Pharmacist-patient medication communication during admission and discharge in specialty hospital settings: implications for person centered healthcare

1

Title Pharmacist-patient medication communication during admission and discharge in specialty hospital settings: implications for person centered healthcare International Journal of Person-Centered Medicine, accepted January 9, 2015, doi: http://dx.doi.org/10.5750/ijpcm.v4i2.446 Author names: Sandra Braaf, BN, MTrngDev, PhD1 Sascha Rixon, BSc/BA (Hons), PhD1 Allison Williams, Grad Dip Adv Nsg(Clin. Nurse Ed.), BaAppSc Nur, MN PhD2 Danny Liew, BMedSc, MBBS(Hons), FRACP, PhD, CertHealthEcon3 Elizabeth Manias, BPharm, MPharm, MNursStud, PhD FACN (DLF) MPS MSHPA 1,3,4

Author affiliations: 1The University of Melbourne, Parkville, Victoria, Australia 2Monash University, Clayton, Victoria, Australia 3Royal Melbourne Hospital, Parkville, Victoria, Australia 4Deakin University, Burwood, Victoria, Australia Corresponding Author: Sandra Braaf Melbourne School of Health Sciences Faculty of Medicine, Dentistry and Health Sciences The University of Melbourne Alan Gilbert Building, level 6 Parkville Victoria 3010 Australia T: +61 3 9035 4933 F: +61 3 8344 5391 Email : [email protected] Running title Pharmacist-patient medication communication

2

Abstract Background Hospital admission and discharge are critical transition points for patients’ medication management. Effective communication between pharmacists and patients at these transition points has the potential to mitigate medication incidents. Previous research has examined communication among community pharmacists and patients. Limited research has explored the complexities of communication between hospital pharmacists and patients during admission and discharge interactions. Objective To explore the complexities of pharmacist-patient medication communication during medication admission and discharge in specialty hospital settings. This process involved examining the socio-cultural and environmental influences on communication processes, the characteristics of actual communication encounters, and the outcomes of these communication interactions. Methods The study was conducted at a metropolitan Australian public hospital. An exploratory qualitative design was used involving the methods of semi-structured interviews and participant observation. Interviews and observations were undertaken in five specialty settings: cardiothoracic care, intensive care, emergency care, oncology care and perioperative care. A comprehensive thematic analysis of the data was performed. Results Twelve pharmacists and 69 patients participated in interviews and observations for the study. Over 200 hours of observational data were collected. In total, 26 medication admissions and 35 medication discharges were observed. Pharmacists regulated communication with patients by using structured communication tools. When providing or gathering information, pharmacists controlled the level of patient engagement. Words used in pharmacist-patient communication were sometimes ambiguous and occasionally miscommunication resulted. Patients sought minimal information from pharmacists. Conclusions Pharmacists need to apply patient-centered principles, and encourage active patient involvement, in admission and discharge conversations. Key Words Communication; health care, acute / critical; medication; observation, participant; research, qualitative.

3

Hospital admission and discharge are critical transition points for patients’ medication management and the delivery of continuous health care. [1] At admission and discharge to hospital, there are different health professionals providing patient care, and patients’ medication regimens are often altered. [2, 3] Patients located in specialty hospital settings such as cardiothoracic care, intensive care, oncology care, emergency care and perioperative care receive complex treatments and are frequently administered high risk medications. High risk medications, such as insulin and opioid medications, if used in error can cause significant patient harm. [4] Effective communication between patients and hospital pharmacists is critical for patients in specialty areas to endorse safe and appropriate medication management across patient transitions in and out of hospital. [5]

Miscommunication about medications plays a key role in the development of medication errors. [6] Preventing miscommunication at pharmacist-patient interactions, such as during hospital admission and discharge, therefore has the potential to avert detrimental patient outcomes. During a medication admission, pharmacist-patient communication involves pharmacists formulating a best possible medication history and a list of current medications. At medication discharge (sometimes referred to as patient counseling), pharmacists and patients communicate about the medications a patient will use after leaving the hospital. [7] Miscommunication at either of these two transition stages might result in inaccurate and incomplete patient medication lists, and the risk that patients may not take their prescribed medications in a safe and effective manner.

The application of patient-centered communication by pharmacist when interacting with patients about medications can facilitate clear and effective communication, as patients are actively involved in the encounter. [8] Active patient communication and collaboration with pharmacists, opens up opportunities for building rapport, sharing knowledge and perspectives, and developing understandings about medication taking. [9, 10] An open and patient centered communication environment also enables patients to freely raise concerns, to express previous experiences with medications, and to ask questions. [11] Further, to optimize patient education, pharmacists can customize teaching and information giving about the appropriate use of medications to a patient’s preferred learning style by using patient-centered principles. [8]

There has been extensive past research on pharmacist-patient communication in the community setting. An ethnographic study in the United Kingdom (UK) involved 15 community pharmacists examining 350 videotaped interactions with patients to determine what constituted effective communication. [12] Pharmacists rated building rapport with patients as the most important of 11 skills for cultivating effective communication. This skill involved being available, being accessible and engaging with patients. Shah and Chewning [13] used a survey method and structured observations to investigate the communication behaviors of 30 United States (US) community pharmacists. The researchers observed 306 pharmacist-patient interactions and then compared observer reports to patient survey reports. In their communication behavior, pharmacists exhibited a preference for unilateral transmission of information to patients, over two-way transaction. Discrepancies were evident between patient and

4

observer reports in regards to whether a pharmacist asked a question or not during the interaction.

Patient communication encounters with hospital pharmacists, however, are unlike those undertaken with community pharmacists. Hospital and community pharmacists’ roles differ in scope and diversity, and patients vary in their health status between hospital and community settings. Hospital pharmacists communicate with patients who are admitted with urgent and serious health problems. [14] To date, there has been very little research on pharmacist-patient communication in the hospital environment. Although numerous scholars have addressed the topic of medication reconciliation [15], a process involving pharmacists checking patient medication list against their medication administration record, addressing discrepancies and documenting changes [16], few have explored the details and complexities of hospital pharmacist-patient communication.

