve documenting:
(1) Your knowledge and understanding of your medicines
(2) How a DAA might help you manage your medicine
(3) Whether you might have any problems using a DAA, and
(4) What type would be best for you?
2. That community patients sign an agreement with their community pharmacy
before being provided with a DAA.
Before signing the agreement, the pharmacy would spend some time explaining
what is involved in having a DAA packed by the pharmacy and how to use the DAA.
The purpose of this agreement is to ensure that patients and doctors receive all the
information they need about how this service will work and what they will have to do
to make sure that the DAA is available on time and contains your current
medicines. The sort of things that would be covered by this agreement includes:
What type of DAA will you get and how many packs you will need?
How long will each pack last (weekly, fortnightly, monthly)?
What medicines will be packed (prescription, over-the-counter medicines,
vitamins and supplements?)
Will you pick it up or will the pharmacy deliver, when and how?
What is the cost of this service and who should the pharmacy charge?
Your consent to sharing information about your medicines between the
pharmacy, GPs, carers and hospital if required?
Who will tell the pharmacy about any changes to your medicines (you or your
doctor) and how will the pharmacy be notified of the changes?
Will the pharmacy keep your prescriptions at the pharmacy or will you keep
them at home?
Will the pharmacy remind you to visit the doctor when you need a new
prescription or will the pharmacy request prescriptions from your doctor on your
behalf?
What you should do if you drop/lose a tablet or if you open the wrong blister?
What to do if you go away (e.g. holidays or to hospital).
What will happen and what should you do if you have a prescription filled at a
different pharmacy (i.e. a hospital pharmacy).
How the pharmacy will provide you with information about your medicines?
Your consent to re-assess how you are managing with your DAA after a
specified time (e.g. 6 months).
335
3. The community pharmacy should provide you with a written medicine
template (see example on page 3) that is a current and accurate record of all
the medicine you are using (those packed in the DAA as well as non-packed
medicine).
You should carry this sheet with you and present it to your GP and/or any other
doctors and the hospital. Any medicine changes can be written on the template by
the doctor so that you can show this to your pharmacy. The pharmacy should then
update their records and print you a new sheet to reflect your new medicines. The
pharmacy would use this information provided on the form to help ensure your
current medicines are packed in your DAA.
4. The doctor should put any medicine changes in writing for community
patients using a DAA to pass on to their pharmacy (e.g. using the written
medicine template).
5. A fair amount should be paid for DAA services as they take some time to
prepare and cost the pharmacy a lot to provide.
6. Community patients using a DAA should see their GP at least once every
three months (even if it is just to have a prescription renewed).
7. Community patients using DAAs should return the used DAA to the
pharmacy containing any unused medicines so that the pharmacy can
confirm that they are using the DAA correctly.
8. The pharmacist and doctor should regularly (6 monthly) review the currency
of the patient’s medicines and the community patient’s ability to use a DAA.
336
Example patient held template (size reduced)
337
Survey Questions
1. Did your pharmacy assess your ability to use a DAA before providing this service
(e.g. give you a sample DAA to try)? Yes No
Please describe any assessment you may have had:
2. Do you currently have a formal agreement with you pharmacy detailing what is
involved in your DAA service? Yes No
3. Would you be willing to sign an agreement like the one explained in
Recommendation 2? Yes No
4. Please tell us which parts of the agreement you would like to change or leave out
(and why)?
5. Thinking back to when you started your DAA, how helpful would it have been to
have your pharmacist discuss all the topics covered in Recommendation 2?
Extremely helpful Some what helpful Not at all helpful
6. Which of the issues covered in Recommendation 2 did your pharmacist discuss
with you before you started using a DAA?
7. Below is a list of tasks that are required to ensure that DAAs are packed accurately
and to make sure that you don’t run out of medicines. For each task, please
indicate who (e.g. you, your carer, your doctor, your pharmacy or a community
nurse) looks after this task currently and who you believe should look after this task
(you may indicate more than one person). Task Who currently does
this task?
Who should do this
task?
1. Tell the pharmacy about any changes
to your medicines.
2. Store the repeat prescriptions
3. Ask the doctor for a new prescription
when you have no more repeats
8. Does your pharmacy currently provide a full and complete list of your medicines to
you? Yes No
9. Would a medicine template as explained in Recommendation 3 and like the
example on page 3 be useful/helpful to you? Yes No
Why or why not?
10. Would you be willing to carry a medicine information sheet with you when you are
going to your doctors, pharmacy or to the hospital? Yes No
11. Does your doctor currently communicate medicine changes (i.e. ceased medicine
or changes in doses) to you in writing? Yes No
12. Would you consider not having medicines packed in a DAA by your pharmacy if the
price increased? Yes No
13. At what price would you no longer be able to continue to get this service? _ _
per/week
14. Do you currently see your GP at least once every three months (even if it is just to
have a prescription renewed)? Yes No
338
15. Would you be willing to return your used DAA to the pharmacy, including any
unused medicines so that the pharmacy can confirm that you are using the DAA
correctly? Yes No
If No, why not?
16. Would you be happy for your community pharmacist/ and or GP to conduct regular
reviews (e.g. 6 monthly) of your medicine management? This review may involve a
visit to your home? Yes No
If No, why not?
What other comments would you like to make about the recommendations listed on
pages 1 and 2?
17. Please comment on any problems you have experienced with DAA services. Do
you have any suggestions for improving the way DAA services are currently
provided?
Please place this completed survey in the reply paid envelope provided.
Thank you for you assistance with this study.
339
COMMUNITY PHARMACIST BEST PRACTICE FEEDBACK
SURVEY
Pharmacist Feedback on “BEST PRACTICE RECOMMENDATIONS FOR THE PROVISION
AND USE OF DOSE ADMINISTRATION AIDS IN THE COMMUNITY”
The following questions are intended to give structure to the feedback we are requesting from you. The questionnaire is 4 pages, reflecting the pharmacy’s central role in providing DAAs. You may need to refer to the “best practice” document to complete the questions We have targeted some specific areas of the best practice model where we would appreciate your specific comments. We are, however, interested in any other comments about the model, it’s potential advantages and disadvantages, its feasibility and the impacts on your practice were the model to be adopted.
1. For recommendation 1.3.1 on assessing patients’ need for a DAA (see also Figure 1.2):
a) How feasible are recommendations related to structured patient assessment? If not feasible, please indicate why.
b) Currently, do you consult the patient’s GP when making the decision to initiate DAA services for a community patient? Yes No?
c) What if any, are the advantages and disadvantages of the community pharmacy consulting with the GP prior to initiating DAA services for a community patient?
d) What changes would you need to make to your DAA initiation process to adopt this ‘best practice’ model? What are the cost, staffing and other impacts?
e) How could Recommendation 1.3.1 be improved?
2. Would you be willing have a written tripartisan agreement as specified in 1.3.2 prior to your community pharmacy beginning to provide a DAA service? Yes No
3. Given that the purpose of this agreement is to ensure that all parties are aware of what is involved in DAA services and can discuss preferences for how to carry out the requirements of providing DAAs prior to beginning this service, which aspects of this agreement would you consider important/useful?
4. Which aspects of this agreement would you change or leave out and why?
5. How feasible is recommendation 1.3.2? What are the likely cost, staffing and other impacts on your pharmacy were you to adopt this ‘best practice’ model?
6. Below is a list of tasks that are required to ensure that DAAs are packed accurately and provided on time. For each task, please indicate who (e.g. the patient, doctor, surgery staff, pharmacy or a community nurse) currently looks after this task and who you believe shouldbe responsible for this task (you may indicate more than one person).
Task Done now by? Should do it?
Recommend that a patient use a DAA Assess a patient’s ability to use a DAA Ensure that the pharmacy has a current and complete record of a patient’s medicines Decide what should be packed in the DAA Decide what is the optimal schedule for the patient (times of day for each medicine) Store the patient’s repeat prescriptions Provide the patient with a current and complete written record of their medicines Tell the pharmacy about any changes to a patient’s medicines Making sure prescriptions are available to ensure continuity of supply of packed medicines Inform the doctor when a repeat prescription is required Assess how patients are managing with the DAA
340
7. For Recommendation 1.3.3, the use of a patient held template to facilitate communication medication changes to the pharmacy:
a) Would a patient held medicine template (as per Recommendation 1.3.3) be useful/helpful to you in your pharmacy using DAAs? Yes No
Why or why not?
b) Would a patient held medicine template (as per Recommendation 1.3.3) be useful/helpful to your community patients using DAAs? Yes No
Why or why not?
c) What would be the advantages and disadvantages of doctors approving a patient’s medication template/pharmacy packing profile prior to the pharmacy packing the patients medicines into a DAA?
d) What changes would you need to make to your DAA service processes? What are the likely cost, staffing and other impacts?
e) Overall, how feasible is the recommendation? Do you see this process as beneficial in promoting safe, effective and efficient DAA services? In your view, what are the barriers to adopting the patient template and how might these be overcome?
f) How could Recommendation 1.3.3 be improved?
8. For Recommendation 1.3.4, keeping a record of communication about medication changes:
a) How feasible is the recommendation? Will it be beneficial in promoting safe, effective and efficient DAA services? Do you believe it will act as legal risk reduction?
b) What changes would you need to make to your medication change communication process? What are the cost, staffing and other impacts?
c) How could Recommendation 1.3.4 be improved?
9. How often do you receive a discharge summary, when a patient using a DAA is discharged from hospital? Never Rarely Sometimes Mostly Always
10. For Recommendation 1.3.5, about continuity of care between hospitals and the community:
a) Do you see the model in Figure 1.3 as beneficial in promoting safe, effective and efficient DAA services? Is this feasible? If not, please indicate why.
b) What changes would you need to make to your processes related to hospitalisation of patients? What are the cost, staffing and other impacts?
c) Does the recommendation 1.3.5 and model in Figure 1.3 address your reservations about packing a DAA for a community patient based on discharge information? If not, why?
d) How could Recommendation 1.3.5 be improved? Think about any circumstances that might complicate or delay the processes.
11. For Recommendation 1.3.6, about Quality Control (QC) and Quality Assurance (QA):
a) Do you believe that record keeping and monitoring can be beneficial in promoting safe, effective and efficient DAA services, and be a form of legal risk-reduction? If not, why?
b) What changes would you need to make to your QC & QA processes? What are the cost, staffing and other impacts?
c) How could Recommendation 1.3.6 be improved? What process guides, tools or other support would you see as helpful to you in adopting recommendation 1.3.6?
12. Would you be willing to check concordance between the pharmacy packing profile, the GP’s records and the patient’s reports 6-monthly as part of a DAA service? Yes No
If No, why not?
13. Do you believe that the recommendations related to prescription management, packing procedures and roles for staff (see section 1.3.7)
a) Promote safe, effective and efficient DAA services? If not, why?
b) What are the cost, staffing and other impacts?
341
c) How could recommendations in section 1.3.7 be improved?
14. Should a ‘best practice’ model include recommendations about fair payment? Yes No
15. What do you believe would be the impacts of negotiating a realistic price for the provision of DAAs to appropriate patients?
16. Is Recommendation 1.3.9 feasible? If not, why? What barriers would need to be overcome?
17. What changes would you need to make to your DAA processes were you to adopt this recommendation 1.3.9? What are the likely cost, staffing and other impacts?
18. This ‘best practice’ document has recognised the role of the General Practitioner in a number of processes.
a) Overall, to what extent are the GPs you interact with in providing DAAs to community patients already following best practice processes?
Never Rarely Sometimes Mostly Always
Comments?
b) How feasible is the recommendations involving GPs? How central do you see GP roles in ‘best practice’ to the delivery of safe, effective and efficient DAA services? What, if anything, would you change in the ‘best practice’ document to improve the interaction between community pharmacies providing DAAs and GPs providing care to community patients?
19. What do you believe would be the main barriers and implications to implementing these recommendations (1.3.1 to 1.3.9) in your DAA service?
20. What strategies or support (e.g. from pharmacy peak bodies, etc) do you believe would be necessary lead to widespread adoption of this best practice model?
21. Please comment on any other problems you have experienced with DAA services that are not addressed by these recommendations. Do you have any other suggestions for improving the way DAA services are currently provided?
Pharmacist Feedback on “BEST PRACTICE RECOMMENDATIONS FOR THE PROVISION AND USE OF DOSE ADMINISTRATION AIDS IN RESIDENTIAL CARE FACILITIES (RCFs)”
The following questions are intended to give structure to the feedback we are requesting from you. The questionnaire is 4 pages, reflecting the pharmacy’s central role in providing DAAs in RCFs. You may need to refer to the “best practice” document to complete the questions We have targeted some specific areas of the best practice model where we would appreciate your specific comments. We are, however, interested in any other comments about the model, it’s potential advantages and disadvantages, its feasibility and the impacts on your practice were the model to be adopted.
1. After considering the steps involved in a DAA service (Figure 1.1) and the implementation model (Figure 1.2):
a) How feasible are the implementation model and recommendations related to tendering for a DAA service? If not feasible, please indicate why.
b) What changes would you need to make to your tendering process? What are the cost, staffing and other impacts?
c) How feasible are the implementation model and recommendations related to monitoring and quality assurance for DAA service quality within the pharmacy and the RCF, and by external assessors? If not feasible, please indicate why.
d) What changes would you need to make to address QA for pharmacy DAA processes? What are the cost, staffing and other impacts?
e) How could Recommendation 1.3.1 be improved?
2. Would you be willing have a written agreement as specified in 1.3.2 prior to your community pharmacy beginning to provide a DAA service to a facility? Yes No
342
3. Given that the purpose of this agreement is to ensure that all parties are aware of what is involved in DAA services and can discuss preferences for how to carry out the requirements of providing DAAs prior to beginning this service, which aspects of this agreement would you consider important/useful?
4. Which aspects of this agreement would you change or leave out and why?
5. Below is a list of tasks that are required to ensure that DAAs are packed accurately and provided on time. For each task, please indicate who (e.g. facility staff, doctor, Dr’s surgery staff, or pharmacy) currently looks after this task (choose a representative RCF) and who you believe should be responsible for this task (you may indicate more than one person).
Task Who does it now?
Who should do it?
Assess a self-medicating residents ability to use a DAA Ensure that the pharmacy has a current and complete record of residents’ medicines Decide what should be packed in the DAA Decide what is the optimal schedule for a resident (times of day for each medicine) Store the residents’ repeat prescriptions Tell the pharmacy about any changes to residents’ medicines Making sure prescriptions are available to ensure continuity of supply of packed medicines Inform the doctor when a repeat prescription is required
6. For Recommendation 1.3.3, the process of communicating medication changes to the pharmacy in writing AND keeping a record of that communication:
a) How feasible is the recommendation? Do you see this process as beneficial in promoting safe, effective and efficient DAA services?
b) What changes would you need to make to your medication change communication process? What are the cost, staffing and other impacts?
c) How could Recommendation 1.3.3 be improved?