In a Nigerian interview study Babalola and Erhun [17] identified the communication patterns between four hospital pharmacists and 30 older patients with medical and renal problems during medication history taking. They used a socio-linguistic model to analyze 20 audio and video-taped transcribed interactions. Pharmacist-patient miscommunication occurred because of noise, use of medical terminology, conflicting statements and ambiguity over the intent of a message. Outcomes included prolonged interaction times, inaccurate and incomplete messages from the pharmacist, and inaccurate feedback from the patient to the pharmacist. Also using a linguistic analysis, Pilnick [18] identified four approaches used by pharmacists to initiate medication counseling sequences by recording 45 hospital pharmacist out-patient interactions in a UK pediatric oncology setting. A patient or carer requesting information, rather than advice, from the pharmacist was one approach to counseling. Another approach involved pharmacists unilaterally providing information without checking if the patient or carer had prior knowledge of a medication. A pre-sequence approach concerned pharmacists making a statement of their intention to commence, or an offer to provide, counseling. A predictable communication pattern in the first three counseling approaches was pharmacists giving information to predominantly passive patients. The fourth approach entailed pharmacists posing questions to patients to actively involve them in the interaction.

Focusing on the type of interaction between pharmacists and patients, Murad et al [9] conducted a meta-narrative review to identify the extent to which pharmacist-patient interactions revealed the presence of biomedical or patient-centered communication. (removed sentence on Pt centered communication as it was repetitive) Biomedical models of care de-emphasize patient interaction with health professionals, as patients are directed in the use of their medicines. [9] The review, which included hospital and community pharmacists, revealed biomedical models of communication (23 studies) prevailed over patient-centered models of communication (8 studies) in pharmacist-patient interactions.

Gaps in pharmacist-patient communication research include a lack of focus on hospital pharmacist communication with adult patients in diverse hospital settings. Little is known about the context and complexities of pharmacist-patient communication during hospital admission and discharge processes. Moreover, few authors in past studies have used a participatory observational approach or multiple methods of data collection. No published study, to our knowledge, has been designed

5

to investigate pharmacist-patient medication communication during admission and discharge in specialty hospital areas. Hence, our objective in this study was to explore the complexities of pharmacist-patient medication communication during medication admission and discharge in specialty hospital settings.

Methods Design and Setting We used an exploratory qualitative design for this investigation. The study was conducted at a metropolitan Australian public hospital and involved the methods of semistructured interviews and participant observation. We interviewed pharmacists and patients in five specialty settings: cardiothoracic care, intensive care, emergency care, oncology care and perioperative care. We conducted observations of pharmacist-patient communication in the same five specialty settings, except perioperative care, which was excluded as interviews revealed most pharmacist communication took place with health care professionals rather than with patients. Furthermore, pre and postoperative patients were located in the other specialty areas where observations had been conducted. Ethics approval was granted by the hospital Human Research and Ethics Committee. Recruitment for Observation and Interviews We recruited participants for interviews between November 2011 and March 2012. The recruitment for observations took place prior to data collection in each of the four specialty areas, and occurred from October 2012 until May 2013.

Patients were considered for inclusion in the study if they were 18 years and older, cooperative and were admitted to or discharged from a specialty area. Only patients able to read and speak English were interviewed. For observations, if a patient was unable to read and speak English, verbal consent was sought from the patient through an English speaking family member who spoke the patient’s native language. Written consent was then gained from the family member. In the intensive care setting where some patients were unconscious, consent for participation in observations was sought from family members.

To recruit patients for interview, a member of the research team approached nurse unit managers to gain appropriate referrals. Patients needed to be medically stable, alert and orientated, and be receiving prescription medications. If a patient expressed an interest in the study, we provided spoken and written information. Patients with varying medical disorders were purposively selected.

We recruited pharmacists for observation or interview through multiple strategies. Pharmacists were considered eligible for participation in the study if they were qualified and employed in a specialty hospital setting. Information sessions were delivered in all specialty areas and pharmacists who displayed an interest in the study were provided with written information. Pharmacists were also recruited through referrals from clinical staff who worked in the study areas. Following a referral, we approached pharmacists through email or in person and invited them to participate in the study. At this time the pharmacist were informed of the objectives of the study and provided with a detailed information and consent form. Pharmacists were purposively selected according to their clinical role and years of experience.

6

Data Collection Data collection commenced in December 2011 and concluded in June 2013. Five pharmacists and 11 patients participated in semistructured individual interviews. Interviews occurred at a mutually agreeable time for the patient/pharmacist and researcher. We interviewed patients at the bedside, and pharmacists were interviewed in a quiet private room within the hospital. Patients were asked questions about how hospital pharmacists communicated with them regarding their medications. We asked pharmacists questions about how they communicated with patients about medications during admission and discharge processes. Pharmacist interviews usually lasted 30 minutes, whereas patient interviews were generally about 15 minutes long. The interview schedules for patients and pharmacists are shown in Table 1.

Table 1. Interview Schedule for Patients and Pharmacists

Interview questions for patients

What information about your medications have you received in hospital? What things do you think are important to discuss with the pharmacists about your medications? What are your overall impressions of how pharmacists have talked to you about your medications during your hospital stay? How involved do you feel in the decisions made about your medications during your hospital stay? Have you ever raised a concern about any of your medications? How confident do you feel about how to use your medications when you leave hospital?

Interview questions for pharmacists

What do you think gets in the way of communicating effectively with patients? What do you think helps in communicating effectively with patients? What do you say to patients about their medications? How are patients involved in decisions about their medications? What resources do you supply for patients about their medications? What are the main issues affecting patients’ knowledge about their medications? What do you think are the main issues affecting patients’ knowledge about their medications?

More than 200 hours of data were collected through participant observation

involving eight pharmacists. As pharmacists approached patients to engage in communication, we sought consent from cooperative and awake patients to participate in the study. After meeting pharmacists at an agreed time, observations commenced for a period of two to four hours. Observations took place at different times of the day and on different days of the week (excluding weekends when pharmacists did not routinely work). Audio recording commenced at the beginning of the observation period and concluded at the end of the observation. We notified pharmacists when audio recording started, as we carried a portable Olympus® DM-5 digital voice recorder in our pocket and wore a lapel omnidirectional microphone.

7

Pharmacists were shadowed at a distance that enabled a clear recording of speech, but was not so intrusive or disturbing to interrupt workflow. The decision to observe pharmacists, as opposed to simply provide them with the audio recorder, was multifactorial. The observation of pharmacists would enable researchers trained in communication research to see how the pharmacists verbally and non-verbally interacted with patients. Rich contextual information could then be recorded in field notes, which would provide relevant and comprehensive details particular to pharmacist-patient communication interactions. This information was considered essential for interpreting these data and to producing a valid analysis. Additionally, it was anticipated that less logistical and technical issues might arise with experienced researchers performing the audio recording. Some potential issues of concern were failed or poor quality recordings, and people interacting with participants and being unaware of the audio recorder.