7. How often do you receive a discharge summary, when a resident using a DAA is discharged from hospital? Never Rarely Sometimes Mostly
Always
8. For Recommendation 1.3.4, about continuity of care between hospitals and RCFs:
a) Do you see the model in Figure 1.3 as beneficial in promoting safe, effective and efficient DAA services? Is this feasible? If not, please indicate why.
b) What changes would you need to make to your processes related to hospitalisation of residents? What are the cost, staffing and other impacts?
c) Does the recommendation 1.3.4 and model in Figure 1.3 address your reservations about packing a DAA for an RCF resident based on discharge information? If not, why?
d) How could Recommendation 1.3.4 be improved? Think about any circumstances that might complicate or delay the processes.
9. For Recommendation 1.3.5, about Quality Control (QC) and Quality Assurance (QA):
a) Do you believe that record keeping an monitoring can be beneficial in promoting safe, effective and efficient DAA services, and be a form of legal risk-reduction? If not, why?
b) What changes would you need to make to your QC & QA processes? What are the cost, staffing and other impacts?
c) How could Recommendation 1.3.5 be improved? What process guides, tools or other support would you see as helpful to you in adopting recommendation 1.3.5?
343
10. Would you be willing to check concordance between the pharmacy packing profile, the GP’s records and the residents’ chart 6-monthly as part of a DAA service? Yes No
If No, why not?
11. Do you believe that the recommendations related to prescription management, packing procedures and roles for staff (see section 1.3.7)
a) Promote safe, effective and efficient DAA services? If not, why?
b) What are the cost, staffing and other impacts?
c) How could recommendations in section 1.3.7 be improved?
12. This ‘best practice’ document has recognised the role of the General Practitioner in a number of processes.
a) Overall, to what extent are the GPs you interact with in providing DAAs to RCFs already following best practice processes?
Never Rarely Sometimes Mostly Always
Comments?
b) How feasible is the recommendations involving GPs? How central do you see GP roles in ‘best practice’ to the delivery of safe, effective and efficient DAA services? What, if anything, would you change in the ‘best practice’ document to improve the interaction between community pharmacies providing DAAs and GPs providing care to RCF patients?
13. What do you believe would be the main barriers and implications to implementing these recommendations (1.3.1 to 1.3.8) in your DAA service?
14. What strategies or support (e.g. from Aged Care, pharmacy peak bodies, etc) do you believe would be necessary lead to widespread adoption of this best practice model by the Residential Aged Care Sector?
15. Please comment on any other problems you have experienced with DAA services that are not addressed by these recommendations. Do you have any other suggestions for improving the way DAA services are currently provided?
Thank you for you assistance with this study.
HOSPITAL PHARMACISTS FEEDBACK QUESTIONS
Hospital Pharmacist Feedback on “BEST PRACTICE RECOMMENDATIONS FOR THE PROVISION AND USE OF DOSE ADMINISTRATION AIDS IN THE COMMUNITY” and “BEST PRACTICE RECOMMENDATIONS FOR THE PROVISION AND USE OF DOSE
ADMINISTRATION AIDS IN RESIDENTIAL CARE FACILITIES (RCFs)”
The following questions are intended to give structure to the feedback we are requesting from you. You may need to refer to the “best practice” document to complete the questions We have targeted some specific areas of the best practice model where we would appreciate your specific comments. We are, however, interested in any other comments about the model, it’s potential advantages and disadvantages, its feasibility and the impacts on your practice were the model to be adopted.
1. For recommendation 1.3.1 on assessing community patients’ need for a DAA (also Figure 1.2):
a) When a community patient is discharged with a recommendation that a DAA be started, is it feasible for the hospital pharmacists to undertake structured patient assessment? If not feasible, please indicate why.
b) Currently, do you consult the patient’s GP when making the decision to initiate DAA services for a community patient? Yes No?
344
c) Currently, do you consult the patient’s usual pharmacy when making the decision to initiate DAA services for a community patient? Yes No?
d) What changes would you need to make to your DAA initiation process to adopt this ‘best practice’ model? What are the cost, staffing and other impacts?
e) How could Recommendation 1.3.1 be improved?
2. The community model recommends a written tripartisan agreement as specified in 1.3.2. Given that the purpose of this agreement is to ensure that all parties are aware of what is involved in DAA services and can discuss preferences for how to carry out the requirements of providing DAAs prior to beginning this service, which aspects of this agreement would you consider important/useful?
3. Which aspects of this agreement would you change or leave out and why?
4. For Recommendation 1.3.3 in the community model, the use of a patient held template to facilitate communication medication changes to the pharmacy:
a) Would a patient held medicine template (as per Recommendation 1.3.3) be useful/helpful to you in your hospital pharmacy practice? Yes No
b) Why or why not?
c) What changes would need to be made to the recommendation to improve how useful such a the template would be to you?
d) Overall, how feasible is the recommendation? Do you see this process as beneficial in promoting safe, effective and efficient DAA services? In your view, what are the barriers to adopting the patient template and how might these be overcome?
5. How often do you send a discharge summary, when a community patient using a DAA is discharged from hospital?
Never Rarely Sometimes Mostly Always
6. How often do you receive a discharge summary, when a resident using a DAA is discharged from hospital? Never Rarely Sometimes Mostly Always
7. For Recommendations 1.3.5 (community) and 1.3.4 (RCF model), about continuity of care between hospitals and the community:
a) Do you see the model in Figure 1.3 as beneficial in promoting safe, effective and efficient DAA services? Is this feasible? If not, please indicate why.
b) What changes would you need to make to your processes related to hospitalisation of patients? What are the cost, staffing and other impacts?
c) To what extent do the best practice model documents (and model in Figure 1.3) address your reservations about using medication regimen information from RCFs or from the DAA pack or patient held template? If not, why?
d) How could Recommendation 1.3.5/1.3.4 be improved? Think about any circumstances that might complicate or delay the processes.
8. These ‘best practice’ documents have recognised the role of the General Practitioner in a number of processes. How feasible is the recommendations involving GPs? How central do you see GP roles in ‘best practice’ to the delivery of safe, effective and efficient DAA services?
9. What do you believe would be the main barriers and implications to implementing these recommendations (1.3.1 to 1.3.9, where relevant for community and RCF patients) in your practice?
10. Do you think that these ‘best practice’ models compliment and integrate with other hospital-based initiatives related to continuity of care? Why or why not?
11. What strategies or support (e.g. from pharmacy peak bodies, etc) do you believe would be necessary lead to widespread adoption of this best practice model?
12. Please comment on any other problems you have experienced with DAA services that are not addressed by these recommendations. Do you have any other suggestions for improving the way DAA services are currently provided?
345
EXPERT PANEL QUESTIONS
Dear Expert Panel member,
Re: Effectiveness and cost-effectiveness of DAAs – Phase 3 feedback on models
We have prepared two ‘best practice’ models (one for community patients and one for RCFs), attached with this email (together with some of the supporting tools). We hope that you will be able to give us feedback on the models. The documents are detailed, and while we hope you will comment on both, we understand if you concentrate on only one of the documents.
These models have arisen out of broad consultation including consultation with RCF management and staff, pharmacy management and staff, hospital pharmacy staff, peak pharmacy bodies and the Phase 2 GP survey and community patients. We have also sought a legal opinion on aspects of the model as they relate to legal risk. On the question of stability of medications in DAAs, we have sought input from TGA and colleagues from Griffith and James Cook University. We will be undertaking more consultation based on these draft models.
The documents are written from a multidisciplinary perspective but as a community pharmacy has a key role in service provision, many recommendations relate to pharmacy.
We are interested in comments about the models, their potential advantages and disadvantages, their feasibility and the impacts on practice were the models to be adopted.
The following questions are intended to give some structure to the feedback we are requesting from you. You may need to refer to the “best practice” document to complete the questions. We have targeted some specific areas of the best practice model where we would appreciate your specific comments. You will note that certain of the recommendations are almost the same in the two model documents (as they reflect core activities), so you may wish to make a single comment that that covers the sections common to both models (e.g. Community model recommendations 1.3.6 & 1.3.7 and RCF model recommendations 1.3.6 & 1.3.7 ). For simplicity, we have included separate questions.
Part 1: Feedback on “BEST PRACTICE RECOMMENDATIONS FOR THE PROVISION AND
USE OF DOSE ADMINISTRATION AIDS IN THE COMMUNITY”
1. For recommendation 1.3.1 on assessing patients’ need for a DAA (see also Figure 1.2): f) How feasible are recommendations related to structured patient assessment? If not
feasible, please indicate why. What are the cost, staffing and other impacts? g) How could Recommendation 1.3.1 be improved? Note that as more research on criteria
for patient selection becomes available, the model can be updated.
2. Recommendation 1.3.2 covers a written tripartisan agreement prior to a community pharmacy beginning to provide a DAA service. a) Given that the purpose of this agreement is to ensure that all parties are aware of what
is involved in DAA services and can discuss preferences for how to carry out the requirements of providing DAAs prior to beginning this service, which aspects of this agreement would you consider important/useful?
b) Which aspects of this agreement would you change or leave out and why? c) How feasible is recommendation 1.3.2? What are the likely cost, staffing and other
impacts on pharmacies were this ‘best practice’ model to be adopted?
3. For Recommendation 1.3.3, the use of a patient held template to facilitate communication medication changes to the pharmacy: a) Overall, how feasible is the recommendation? Do you see this process as beneficial in
promoting safe, effective and efficient DAA services? In your view, what are the barriers to adopting the patient template and how might these be overcome?
b) Do you believe a patient held medicine template (as per Recommendation 1.3.3) would be useful/helpful to: community patients using DAAs? Yes No pharmacies providing DAAs? Yes No GPs caring for patients who use a DAA? Yes No
346
c) What would be the advantages and disadvantages of doctors approving a patient’s medication template/pharmacy packing profile prior to the pharmacy packing the patients medicines into a DAA?
d) If not already mentioned above, what are the likely cost, staffing and other impacts? How could Recommendation 1.3.3 be improved?
4. For Recommendation 1.3.4, keeping a record of communication about medication changes: a) How feasible is the recommendation? Will it be beneficial in promoting safe, effective
and efficient DAA services? Do you believe it will act as legal risk reduction? b) What are the cost, staffing and other impacts? How could Recommendation 1.3.4 be
improved?
5. For Recommendation 1.3.5, about continuity of care between hospitals and the community: a) Do you see the model in Figure 1.3 as beneficial in promoting safe, effective and
efficient DAA services? Is this feasible? If not, please indicate why. b) What are the cost, staffing and other impacts? How could Recommendation 1.3.5 be
improved?
6. For Recommendation 1.3.6, about Quality Control (QC) and Quality Assurance (QA): a) Do you believe that record keeping and monitoring can be beneficial in promoting safe,
effective and efficient DAA services, and be a form of legal risk-reduction? If not, why? b) What are the cost, staffing and other impacts? If applicable to you, what changes would
you need to make to your QC & QA processes? c) How could Recommendation 1.3.6 be improved? What process guides, tools or other
support would you see as helpful to you in adopting recommendation 1.3.6?
7. Should a DAA service include a 6-monthly concordance check between the pharmacy packing profile, the GP’s records and the patient’s reports? Yes No If No (or conditional answer) why not?
8. Do you believe that the recommendations related to prescription management, packing procedures and roles for staff (see section 1.3.7) a) Promote safe, effective and efficient DAA services? If not, why? b) What are the cost, staffing and other impacts? c) How could recommendations in section 1.3.7 be improved?
9. Should a ‘best practice’ model include recommendations about fair payment? Yes No
10. What do you believe would be the impacts of negotiating a realistic price for the provision of DAAs to appropriate patients?
11. Is Recommendation 1.3.9 feasible? If not, why? What barriers would need to be overcome?
12. What are the likely cost, staffing and other impacts of recommendation 1.3.9?
13. This ‘best practice’ document has recognised the role of the General Practitioner in a number of processes. How feasible is the recommendations involving GPs? How central do you see GP roles in ‘best practice’ to the delivery of safe, effective and efficient DAA services? What, if anything, would you change in the ‘best practice’ document to improve the interaction between community pharmacies providing DAAs and GPs providing care to community patients?
Part 2: Feedback on “BEST PRACTICE RECOMMENDATIONS FOR THE PROVISION AND
USE OF DOSE ADMINISTRATION AIDS IN RESIDENTIAL CARE FACILITIES (RCFs)”
1. After considering the steps involved in a DAA service (Figure 1.1) and the implementation model (Figure 1.2): a) How feasible are the implementation model and recommendations related to tendering
for a DAA service? If not feasible, please indicate why.
347
b) How feasible are the implementation model and recommendations related to monitoring and quality assurance for DAA service quality within the pharmacy and the RCF, and by external assessors? If not feasible, please indicate why. What are the cost, staffing and other impacts?
c) How could Recommendation 1.3.1 be improved?
2. Recommendation 1.3.2 covers a written agreement prior to a community pharmacy beginning to provide a DAA service to an RCF: a) Given that the purpose of this agreement is to ensure that all parties are aware of what
is involved in DAA services and can discuss preferences for how to carry out the requirements of providing DAAs prior to beginning this service, which aspects of this agreement would you consider important/useful?
b) Which aspects of this agreement would you change or leave out and why? c) How feasible is recommendation 1.3.2? What are the likely cost, staffing and other
impacts on pharmacies were this ‘best practice’ model to be adopted?
3. For Recommendation 1.3.3, the process of communicating medication changes to the pharmacy in writing AND keeping a record of that communication: a) How feasible is the recommendation? Do you see this process as beneficial in
promoting safe, effective and efficient DAA services? b) How could Recommendation 1.3.3 be improved? What are the cost, staffing and other
impacts?
4. For Recommendation 1.3.4, about continuity of care between hospitals and the RCF (note that the needs of an RCF for a legal order to administer are included in the model: a) Do you see the model in Figure 1.3 as beneficial in promoting safe, effective and
efficient DAA services? Is this feasible? If not, please indicate why. b) What are the cost, staffing and other impacts? How could Recommendation 1.3.4 be
improved?
5. Recommendation 1.3.5 in the RCF model is similar to 1.3.5 in the community model. The difference relates to educating RCF staff about medication storage and involving them in capturing evidence of drug instability. How feasible is the recommendation?
6. How feasible is a 6-monthly concordance check between the pharmacy packing profile, the RCF charts and the GP’s records suggested in Recommendation 1.3.6? What are the cost, staffing and other impacts?
Recommendation 1.3.7 in the RCF model essentially the same as 1.3.7 in the community model.
Overall Comments on the models
Of all the issues you have raised in response to the preceding questions, what do you believe would be the main barriers and implications to implementing these recommendations (1.3.1 to 1.3.9) in a DAA service for an RCF? And the main barriers/implications for community patients?
What strategies or support (e.g. from pharmacy peak bodies, etc) do you believe would be necessary lead to widespread adoption of this best practice model?
Any other comments?
348
GENERAL PRACTITIONER FEEDBACK QUESTIONS
BEST PRACTICE RECOMMENDATIONS FOR THE PROVISION AND USE OF DOSE
ADMINISTRATION AIDS IN THE COMMUNITY
1. In current practice are you consulted in the decision to initiate DAA services for a community patient? Yes No
2. From your perspective, what would be the benefits of the community pharmacy consulting with you prior to initiating DAA services for a community patient?