A clear benefit of the researchers performing the observations was it enabled events to be clarified at, or close to, the time of the communication interaction. At times we engaged with pharmacists and patients during breaks in conversation. While performing observations, family members sometimes contributed to pharmacist-patient interactions or spoke on behalf of the patient, particularly when a family member was responsible for managing a patient’s medications. Occasionally, during observations, pharmacists initiated communication with us. Such interactions were usually to ensure we were following a complex sequence of events. Field notes were documented throughout the observation period. Details recorded in field notes included attributes of medication communication events such as the time, discussion details, individuals involved, location, patient’s language spoken, and the details of expected or unexpected outcomes.

Two researchers (first and second authors) were involved in data collection for the study. One researcher was present in each individual interview and observation. Both researchers possessed extensive knowledge of medications and had substantial experience in communications research. We collected demographic data from all study participants. Codes were allocated to protect participants’ identities and to enable tracking of individuals and settings. All interviews and observations were audio recorded and were transcribed ad verbatim. Data collection was ceased when repeated communication patterns emerged in the various specialty settings.

Data Analysis Pharmacist and patient demographic details were entered into Microsoft Excel (2010, Version 14 Redmond, Washington), and summarized using descriptive statistics. We prepared interview and observational data for analysis by listening to audio recordings and identifying medication communication for transcription. Field notes were added to the transcribed document to provide contextual information relevant to the many communication encounters that occurred. Three researchers (first, second and last authors) performed the analysis, meeting frequently to discuss emerging findings and to refine developing ideas. Regular discussion and reflection led to agreement of key themes and subthemes. Rigor of the research investigation involved processes to ensure dependability, transferability and credibility (Table 2).

8

Table 2. Rigor of Research Process

Dependability: ensured by detailed and transparent records pertaining to methodological processes, issues and decisions, and regularly meeting with multiple experienced and knowledgeable researchers to discuss project progress and management [19]. Creditability: ensured by prolonged engagement in the research settings; data obtained from many diverse research participants; use of an exploratory qualitative approach to answer the research question; triangulation of the data through multiple research methods; and through taking into account data generated by both pharmacists and patients [20, 21]. Transferability: to enable the findings to be transferred to environments with a similar context purposive sampling was used to ensure a range of patient ages and medical conditions; that pharmacists varied in their work settings and years of experience; and the provision of detailed findings and analysis description [20, 22].

We performed a comprehensive thematic analysis using a framework approach, as

described by Ritchie and Spencer [23]. This approach involved five steps: familiarization, identifying a thematic framework, indexing, charting, and mapping and interpretation. To gain an overview of the complexity, extent and diversity of the data, we read and reread the interview and observational transcripts. While reading, notes were made to document emerging ideas. Recurrent themes and issues were then identified to reveal connected and discrete patterns of interactions. This process involved repeated reference to the transcripts and the research question. We subsequently developed a framework of key themes and subthemes. The thematic framework was then applied to the data set, and index headings relating to key themes and subthemes were recorded. Indexing organized the data into a system that linked and traced identified themes to passages in the transcripts. To construct an overall picture of the data, we assembled charts to group and order passages of transcripts under applicable themes or subthemes. In mapping and interpretation, we examined the charts for associations and patterns, which supported interpretation of the data as a whole.

Results Overall, 12 pharmacists participated in observations and interviews, with one pharmacist involved in both. We observed and interviewed 69 patients for their interactions with pharmacists. Most patients were located in emergency and oncology care, as these areas had a consistently high throughput of patients. As many patients were medically unstable or sedated in intensive care and perioperative care, only four and two patients respectively participated in the research from these areas (Table 3).

9

Table 3. Demographic Characteristics of Pharmacist and Patient Participants

Demographic characteristics of participants Patient (n=69) Pharmacist (n=12)

n % n %

Gender Female 35 50.8 10 83.3

Male 34 49.2 2 16.7

Age in years Mean in years (SD) 60.7(18) 29.8(4.2)

Specialty practice setting Cardiothoracic care 11 16.0 3 25

Emergency care 35 50.7 3 25

Oncology care 17 24.6 2 16.7

Intensive care 4 5.80 2 16.7

Perioperative care 2 2.90 2 16.7

Patient country of birth Australia 50 72.5

Overseas 19 27.5 Patient primary medical condition at admission

Cardiovascular 21 30.4

Neurological/ Mental health 12 17.4

Hematology / Oncology 9 13.1

Surgical 6 8.70 Gastrointestinal 6 8.70 Respiratory 4 5.80 Renal 4 5.80 Integument 3 4.30 Fluid and electrolyte 3 4.30 Suspected fracture 1 1.50 Pharmacists’ employment status

Full time 11 91.7

Part time 1 8.30

Pharmacists’ highest education qualification

Bachelor 0 0.00

Graduate Certificate/ Diploma 4 33.3

Masters / PhD 8 66.7

Pharmacist years’ experience Mean in years (SD) 7.3(4.3)

In total, we observed 26 medication admissions and 35 medication discharges (Tables 4 and 5). The mean time taken to complete a pharmacist-patient admission interaction was nine minutes and 36 seconds. At admission encounters, pharmacists’ interactions with most patients involved communication about multiple medications. We identified that many patients did not have a medication list, use a medication aid, or have family members present during admission encounters. The mean time to complete pharmacist-patient discharge communication was seven minutes and 42 seconds. During discharge encounters, pharmacist-patient communication sometimes included references to numerous medication names, often many more than the number of actual discharge medications. No discharge communication occurred in intensive care as all patient transfers from the area were to a hospital ward.