3. Please describe how you foresee your role, as a GP, in a structured assessment of a community patient (as per Recommendation 1.3.1) prior to starting DAA services?
4. Would you be willing to be involved in a tripartisan agreement as specified in 1.3.2 prior to a community patient beginning a DAA service? Yes No
5. Given that the purpose of this agreement is to ensure that all parties are aware of what is involved in DAA services and can discuss preferences for how to carry out the requirements of providing DAAs prior to beginning this service, which aspects of this agreement would you consider important/useful?
6. Which aspects of this agreement would you change or leave out and why?
7. Under what circumstances would you be willing to write a prescription for a community patient using a DAA without seeing the patient and at what if any cost?
8. What information do you believe should be included on a prescription reminder/request to promote best practice?
9. What would you add/change in the recommendations to address the issues you may have with prescription management for DAA patients and to improve the interaction you have with community pharmacies providing DAAs for your patients?
10. Below is a list of tasks that are required to ensure that DAAs are packed accurately and provided on time. For each task, please indicate who (e.g. the patient, doctor, surgery staff, pharmacy or a community nurse) currently looks after this task and who you believe shouldbe responsible for this task (you may indicate more than one person).
Task Who currently does this task?
Who should do this task?
Recommend that a patient use a DAA Assess a patient’s ability to use a DAA Ensure that the pharmacy has a current and complete record of the patient’s medicines Decide what should be packed in the DAA Decide what is the optimal schedule for the patient (times of day for each medicine). Store the patient’s repeat prescriptions Provide the patient with a current and complete written record of their medicines Tell the pharmacy about any changes to the patient’s medicines Making sure prescriptions are available to ensure continuity of supply of packed medicines. Inform the doctor when a repeat prescription is required Assess how patients are managing with the DAA
11. Would a patient held medicine template (as per Recommendation 1.3.3) be useful/helpful to your community patients using DAAs? Yes No
Why or why not?
12. Would a patient held medicine template (as per Recommendation 1.3.3) be useful/helpful to you in your practice? Yes No
Why or why not?
349
13. What would be the advantages and disadvantages of doctors approving the patients’ medication template/pharmacy packing profile prior to the pharmacy packing the patients medicines into a DAA?
14. Do you currently communicate changes to the drug regimens of patients who use a DAA (including those changes not needing a prescription i.e. ceased medicine or changes in doses) in writing? Yes No
If Yes, please describe when you would do this and what is the format:
15. How often do you receive a discharge summary, when a patient using a DAA is discharged from hospital? Never Rarely Sometimes Mostly Always
16. When a patient is discharged from hospital with new medicines do you communicate with the community pharmacy packing the patients DAA about what should be packed/changed (as per Figure 1.3) and do you believe this step is necessary?
17. Would you be willing to conduct 6-monthly concordance checks between your records and the pharmacies packing profile for patients receiving DAA services? Yes No
If No, why not?
18. What do you believe would be the main barriers and implications to implementing these recommendations (1.3.1 to 1.3.9) in your practice?
19. Please comment on any other problems you have experienced with DAA services that are not addressed by these recommendations. Do you have any other suggestions for improving the way DAA services are currently provided?
BEST PRACTICE RECOMMENDATIONS FOR THE PROVISION AND USE OF DOSE
ADMINISTRATION AIDS IN RESIDENTIAL CARE FACILITIES (RCFs)
1. From your perspective, what would be the benefits of a RCF (where you provide care for residents) involving you in the process of tendering/ contracting DAA services from a community pharmacy?
2. Would you be willing to be involved in a written agreement as specified in 1.3.2 prior to a community pharmacy beginning to provide a DAA service to a facility where you provide care for residents? Yes No
3. Given that the purpose of this agreement is to ensure that all parties are aware of what is involved in DAA services and can discuss preferences for how to carry out the requirements of providing DAAs prior to beginning this service, which aspects of this agreement would you consider important/useful?
4. Which aspects of this agreement would you change or leave out and why?
5. Below is a list of tasks that are required to ensure that DAAs are packed accurately and provided on time. For each task, please indicate who (e.g. facility staff, doctor, your surgery staff, or pharmacy) currently looks after this task and who you believe should be responsible for this task (you may indicate more than one person).
Task Who currently does this task?
Who should do this task?
Assess a self-medicating residents ability to use a DAA Ensure that the pharmacy has a current and complete record of the residents medicines Decide what should be packed in the DAA Decide what is the optimal schedule for the resident (times of day for each medicine). Store the residents repeat prescriptions Tell the pharmacy about any changes to the residents medicines Making sure prescriptions are available to ensure continuity of supply of packed medicines. Inform the doctor (you) when a repeat prescription is required
350
6. Under what circumstances would you be willing to write a prescription for a resident using a DAA without seeing the resident and at what if any cost?
7. What information do you believe should be included on a prescription reminder/request to promote best practice?
8. What would you add/change in recommendation 1.3.7.1 to address the issues you may have with prescription management for DAA patients and to improve the interaction you have with community pharmacies providing DAAs for your patients?
9. What would be the advantages and disadvantages of doctors approving the residents’ medication template/pharmacy packing profile prior to the pharmacy packing the patients medicines into a DAA?
10. Do you currently communicate changes to the drug regimens of residents who use a DAA (including those changes not needing a prescription i.e. ceased medicine or changes in doses) in writing? Yes No
If Yes, please describe when you would do this and what is the format:
11. How often do you receive a discharge summary, when a resident using a DAA is discharged from hospital? Never Rarely Sometimes Mostly Always
12. When a resident is discharged from hospital with new medicines do you communicate with the community pharmacy packing the patients DAA about what should be packed/changed (as per Figure 1.3) and do you believe this step is necessary?
13. When a resident is discharged from hospital to a RCF with new medicines what is your procedure for updating the residents chart and in what timeframe can you preform this task?
Please describe any circumstances that complicate or delay this process:
14. Would you be willing to conduct 6-monthly concordance checks between your records, the residents chart and the pharmacies packing profile for patients receiving DAA services? Yes No
If No, why not?
15. What do you believe would be the main barriers and implications to implementing these recommendations (1.3.1 to 1.3.8) in your practice?
Please comment on any other problems you have experienced with DAA services that are not addressed by these recommendations. Do you have any other suggestions for improving the way DAA services are currently provided?
RCF MANAGEMENT FEEDBACK QUESTIONS
RCF Feedback on “BEST PRACTICE RECOMMENDATIONS FOR THE PROVISION AND USE OF DOSE ADMINISTRATION AIDS IN RESIDENTIAL CARE FACILITIES (RCFs)”
The following questions are intended to give structure to the feedback we are requesting from you. You may need to refer to the “best practice” document to complete the questions We have targeted some specific areas of the best practice model where we would appreciate your specific comments. We are, however, interested in any other comments about the model, it’s potential advantages and disadvantages, its feasibility and the impacts on your facility were the model to be adopted.
1. After considering the steps involved in a DAA service (Figure 1,1) and the implementation model (Figure 1.2):
a) How feasible are the implementation model and recommendations related to tendering for a DAA service? If not feasible, please indicate why.
b) What changes would you need to make to your tendering process? What are the cost, staffing and other impacts?
c) How feasible are the implementation model and recommendations related to monitoring and quality assurance for DAA service quality within the RCF and by Aged Care assessors? If not feasible, please indicate why.
351
d) What changes would you need to make to address QA for RCF processes that support the use of DAAs in medication administration? What are the cost, staffing and other impacts?
e) How could Recommendation 1.3.1 be improved?
2. Would you be willing have a written agreement as specified in 1.3.2 prior to a community pharmacy beginning to provide a DAA service to a facility? Yes No
3. Given that the purpose of this agreement is to ensure that all parties are aware of what is involved in DAA services and can discuss preferences for how to carry out the requirements of providing DAAs prior to beginning this service, which aspects of this agreement would you consider important/useful?
4. Which aspects of this agreement would you change or leave out and why?
5. Below is a list of tasks that are required to ensure that DAAs are packed accurately and provided on time. For each task, please indicate who (e.g. facility staff, doctor, Dr’s surgery staff, or pharmacy) currently looks after this task and who you believe should be responsible for this task (you may indicate more than one person).
Task Who currently does this task?
Who should do this task?
Assess a self-medicating residents ability to use a DAA Ensure that the pharmacy has a current and complete record of the residents’ medicines Decide what should be packed in the DAA Decide what is the optimal schedule for the resident (times of day for each medicine). Store the residents’ repeat prescriptions Tell the pharmacy about any changes to residents’ medicines Making sure prescriptions are available to ensure continuity of supply of packed medicines. Inform the doctor when a repeat prescription is required
6. For Recommendation 1.3.3, the process of communicating medication changes to the pharmacy in writing AND keeping a record of that communication:
a) How feasible is the recommendation? Do you see this process as beneficial in promoting safe, effective and efficient DAA services?
b) What changes would you need to make to your medication change communication process? What are the cost, staffing and other impacts?
c) How could Recommendation 1.3.3 be improved?
7. For Recommendation 1.3.4, about continuity of care between hospitals and RCFs:
a) Do you see the model in Figure 1.3 as beneficial in promoting safe, effective and efficient DAA services? Is this feasible? If not, please indicate why.
b) What changes would you need to make to your processes related to hospitalisation of residents? What are the cost, staffing and other impacts?
c) How could Recommendation 1.3.4 be improved? Think about any circumstances that might complicate or delay the processes.
8. What changes would you need to make to your processes related to Recommendation 1.3.8? What are the cost, staffing and other impacts
9. What do you believe would be the main barriers and implications to implementing these recommendations (1.3.1 to 1.3.8) in your facility?
10. What strategies or support (e.g. from Aged Care peak bodies, etc) do you believe would be necessary lead to widespread adoption of this best practice model by the Residential Aged Care Sector?
11. Please comment on any other problems you have experienced with DAA services that are not addressed by these recommendations. Do you have any other suggestions for improving the way DAA services are currently provided?
352
RCF RESIDENT INTERVIEW
My name is XXXXXX and I am working in a research project being conducted by the School of Medicine at the University of Queensland.
The project is about defining the best way to provide medicines packed in dose administration aids to Aged Care facilities. [Show picture of DAA]
Dose administration aids (DAAs) are devices packed by your pharmacy to help the nurses administer medications to residents.
In this interview, I want to ask about your experiences when you first moved to This facility and how the move affected things related to your medicines. Your participation is voluntary and you can stop at any time in the interview
Your responses will be confidential. In any report on the study, no-one will be able
to identify who said what.
To help me report what you say accurately, it would help me to make a tape recording of our conversation. After I transcribe your responses and correct my notes, the tape will be wiped.
Do I have your permission to tape?
1. How long ago did you move to this facility?
2. Where were you just before you moved to this facility (e.g. at home, hospital, other
RCF)
3. Who looked after your medicines there, making sure that you had a supply and that
you had a current prescription when you needed it?
When you moved to this facility:
4. Were the process involved in a smooth transition to ensure you had your correct
medicines available when you needed them explained to you by the Nurses?
5. (a) Did you have to change your usual Doctor?
(b) Did you have to change your usual Pharmacy?
6. Were there any initial problems for you and/or your carer/family related to
transferring your medicine supply and current prescriptions to This facility and to
the usual This facility Pharmacy? In the early days of your move to This facility,
were the medicines and prescriptions you needed always available when required?
Or did it take some work to organise? Please describe what happened.
7. If YES to Q6 and NO to Q4, would it have been helpful and prevented or reduced
these problems if you had had the steps needed to make a smooth transition
explained to you?
8. Were you aware of any problems the Nurses had with your medicine supply and
prescriptions when you first moved to This facility? If so, please describe.
353
In the best practice Dose Administration Aids service, we have suggested that at the
time a person moves into an aged care facility, that there is an agreement that
specifies up front the following details of the service to be provided [give large print
version to resident]:
Who is the pharmacy providing the DAA service.
What are the costs of the service and who should the pharmacy charge
What the person’s costs are (for dispensed medicines and any other pharmacy
items like vitamins)
What the billing or account procedures are.
What type of DAA will be provided? How long will each pack last (weekly,
fortnightly, monthly)?
What medicines will be packed (prescription, over-the-counter medicines, vitamins
and supplements?)
Your consent to sharing information about your medicines between the pharmacy,
GPs, carers and hospital if required?
Who will tell the pharmacy about any changes to your medicines (you, the Nurses
or your doctor) and how will the pharmacy be notified of the changes?
What will happen to medicines in a pack that has to be changed and returned to the
pharmacy.
Will the pharmacy keep your prescriptions at the pharmacy or will you or the nurses
keep them?
Will the pharmacy remind you or the nurses to arrange a doctor’s visit when you
need a new prescription or will the pharmacy request prescriptions from your doctor
on your behalf?
What to do if you go away (e.g. holidays or to hospital).
What will happen and what should you do if you have a prescription filled at a
different pharmacy (i.e. a hospital pharmacy).
Procedures for returning unused medicines to the pharmacy.
How the pharmacy will provide you with information about your medicines?
9. Do you currently have a formal agreement with you pharmacy detailing
what is involved in your DAA service? Yes No
If YES, what aspects are covered/were discussed with you?
10. Would you be willing to sign an agreement that covered the things above?
Yes No
11. Please tell us which parts of the agreement you would like to change or leave out
(and why)?
12. How helpful to you think such an agreement would be to people moving into an
aged care facility? (Think back to the time you moved into This facility, how helpful
would it have been to have all the topics discussed)
Extremely helpful Some what helpful Not at all helpful
13. Did the pharmacist at your usual pharmacy, the one you used before moving to
This facility, give you (or your carer as applicable) information either verbally, in
conversation or as a leaflet or brochure about your various medicines? [show CMI
example]
If yes, what was it, how often and did you find this helpful?
354
14. Have you ever received information about your medicines either verbally, in
conversation or in writing (e.g. Medicines Information for Consumers brochure)
since you have lived in This facility?
If yes, who provided the information, what was it, how often and did you find this
helpful?
15. Do you think it is important that you receive information about your medicines (for
example, what your medicines are for, how to take them, any precautions and
possible side effects).
16. If family/carer has involvement in medicines, is it important that your family/carer
receives information about your medicines?
17. In what situations would you like to be given written information like the package
inserts or Medicines Information for Consumers [show CMI example]? In which of
the following situations?
Only when you start a new medicine
When the information has significantly changed
When, for example, you change from a tablet to a capsule of the same medicine
When there are special reasons to remind you about how to use the medicine and
its precautions
Only when you request this information
For long term treatments, routinely, say every 6 months just as an update.
Not required: (please give details of reason/s why not)
Are there any other comments you would like to make about how to make the provision
of medicines packed in Dose Administration Aids a better service?
Thank you for your time and input today.
RCF/Patient identifier:
Date of Interview:
355
APPENDIX H: COMMUNITY PATIENT FOLLOW-UP
SURVEY
Community patient Phase 3 introduction letter
Community patient telephone interview script
Community patient mail-out version
356
The Project Officer
Dose Administration Aid Study
PO Box 6067
Buranda, QLD, 4102.