10

Table 4. Attributes of Pharmacist-Patient Communication Interactions at Admission

Patient admission number

Setting Length of interaction

Number of regular medications discussed at admission

Number of PRN medications discussed at admission

Presence of a medication list or medication aid

Presence of a family member at admission

1, part 1a CTU 17m42s 4 2 Yes Wife 1, part 2a CTU 05m28s 2 0 Yes Wife 2 CTU 09m00s 9 0 Yes No 3, part 1 ONC 05m25s 5 1 No No 3, part 2 ONC 02m50s 0 3 No No 4, part 1 ONC 07m08s 8 1 Yes No 4, part 2 ONC 03m30s 7 1 Yes No 5, part 1 ONC 10m10s 10 1 No Husband 5, part 2 ONC 05m09s 3 0 No Husband 6, part 1 ONC 09m27s 8 1 No No 6, part 2 ONC 01m23s 3 0 No No 7 ONC 15m57s 4 0 No No 8 ONC 16m09s 9 1 No Son 9 ONC 24m27s 4 2 No No 10 ED 04m52s 2 0 No No 11 ED 01m32s 0 1 DAAb No 12 ED 06m05s 6 0 No No 13, part 1 ED 04m57s 4 2 DAA No 13, part 2 ED 01m29s 0 0 DAA No 14 ED 07m34s 7 0 No No 15 ED 06m11s 0 4 DAA Daughter 16, part 1 ED 02m24s 1 0 DAA Daughter 16, part 2 ED 10m47s 3 0 DAA Daughter 17, part 1 ED 10m30s 9 0 No Daughter 17, part 2 ED 00m30s 0 0 No Daughter 18 07m24s 10 0 No No 19 ED 05m33s 7 0 No No 20, part 1 ED 04m01s 3 1 DAA by

patient Husband

20, part 2 ED 10m18s 4 2 DAA by patient

Husband

21 ED 06m09s 5 0 No No 22 ED 04m39s 5 1 No Husband 23 ED 03m15s 2 0 No Daughter 24 ED 08m23s 8 2 Yes Son 25, part 1 ICU 07m27s 7 1 Yes Granddaughter 25, part 2 ICU 01m44s 1 0 Yes Granddaughter 26 ICU 04m43s 4 0 No No

a Reasons for divided (part1/part2) admission interactions included: pharmacists returned to patients to give feedback on the medication

reconciliation process, interruptions to the admission process by other health care professionals; pharmacists returned to patients to obtain further details of medications or pharmacists returned to patients to inspect medications brought in by a family member. b DAA Dosage Administration Aid

CTU cardiothoracic care; ONC oncology care; ED emergency care; ICU intensive care

11

Table 5. Attributes of Pharmacist-Patient Communication Interactions at Discharge

Patient Discharge Number

Setting Length of interaction

Number of medications

mentioned at discharge taken by

patient prior to hospital stay

Number of new medications

mentioned at discharge

Total number of medication

names mentioned at

discharge

Presence of family

member

1 CTU 14m09s 18 1 22 Daughter

2 CTU 18m01s 5 3 16 Daughter

3 CTU 14m27s 0 5 16 No

4 CTU 21m47s 3 5 17 No

5 CTU 14m24s 3 3 17 Husband

6 CTU 17m04s 0 5 9 No

7 CTU 16m26s 6 3 17 Wife

8 ONC 03m23s 1 1 2 Wife

9 ONC 01m01s 0 1 1 No

10 ONC 02m54s 0 3 5 No

11, part 1a ONC 02m33s 3 1 6 No

11, part 2a ONC 00m49s 3 1 6 No

12 ONC 00m48s 0 2 2 No

13 ONC 08m33s 5 4 10 No

14, part 1 ONC 00m49s 2 0 4 No

14, part 2 ONC 00m59s 0 1 2 No

15 ONC 01m47s 3 0 4 Son

16 ONC 06m32s 3 3 7 Wife

17 ED 08m53s 4 1 8 No

18 ED 07m23s 3 1 4 No

19 ED 01m29s 1 1 2 No

20, part 1 ED 00m38s 0 3 3 No

20, part 2 ED 02m52s 0 3 6 No

21, part 1 ED 10m03s 8 1 15 No

21, part 2 ED 07m59s 4 2 9 No

22 ED 01m43s 0 3 3 Partner

23, part 1 ED 02m44s 0 3 3 No

23, part 2 ED 05m54s 0 3 5 No

24, part 1 ED 07m08s 5 4 10 No

24, part 2 ED 02m22s 0 4 7 No

25 ED 07m36s 3 2 5 Husband & son

26, part 1 ED 00m59s 1 1 1 No

26, part 2 ED 01m55s 0 1 3 No

27, part 1 ED 00m38s 2 0 2 No

27, part 2 ED 01m28s 2 1 3 No

28 ED 09m06s 5 1 7 Husband

29, part 1 ED 00m36s 0 1 2 Mother

29, part 2 ED 06m04s 0 4 11 Mother

30 ED 01m49s 2 1 3 Wife

31 ED 11m51s 1 2 3 Daughter

32 ED 08m47s 0 3 10 Husband

33 ED 01m41s 0 1 5 Relative

34, part 1 ED 04m17s 8 0 10 Daughter

34, part 2 ED 03m43s 1 2 6 Daughter

35, part 1 ED 01m30s 5 2 8 Husband

35, part 2 ED 03m10s 0 2 6 Husband

a Divided discharge interactions occurred as pharmacists sometimes gathered preliminary information from the patient before dispensing discharge medications

CTU cardiothoracic care; ONC oncology care; ED emergency care

12

Four major themes were evident in the data: pharmacist information gathering at admission, pharmacist information providing at discharge, words and phrases used in pharmacist-patient interactions and patient engagement. Themes and subthemes are listed in Table 6.

Table 6. Themes and Subthemes arising from analysis of Interviews and Observations of Hospital Pharmacists and Patients

Pharmacist information gathering at admission Structured admission communication Direct and focused patient questioning Scanning for problems Multiple admission interactions Pharmacist information providing at discharge Structured discharge communication Risk of overloading patients with information Undesirable and noisy environments Words and phrases used in pharmacist-patient interactions No agenda communicated or negotiated Pharmacists’ language ambiguous Multiple medication names Patients not confident with medication information Patient engagement Unilateral information transmission Limited patient contributions and questions Family members acting as interpreters Interruptions and distractions

Pharmacist Information Gathering at Admission Structured admission communication When communicating with patients at admission, pharmacists focused on gathering specific information to document the best possible medication history. Pharmacists garnered information from multiple sources, but patient interview was the customary starting point. Pharmacists used a medication history on admission form to guide their spoken communication with patients, leading to a similarity of talk among the different pharmacists. This structured form produced by the health care organization, once completed, was filed in a patient’s medical record. Information collected by pharmacists, as directed by the form, included the names, forms, strengths, doses, routes and frequency of administration of patients’ medications. Pharmacists also collected data on adverse drug reactions and allergies. A cardiothoracic pharmacist at interview described:

13

So we’re guided by our [medication history on admission] form, which we fill in. On admission we interview the patient. So every patient is interviewed within 48 hours of coming into hospital. So we can get the sources from different- We try the patient – that’s the best source.

Direct and focused patient questioning Using the medication history on admission form, pharmacists formulated many

direct questions to elicit the information necessary to complete the document. With these questions, pharmacists tested patients’ knowledge and recall of their current medications. Pharmacists did not routinely explore patients’ understandings of their medications, such as the general indication for a medication (e.g. for the heart), or how a medication related to their medical condition (e.g. reduce the risk of clots in the coronary arteries following a heart attack), or patients’ awareness of the main side effects of their medications. In the following observation excerpt, the pharmacist was completing a medication history for a patient admitted to intensive care with hyponatremia:

Pharmacist (Pha): Alright. So, we’ve got, so Atacand [candesartan cilexetil] you take two tablets every morning?

Patient (Pat): Yes. Pha: Okay. What about the Nexium [esomeprazole]? Pat: Yes. Pha: How many do you have a day? Pat: I have one. Pha: One in the morning? Pat: Yes. Pha: Okay. And this one here Vasocardol [diltiazem hydrochloride], Pat: Yes. Pha: you have one tablet a day? One capsule? Pat: Yes. Pha: In the morning or at night? Pat: Yes. Pha: One in the morning? Pat: In the morning.

Pharmacists aimed to “control” the topic and content of patient communication

interactions to ensure specific and timely information was gathered. This aim was revealed in a reflective teaching session we observed between an oncology pharmacist and pharmacy intern:

Pha: And he [the patient] liked to chat. You kind of lost a little focus here and there.

Pharmacy intern (PhaI): Mm. Pha: So I felt like you could have taken more control of the interview. You

know what you PhaI: Yep. Pha: need to ask the patient. You know what information you need to get from

them.

14

Scanning for problems While listening and documenting patient responses to obtain a medication history,

pharmacists scanned the information provided for evidence of medication problems such as drug interactions, medication misuse and lack of medication adherence. Pharmacists, when confronted with reports of potentially problematic medication management, usually directed and focused questions to clarify the issue. With these questions, however, pharmacists infrequently explored physical, emotional, social or financial reasons for the medication problem. This was evident in the following admission interview we observed between an intensive care pharmacist and family member. The patient was admitted for treatment of cardiac ischemia and his granddaughter assisted with the management of his seven medications:

Pha: And plus the patch [Transiderm nitro patch (glycerol trinitrate)]. Is- would it be possible for you to bring if you’ve got a box that’s filled?

Granddaughter (Gra): They finished. Pha: Oh. Gra: It all finished. Pha: Okay. Gra: And before he came, Pha: Okay. Yep. Gra: for two days he didn’t use the patch because he had run out. Pha: Oh. Okay. Alright. Gra: Just letting you know. Pha: Okay. So we’ll do some more patches. If he’s still on it when he goes home. Gra: Yep. Pha: What about the other tablets? Is there anything that you’re running low

on? Gra: He’s run out. Yeah. Pha: Oh, he’s run out of everything? Gra: Yeah. Pha: Okay. That’s fine.

In this interaction the pharmacist discovered the patient had not refilled his

medications and was therefore not taking them. However, the pharmacist did not explore any reasons for why the patient had run out of his medications, such as financial issues with refilling the prescriptions or because he did not have any more prescriptions, or he did not let his granddaughter know.

Multiple admission interactions

To assist with the efficient acquisition of accurate, contemporary and complete information from patients, pharmacists attempted to inspect patients’ actual medications as well as refer to medication lists. However, many patients either did not have a medication list or the list was not in their possession. Additionally, sometimes patients did not bring their medications into hospital, or if the medications were present at admission, they were sent home before a pharmacist could inspect them. Thus, in all study environments, initial admission communication interactions between pharmacists and patients were often prolonged, and the admission process was commonly completed in parts. To accomplish an accurate medication history

15

pharmacists often visited patients many times in a day and engaged with community based health care professionals. A conversation we observed between an oncology pharmacist and a patient, provided insights into some of the challenges faced by pharmacists when gathering information:

Pat: If I’ve got my wallet, wallet with me I’ve got a [medication] list in there. Pha: Oh, do you? Where about is your wallet? Pat: My wallet is at home. Pha: Do you have an arm band on? I’ll double check. [Checks patient]. So it says

here- but it doesn’t actually have your allergies. I’ll have a look at your history and then I’ll come back and hopefully that will jog your memory. Okay? I’m just going to borrow these [medications] to write down the doses, and give the local pharmacy a call as well.

Pharmacist Information Providing at Discharge Structured discharge communication A major component of pharmacist-patient communication was pharmacists providing information to patients going home with medications. Pharmacist-patient communication occurred in both spoken and written forms and was usually delivered in a single interaction at discharge. The exception to this situation was in cardiothoracic care where the pharmacist sometimes met a patient on more than one occasion to discuss discharge medications; in particular, the anticoagulant warfarin. As patients admitted to cardiothoracic care generally stayed for more than a day, pharmacists were able to return to patients for repeated education sessions. However, in emergency care, we noted pharmacists delivered warfarin education in a single discharge interaction. Owing to a government policy mandating the discharge or movement of patients from the emergency setting within a four-hour period, pharmacists delivered warfarin education in restricted timeframes.

In the oncology, cardiothoracic and emergency settings, where patients were discharged home, pharmacists usually supplied patients on multiple medications with a medication list. The hospital’s pharmacists generated printed lists with individualized information on a patient’s generic and brand name medications, the medication strength and dose to take. They also included information such as directions and time for medication administration, the general purpose of the medication, and special instructions. Pharmacists used the structure of the printed information categories to configure their spoken communication to patients, resulting in a similarity of talk among varying pharmacists. A cardiothoracic pharmacist supplied a medication list for a patient going home on eight medications following his admission for chest pain. We observed the pharmacist communicate with the patient and his daughter:

Pha: Okay. Alright. There we go. Here we go. So, this is a [medication] list, of everything that we want you to take, okay? So how you read it – this column here is the name of your medication, or the ingredient and the strength.

Pat: Yeah. Pha: [In] this column are the brand names that you can buy from the pharmacy Daughter (Dau): Yeah. Pha: because they often come in different brands.

16

Dau: Yeah. Pha: But as long as they’ve got that ingredient and that strength, they’re all the

same. Okay? Pat: Oh yeah. Pha: Then directions. You hold it so you can see. Then we’ve got directions in

this column and then number of tablets to take, then what time of day to take them.

Dau: Oh yeah.