««AddressBlock»»
1 February 2005
««GreetingLine»»
We are writing to thank you for your participation in the Pharmacy Guild of Australia
Dose Administration Aids (DAAs) study run by the University of Queensland/ Princess
Alexandra Hospital. We are pleased to inform you that this study was a great success
thanks to your time and effort and the time and effort of all the community pharmacies
and their customers who participated around Australia.
The information you provided helped us a lot and also raised some more questions that
we would like to collect information on, so that we can better understand how helpful
DAAs are. We are contacting you to ask for your continued assistance with this study.
WHAT IS REQUIRED OF YOU
We are asking you to participate in a telephone interview conducted by a project staff
member. We expect that the interview will take approximately 15 minutes and we
would like to ask you questions about DAAs, your health, your medicine use and your
health care visits. Your participation is entirely voluntary and you may stop the
interview at any point. All personal information collected for the trial will be treated
confidentially and will not be disclosed to any person body or agency.
Your decision about being involved in the study will not in any way affect your
pension, benefits or any health services you are entitled to.
357
A project staff member will telephone you in February or early March to ask you to
participate in the interview or to make an appointment to talk with you at a convenient
time. It is also possible for us to mail you a survey to be completed at your leisure and
returned in a reply paid envelope.
For one of the questions about your health we ask you to look at a scale from 0 to 100.
This scale has been included with this letter. If possible please keep this scale near
the telephone as we will refer to it when we call.
The study office will call you so you will not incur any expenses. We also have a free
call number if you would like to contact us free of charge 1800 555 803.
We would also like to let you know that if you are interested in the results of this study
they will be available from the Pharmacy Guild of Australia website (www.guild.org.au)
or you can call us here at the study office and we will send you a summary of our
findings.
Thank you for you assistance and we look forward to talking with you soon.
Regards,
Clare Ientile
Project Officer
358
DAA STUDY Phase 3 – Patient Telephone interview
PHONE NUMBER: XXXXXXXXX Gender: Pt ID: XXXXX
1. My name is ____________; I am calling from the School of Medicine University of
Queensland. I would like to speak to XXXXXXXX about a study on Dose Administration
Aids (a device supplied by the pharmacy to help you take you medicine) that XXXXXXXX
participated in last year with XXXXXXX Pharmacy.
Verify that you are talking to participant and go to 3. If you are unable to speak to the participant
go to 2.
2. We would like to follow-up with XXXXXXX, to see how he/she is doing and to ask if they
would be willing to do another brief survey. Would you be able to let us know when/how
4. Did you require treatment for these symptoms or health problems caused by your
medicines? YES (go to 5) NO (go to Part D)
5. I will list different types of treatments one at a time and ask you to tell me whether you
had that treatment and the number of visits or duration of the treatment: I am interested in
the type of treatment you had in the last 12 months for the medicine related symptoms only.
361
ADE related health service use in last 12 months No. of Times Total no. days
Did you see your GP?
Did you start a new medication or increase your dose? YES NO
Did you visit a hospital emergency department?
Were you admitted to hospital?
Did you require respite care?
Were you admitted to a residential care facility?
Did you require assistance from community nurses?
Any additional treatment: ______________________
PART D – This part of the survey is about your current health status
1. Have you experienced any of the following conditions in the past four weeks? (Read
through the list and indicate YES or NO in the middle column. If YES, ask “have you
experienced this condition in the last two days?” and indicate YES or NO in the far right
column).
Condition Past four weeks Past two days
Falls
Dizziness
Headaches
Fever
Feeling depressed or blue
Pain (back, chest, joints/limbs or abdominal)
Extreme fatigue/tiredness
Shortness of breath
Nausea, constipation or diarrhoea
Feeling anxious or worried
Vision problems
Trouble sleeping
Memory trouble
Hearing problems
Do you have the letter with the picture of the EQ-5D scale from 0-100 (looks like a
thermometer)? YES (please refer to this now) NO (go to 2)
2. To help people say how good or bad their health state is, we ask people to think of a
scale (rather like a thermometer) on which the best health state you can imagine is marked
100 and the worst health state you can imagine is marked 0. Using that scale, could you
please tell us a number, from 0 to 100, that tells us, in your opinion, how good or bad your
own health is today?
Score: _______
362
The following five questions are about a number of areas related to your level of functioning. In
each case please tell us which statement best describes you today
2a. Which of the following 3 categories best describes your level of mobility today?
I have no problems in walking around
I have some problems in walking around
I am confined to bed
2b. Which of the following 3 categories best describes your ability to conduct your personal
care activities today?
I have no problems with personal care
I have some problems washing or dressing myself
I am unable to wash or dress myself
2c. Which of the following 3 categories best describes your ability to conduct your usual activities today (e.g. work, study, housework, family or leisure activities)? I have no problems with performing my usual activities
I have some problems with performing my usual activities
I am unable to perform my usual activities
2d. Which of the following 3 categories best describes the level of pain or discomfort you have
experienced today (e.g. back pain/discomfort, headaches etc.)?
I have no pain or discomfort
I have moderate pain or discomfort
I have extreme pain or discomfort
2e. Which of the following 3 categories best describes your mood today in relation to feelings of
depression or anxiety?
I am not anxious or depressed
I am moderately anxious or depressed
I am extremely anxious or depressed
PART E - Health service use in last four weeks.
This part of the survey is about your health care services, the cost to you (or your family) and
the level of support you have had in the last 4 weeks. (Write 0 if no visits or no days)
Health services used in the last four weeks No. of
times
Total $ to pt.
1. In the last four weeks, how many times have you seen a doctor
(GP or specialist) at home, in the surgery or elsewhere?
2. In the last four weeks, how many times have you seen a
community nurse?
3. In the last four weeks, how many times have you seen a health
professional other than a doctor (eg physiotherapist, podiatrist,
chiropractor, dietician, naturopath or similar? (NOTE: DO NOT
INCLUDE PHARMACISTS IN THIS FIGURE)
4. In the last four weeks, how many times have you been treated
in casualty or the emergency department of a hospital but not
admitted?
5. In the last four weeks, how many times have you been treated
as an outpatient or at clinics at the hospital (seen a Dr. or other
healthcare professional)?
6. Have there been any other costs needed to meet your health
care needs in the last four weeks such as pathology (eg blood
tests), biopsies, x-rays, etc.?
363
Health services used in the last four weeks No. of times Total $ to pt.
7. Have you used in the last four weeks any support services
(home nursing, Meals-on-wheels, home help)? Please
describe:_______________________
8. Have you had any other types of health care in the last four
weeks? ____________________________
Hospitalisation in the last 12 months No. of times Total $ to pt.
9. In the last 12 months, how many times have you been
admitted to a public hospital?
10. In total, how many days were you in a public hospital?
11. In the last 12 months, how many times have you been
admitted to a private hospital?
12. In total, how many days were you in a private hospital?
13. In the last 12 months, how many days did you spend in other
care facilities (eg respite care or hostel)?
PART F- These questions are how much a DAA would be worth to you.
[Note: the order in which these questions are asked varies from interview to interview]
1. How much would you be willing to pay to have your medicines packed in a device and
delivered to you so that your medication taking was easy and convenient?
$----------/week
2. How much would you be willing to pay to have your medicines packed in a device that
would make remembering to take your medications easier?