In this excerpt, the pharmacist orientated the patient and his daughter to the format of the medication list. This explanation was provided so the patient and his daughter could follow the pharmacist’s talk during the interaction. The pharmacists’ communication digressed minimally from the structured printed medication list, and invited little participation from the patient or his daughter.

Risk of overloading patients with information

Pharmacists frequently communicated large quantities of information to patients at discharge. They often transmitted considerable information pertaining to a single medication. This lengthy process risked overloading patients with information. Pharmacists in all settings shared this view. At interview, a cardiothoracic pharmacist commented:

I think getting a lot of information at once is a major issue. You know, even for a particularly, you know, intelligent person it can be a lot. Three other people have seen them and then we’re seeing them. And if we’re seeing them on the day of discharge that’s particularly the risky period because there’s too much information. They’re itching to go home. So I think it’s a lot of information at once.

Undesirable and noisy environment Interruptions to communication were common. Noise from vacuum cleaners,

overhead announcements, alarms, pagers, and conversations, was ever present in all the study settings. We observed some pharmacists to be desensitized to the intrusion of noise, as they would seldom pause when speaking to patients when loud noises occurred. Patients, however, at times redirected conversations with pharmacists to inquire about the source of a noise, or to request information to be repeated. Interruptions to pharmacist-patient admission and discharge communication were also commonplace. As these interactions usually lasted an average of eight to ten minutes, sometimes multiple interruptions occurred in a single encounter. Sources of interruptions were predominantly electronic pagers, phone calls for pharmacists and patients, nursing staff clarifying or delivering information, offers for tea and coffee, and doctors’ ward rounds. In an interview, an emergency pharmacist cited the distraction of a busy and stressful work environment to be of concern when conveying medication information. The pharmacist also suggested that patients in emergency did not want to be involved in lengthy conversations about their medications after they were told they could go home:

Well ED [emergency department] is a busy place, and a distracting place. And it’s not the best place to have a half-hour conversation about warfarin. And it’s also

17

the government doesn’t want us to have a half-hour conversation about warfarin in an ED – they want us to kick ’em out of EDs! So you’ve got the pressures of the bed-management side of it, you’ve also got the patient who- they’ve been told they can leave – and they want to leave.

In this excerpt, the pharmacist conveyed concern about problems specific to the

emergency setting, as well as his lack of control over the issues. Productivity and efficiency concerns were ever present in the emergency setting because of the risk of financial penalties if patients were not moved from the area within a four-hour time period. The pharmacist perceived this restricted timeframe to be incongruent with quality patient communication.

Words and Phrases used in Pharmacist-Patient Interactions When communicating with patients, pharmacists used words and phrases that were sometimes ambiguous and resulted in miscommunication as meaning was lost or misinterpreted. Patients, during admission interactions with pharmacists, regularly used words that expressed uncertainty in the information they were providing. No agenda communicated or negotiated

Pharmacists used words and phrases to introduce the admission process to patients or their family members that rarely indicated the communication agenda. A typical introduction by the pharmacist when meeting a patient to complete a medication admission was “My name’s [first name of pharmacist] I’m the pharmacist. I was hoping to ask you a few questions?” As pharmacists gave patients little insight into the purpose and extent of admission communication interactions, patient input was generally restricted.

Similarly, at discharge, pharmacists routinely indicated their intent to impart information, but they infrequently shared specific content details and how long the interaction would take. A common introduction to discharge medication communication was “Now I’ve got a few bits of information so I’ll go through the information leaflets at the same time as I go through the medicines.”

Pharmacists’ language ambiguous

Patients were sometimes bewildered by the words chosen by pharmacists during medication communication. Medical jargon, multiple drug names or the use of no medication names when providing information, all had the potential to confuse patients. In the following emergency admission interaction we observed a patient interpret the word “allergies” to have a different meaning to the one intended by the pharmacist:

Pha: Do you have any allergies? Pat: Mm. Not really. No. I don’t think I would say I’m allergic. Pha: So the one we want to give you are the strong pain relievers. Pat: I am very sensitive to drugs. Pha: Mm-hmm. Pat: Because I don’t like taking drugs. Pha: Mm-hmm.

18

Pat: And I have reactions, I think it’s on the file there, to some of the tablets I have taken.

Pha: Oh, that’s what I mean by, have you got any allergies. Pat: Well allergies, yeah with, that’s it with the drugs. Yeah.

When pharmacists swapped between brand and generic medication names in

communication, some patients sought clarification. We observed the following interaction between an emergency pharmacist and a patient:

Pha: Okay. Good. And your digoxin? Pat: Oh. What’s that? Pha: Uh for your heart. Lanoxin is the brand.

Sometimes, when talking about medications with patients, pharmacists did not use

medication names. Some pharmacists chose to refer to a medication as “this one” and show patients the medication package. This lack of use of the medication name forced patients to read medication labels or rely on visual recognition of the package. Moreover, pharmacists commonly simplified their communication and spoke of a medication in general terms, rather than using the medication name. We noted pharmacists referred to medications as a “cholesterol tablet”, “fluid tablet”, “morphine like medicine”, or “steroid tablet.”

Multiple medication names

During discharge interactions, pharmacists often mentioned multiple medications in communication with patients. To educate patients on the variety of terms used to identify a single medication, pharmacists routinely spoke about alternative medication names. Furthermore, providing education about one medication frequently led to talk of other medications. A patient admitted to emergency care following a motor vehicle accident was observed to be confused when the emergency department pharmacist mentioned multiple medication names during discharge education:

Pha: . . . but at the moment your normal dose of Panamax [paracetamol or acetaminophen] would be two tablets four times a day.

Pat: Panamax? Pha: Panamax. Pat: That’s not Panadol Forte is it? Pha: Osteo? Pat: No, there’s three, we’re talking about, you, you’re talking about Panamax. Pha: Yep. Pat: That’s, that’s an aspirin isn’t it? Pha: No. They’re all paracetamol. Pat: Right. So that doesn’t have the other ingredients in it? Pha: No, this is purely paracetamol. Pat: That’s right. Pha: So- Pat: So that’s just like a, er, ordinary aspirin. Pha: Aspirin? Pat: Well’s it like taking Panadeine. Panadol. Pha: It is-

19

Pat: But it’s bet- Pha: Panadol is Panamax. They’re the same thing. Pat: Oh. I see. Pha: But Panadol Osteo is different.

Patients not confident with medication information

Patients often used words during admission interviews that conveyed to pharmacists that they were not confident in the information they were providing. Words such as “I think” were common in patients’ language, as were descriptions of medications, when patients struggled to recall medication names and details. In the following interaction a patient was admitted to cardiothoracic care for a percutaneous coronary intervention. The patient was observed to be unsure of the details pertaining to a medication when speaking to the pharmacist:

Pat: It [vitamin D] comes with the uh calcium tablet I think? Pha: It does. Okay. Alright. Does it- Pat: [to wife] Darling it comes with a calcium tablet doesn’t it? Wife: Yeah – but I don’t know what’s on the calcium tablet.

Patient Engagement Patient engagement with pharmacists in communication was variable. During admission interviews, patients spoke frequently, but information was mostly offered only in response to pharmacists’ questioning. In contrast, when receiving discharge education, patients were overwhelmingly passive as pharmacists provided information. Patients asked pharmacists minimal questions during admission or discharge interactions. Factors such as patients’ knowledge of their medications and their ability to speak and understand English, affected their engagement in communication. Pharmacists’ workloads and the location of the interaction also affected pharmacist-patient engagement. Unilateral information transmission

Across all settings, we identified that pharmacists’ use of a structured communication process at discharge to convey medication information, did not facilitate patient participation in conversations. Pharmacists mainly imparted information in a unilateral manner to patients, rather than using language that would promote a joint transaction. They often transferred large amounts of information without establishing what particular information a patient needed. Sometimes pharmacists declared their intention to convey medication information with statements that did not encourage patient engagement. An oncology pharmacist announced “So we’ve got all your discharge medications here for you. I’ll go through them for you.”

Pharmacists did not routinely check if a patient had used a medication before providing education. Even when pharmacists became aware that a patient had pre-existing experience with a medication, some pharmacists continued to supply information without exploring the patient’s knowledge and understandings of the medication. In the following observation excerpt, a patient with an infected finger was being discharged home on an antibiotic he had previously taken. The oncology

20

pharmacist, however, did not acknowledge the patient’s prior experience with cephalexin, and did not check what the patient knew about it before providing information:

Pha: It’s called cephalexin. Pat: Yep. Pha: Alright? So the directions are one, four times a day Pat: Yep. I’ve had it before. Pha: with or without food it doesn’t matter Pat: Yep. Pha: until you’re all done so that will be five days. Pat: Yep.

Limited patient contributions and questions

During pharmacists’ discharge information delivery, patients and family members largely provided receipt of information by offering confirming utterances, as pharmacists dominated the communication interaction. In the following observation excerpt, a cardiothoracic pharmacist demonstrated communicative dominance over the patient:

Pha: Now what this medication [pantoprazole] is for, is to protect your stomach against any reflux

Pat: Oh I see. Pha: or feeling like you’re going to vomit Dau: Yeah. Pat: Okay. Pha: or any of that heartburn. Okay? Dau: Yeah. Pha: That’s to help reduce that. It’s also there to prevent any stomach ulcers as

well. Dau: Mhm. Pha: So we don’t want any bleeding in your stomach. This will stop that. Dau: Yeah. Pha: Okay? Pat: Okay. Pha: Now with this one, it’s important that you swallow it whole. Don’t cut it or

crush it or chew it. Pat: No. Dau: No. Pat: No. Pha: It’s got a special coating for it to work. It won’t work if you cut it in half or

anything like that. Dau: Okay. Pat: No. Pha: Okay? So one every morning. Pat: Okay.

21

A practice evident across all settings was that pharmacists rarely posed questions to patients relating to a specific medication to check for patients’ understandings during discharge education. Rather, pharmacists reserved close-ended and nonspecific questions such as “No questions at all for us?” or “Does that make some sense?” or “So, any questions?” until the end of their information delivery. To check if patients were following the information provided, pharmacists frequently used nonspecific words such as “Okay?”. Patients rarely posed questions in response to such general queries.

Although most patients attempted to contribute information about their medication history during admission interviews, many patients did not know their medications’ names, actions or strengths. Patients’ limited recall of medication details restricted pharmacist-patient communication as pharmacists pursued other sources to obtain the necessary information. Some patients were very clear about their limited medication knowledge, as we observed in the emergency care setting:

Pha: Good. Do you know the names of the medicines? Pat: Yes. Coversyl Plus [perindopril and indapamide] Pha: Just bear with me. [Pharmacist writing] Pat: You don’t want to ask me how many, how many milligrams or whatever

they are. I wouldn’t know.

Family members acting as interpreters In observations, pharmacists frequently used family members to engage patients

in communication. Professional interpreters often proved difficult to arrange in short timeframes and at suitable times for the pharmacist. A cardiothoracic pharmacist expressed at interview that she rarely used professional interpreter services:

Um, I’ve only once or twice used the interpreter at [hospital name]. I find it’s really hard to get in with the interpreter because they book it usually for the ward round or the doctors, and I might not be available then.

Using family members as an interpreter, however, risked miscommunication.

Communication breakdown was likely when pharmacists used medical terminology and a family member was unfamiliar with the term in the patient’s native language. We observed this issue arise in the emergency care setting:

Pha: Okay. So I’ll show you the first one. The first one is Stemetil [prochlorperazine]. I don’t know whether you want to translate as we go or if-

Son: I would not know, I would not know how to translate that. Interruptions and distractions

Patient engagement with pharmacists was also shaped by the time available with a pharmacist and a patient’s physical state. Pharmacists sometimes communicated the demands on their time to patients, which had the potential to curtail pharmacist-patient engagement. We observed a busy pharmacist in emergency care say to a patient during medication education, “Um, sorry. I’m just so over run today. They’re [medical and nursing staff] nagging me.”

Moreover, patients in all settings were frequently tired from their illness and lack of sleep. Some patients’ abilities to engage with pharmacists and retain information

22

were subsequently impeded. This was evident in the following observation excerpt where a patient in emergency care fell asleep while the pharmacist was talking:

Pha: We only suggest one spray. Under your tongue. Pat: Mm. Pha: Okay? After you’ve done that spray, we get you to wait for a few minutes,

about five minutes. If the pain is still there after five minutes you can do a second spray, okay? Are you okay? [To patient, but patient is silent, as he has fallen asleep]

Dau: Yeah he’s a bit tired. Pha: Okay. Long night.