$----------------------- /week
3. An adverse drug event is an illness related to medicines (including side effects and
allergies) or injury resulting from using a medicine. How much would you be willing to
pay to have your medicines packed in a device that would reduce your risk of
experiencing an adverse drug event? $----------------------- /week
Thank you for your participation in this interview. If we have any further questions that we’d like
to follow up, would it be ok for us to contact you again? YES NO
364
DAA STUDY Phase 3 – Patient Questionnaire
Name: «Given_Name» «Family_Name» ID:«Patient_ID»
PART A – These questions are about Dose Administration Aids and Medication Use
1. Do you have a Dose Administration Aid?
YES NO (go to Part B) Don’t know
If YES, what type of Dose Administration Aid do you use? Tick the ONE that applies
Webster Multidose Pack MPS Sachet Roll Mediplanner
4. Did you require treatment for these symptoms or health problems caused by your
medicines? YES (go to 5) NO (go to Part D)
5. We are interested in the type of treatment and the number of times you had this
treatment in the last 12 months for the medicine related symptoms only. Please
refer to the table below and indicate whether you had this type of treatment and
how many times you required this treatment for your medicine related symptoms.
Health services used in last 12 months for medicine
related symptoms only.
Had this type
of treatment?
Number of
times?
a) Did you see your GP? YES NO
b) Did you start a new medication or increase your dose? YES NO
c) Did you visit a hospital emergency department? YES NO
d) Were you admitted to hospital? YES NO
e) Did you require respite care? YES NO
f) Were you admitted to a residential care facility? YES NO
g) Did you require assistance from community nurses? YES NO
h) Any additional treatment: ______________________ YES NO
PART D– This part of the survey is about your current health status
1. Have you experienced any of the following conditions in the past four weeks? If
YES, have you experienced this condition in the last two days?
(Please circle YES or NO in each column).
Condition Past four weeks Past two days
Falls YES NO YES NO
Dizziness YES NO YES NO
Headaches YES NO YES NO
Fever YES NO YES NO
Feeling depressed or blue YES NO YES NO
Pain (back, chest, joints/limbs or abdominal) YES NO YES NO
Extreme fatigue/tiredness YES NO YES NO
Shortness of breath YES NO YES NO
366
Part B, Question 1 (continued)
Condition Past four weeks Past two days
Nausea, constipation or diarrhoea YES NO YES NO
Feeling anxious or worried YES NO YES NO
Vision problems YES NO YES NO
Trouble sleeping YES NO YES NO
Memory trouble YES NO YES NO
Hearing problems YES NO YES NO
2. To help people say how good or bad a health state is, we
have drawn a scale (rather like a thermometer) on which the
best state you can imagine is marked 100 and the worst state
you can imagine is marked 0.
We would like you to indicate on this scale how good or bad
you health is today, in your opinion. Please do this by
drawing a line from the box below to whichever point on the
scale indicates how good or bad your health state is today.
Worst
imaginable
health state
Your own
health state
today
Best imaginable health state
367
3. By placing a tick in one box in each group below, please indicate which statement best describes your own health state today.
3a. Which of the following 3 categories best describes your level of mobility today?I have no problems in walking around I have some problems in walking around I am confined to bed
3b. Which of the following 3 categories best describes your ability to conduct your personal care activities today?I have no problems with personal care I have some problems washing or dressing myself I am unable to wash or dress myself
3c. Which of the following 3 categories best describes your ability to conduct your usual activities today (e.g. work, study, housework, family or leisure activities)?I have no problems with performing my usual activities I have some problems with performing my usual activities I am unable to perform my usual activities
3d. Which of the following 3 categories best describes the level of pain or discomfort you have experienced today (e.g. back pain/discomfort, headaches etc.)?I have no pain or discomfort I have moderate pain or discomfort I have extreme pain or discomfort
3e. Which of the following 3 categories best describes your mood today in relation to feelings of depression or anxiety? I am not anxious or depressed I am moderately anxious or depressed I am extremely anxious or depressed
PART E - Health service use in last four weeks.
This part of the survey is about your health care services, the cost to you (or your family) and the level of support you have had in the last 4 weeks. (Write 0 if no visits or no days)
Health services used in the last four weeks Number of times
Total cost to you
1. In the last four weeks, how many times have you seen a doctor (GP or specialist) at home, in the surgery or elsewhere?
2. In the last four weeks, how many times have you seen a community nurse?
3. In the last four weeks, how many times have you seen a health professional other than a doctor (eg physiotherapist, podiatrist, chiropractor, dietician, naturopath or similar? (NOT INCLUDING PHARMACISTS)
4. In the last four weeks, how many times have you been treated in casualty or the emergency department of a hospital but not admitted?
5. In the last four weeks, how many times have you been treated as an outpatient or at clinics at the hospital (seen a Dr. or other healthcare professional)?
6. Have there been any other costs needed to meet your health care needs in the last four weeks such as pathology (eg blood tests), biopsies, x-rays, etc.?
368
Health services used in the last four weeks No. of times
Totalcost.
7. Have you used in the last four weeks any support services (home nursing, Meals-on-wheels, home help)? Please
describe:_______________________
8. Have you had any other types of health care in the last four weeks? Please describe:_______________________
Hospitalisation in the last 12 months No. of times
TotalCost.
9. In the last 12 months, how many times have you been admitted to a public hospital?
10. In total, how many days were you in a public hospital?
11. In the last 12 months, how many times have you been admitted to a private hospital?
12. In total, how many days were you in a private hospital?
13. In the last 12 months, how many days did you spend in other care facilities (eg respite care or hostel)?
PART F- These questions are how much a DAA would be worth to you.
There is no right or wrong answer to these questions and we ask you to estimate the
value of the different potential benefits of DAAs given your needs and what you can
afford to pay per week for this service.
1. An adverse drug event is an illness related to medicines (including side effects and
allergies) or injury resulting from using a medicine. How much would you be willing
to pay to have your medicines packed in a device that would reduce your risk of
experiencing an adverse drug event? $----------------------- /week
2. How much would you be willing to pay to have your medicines packed in a device
and delivered to you so that your medication taking was easy and convenient?
$----------/week
3. How much would you be willing to pay to have your medicines packed in a device
that would make remembering to take your medications easier?
$----------------------- /week
Thank you for your participation in this interview. If we have any further questions that
we’d like to follow up, would it be ok for us to contact you again?
YES NO
Please place this questionnaire in the Reply Paid envelope provided.
369
APPENDIX I: HIC DATA DISTRIBUTIONS AND
MODELS
DISTRIBUTIONS OF AGGREGATED HIC SERVICE USE
VARIABLES
The following figures show the untransformed and transformed service use variables
used in analyses. Cost variables were transformed using a natural logarithm
transformation while variables concerning the number of items were transformed by a
square root transformation or by recoding into a dichotomous variable (e.g. received a
particular service or not).
0 50 100 150 200
No. PBS items in last yr
0
10
20
30
Fre
qu
en
cy
Mean = 85.45Std. Dev. = 37.682N = 224
2.0 4.0 6.0 8.0 10.0 12.0 14.0
Sqrt No. PBS items last year
0
10
20
30
40
Fre
qu
en
cy
Mean = 9.0192Std. Dev. = 2.03075N = 224
0 200
0
400
0
600
0
8000
10
00
0
$ PBS benefit in last yr
0
10
20
30
40
Fre
qu
en
cy
Mean = 2609.6423Std. Dev. = 1626.15511N = 224
6.00 7.00 8.00 9.00
Ln $ PBS benefit in last yr
0
10
20
30
40
50
Fre
qu
en
cy
Mean = 7.6802Std. Dev. = 0.64621N = 224
5 10 15 20 25 30 35 40
No. different PBS items in last year
0
10
20
30
Fre
qu
en
cy
Mean = 16.61Std. Dev. = 7.538N = 224
2.00 3.00 4.00 5.00 6.00
Sqrt No. different PBS items last yr
0
5
10
15
20
25
Fre
qu
en
cy
Mean = 3.9741Std. Dev. = 0.90635N = 224
Appendix Figure 4 PBS service use variables for services in the year before the homes