Discussion The results from our study provide new insights into pharmacist-patient communication at admission and discharge transition points in specialty hospital settings. Pharmacists regulated communication with patients to accomplish the tasks of medication admission and discharge. When providing or gathering information, pharmacists controlled the level of patient engagement, shaping the communication dynamics and flow of the interaction. Pharmacists sometimes used ambiguous words when communicating with patients, changing the way admission and discharge interactions unfolded. At times, miscommunication resulted. Patients sought minimal information during communication encounters, posing minimal questions to pharmacists.

Pharmacists’ communication with patients during admission and discharge encounters was not generally patient-centered. Pharmacists employed a structured communication approach in all the study settings to strategically collect and deliver specific information in an efficient and timely manner. In turn, they were able to see multiple patients with diverse disorders, in areas with high patient throughput. This ability to see multiple patients enabled pharmacists to achieve one of their key performance indicators, which was to complete a medication history for all patients within 48 hours of hospital admission. However, the use of a structured approach for admission and discharge communication militated against pharmacists posing open-ended questions to patients. Patients during, or at the end of, interactions with pharmacists were not encouraged to express concerns or ask questions regarding their medications. Moreover, pharmacists did not tailor information to suit patients’ individualized needs. To implement patient-centered principles pharmacists could ask patients at the start of an interaction to identify any problems or issues, and raise any particular information needs that they would like addressed in the interaction. A mutually negotiated agenda, which takes both the patients’ and pharmacists’ needs into account, could then lead to a personalized collaborative discussion.

Vogelsmeier et al [16] reported similar findings in regard to medication reconciliation forms. Doctors, nurses and pharmacists employed in Veterans Administration hospitals equated using the highly structured forms in patient interactions to a “mechanistic ‘checklist’ task.” Consistent with our results, Murad et al [9] demonstrated that hospital and community pharmacists asked patients more close-ended than open-ended questions, and Greenhill et al [24] found that hospital and

23

community pharmacists did not routinely encourage patient involvement in consultations. Although, Hargie et al [12] reported that community pharmacists valued building rapport through patient engagement as the most important skill for effective communication, we found this was not practiced by hospital pharmacists. Relationship building between patients and hospital pharmacists was hampered because pharmacists dominated interactions, limiting patient participation in admission and discharge communication.

Patient engagement with pharmacists was limited as pharmacists carried out medication conversations in undesirable circumstances, such as when patients were feeling tired and unwell. Pharmacists conversed with patients at times that matched their availability, as opposed to times that best suited the patient, as they were mindful that medication admission processes needed to be completed within 48 hours. Factors such as waiting to speak to family members or community based health care professionals, or waiting to view a patient’s medication or medication list, resulted in delays in finishing this process. In addition, sometimes pharmacists had minimal warning that patients were to be discharged and thus they communicated with patients at the last minute. Furthermore, on occasions, pharmacists’ access to patients being discharged was hampered by the presence of other health professionals. Pharmacists, therefore, conversed with patients when there was an opportunity, rather than when they were most receptive to education.

Sometimes pharmacists’ methods of communication with non-English speaking patients did not reflect a patient-centered approach. We did not observe professional medical interpreters to be used in any pharmacist-patient communication interactions. On the occasions where patients did not speak or understand English, pharmacists relied on using family members to interpret. This practice was routine in all the observed study settings, despite some family members declaring they did not know how to translate the medical terminology used in the conversation. Pharmacists favored the use of family members to act as an interpreter because they were often present at a patient’s bedside at the same time pharmacists were available. Outcomes included restricted information sharing and constrained communication flow. Given we found that many patients lacked detailed medication knowledge, our study highlights the potential risk for medication incidents when pharmacist-patient communication is suboptimal.

Despite involving patients in long discharge or repeated admission encounters, most pharmacists failed to inform patients about the scope and purpose of an interaction, or to negotiate an agenda. Reasons for this lack of collaboration about medication communication included time constraints, heavy workloads, and overfamiliarity with admission and discharge processes from repeatedly using the same communication script. Babalola and Erhun [17] demonstrated communication breakdown between hospital pharmacists and patients owing to pharmacists’ undeclared intentions at the commencement of an interaction. Although the hospital setting and medication admission and discharge processes were routine for pharmacists, these procedures and the environment were often unfamiliar to patients. Strengthening the need for pharmacists to negotiate with patients an agenda for their communication encounters is that the role of hospital pharmacists is not well understood by all patients. [14, 24, 25] Furthermore, patients and pharmacists differ in their perceptions of a pharmacist’s role in creating a patient-centered relationship and

24

in interpersonal communication. [26] Based on our findings, we believe that a clear and agreed agenda for pharmacist-patient interactions could enhance patient contributions and cultivate a more equitable pharmacist-patient relationship, thereby maximizing opportunities for effective communication.

Time constraints also played a role in pharmacists’ limiting patient engagement in communication interactions. Echoing our findings, Dutton et al [27] have shown that the process of gathering information from patients to complete a medication admission is a time consuming task for pharmacists. Restrictions on pharmacists’ time, and use of structured documents, might explain their use of a biomedical approach to communication. It might also reflect a need for pharmacist skill development. [24] While pharmacists’ fostering of patient engagement during admission and discharge communication might take longer to accomplish, it could expand the scope of information sharing and improve the quality of the interactions. [9, 17] In turn, if pharmacists have in-depth knowledge of patients’ expertise in, and perspectives of, medication management, this information can be shared with other health care professionals to optimize patient health outcomes. Additionally, the potential to mitigate medication incidents arises when clear and open communication occurs between pharmacists and patients. [28]

Even though most patients responded to pharmacists’ questions for the purpose of obtaining a medication history at admission, pharmacists did not seek to establish patients’ understandings of their medications. Past research undertaken in the UK on pharmacist-patient communication concurred with our findings. [24] Pharmacists posed questions during admission interactions that were designed to attain specific medication details, not to probe patients’ medication understandings. Rather, they established patients’ abilities to recall their medication names, forms, strengths, doses, routes and frequencies, instead of seeking patients’ comprehension about their medications. At discharge, pharmacists focused their communication on the transmission of often large quantities of medication information. Similarly, Pilnick [18] identified pharmacists’ information giving, to which patients intermittently responded with acknowledging utterances, to be a common approach used by hospital pharmacists at discharge. In our study, pharmacists exhibited unilateral communication with patients by rarely repeating or summarizing information, seldom testing patients’ knowledge of information provided, and largely ignoring patients’ prior experiences with medications.

Furthermore, opportunities for patients to achieve understandings about their medications were moderated by pharmacists’ dominance during discharge communication, and through pharmacists’ use of ambiguous language to describe medications. Pharmacists in their communication to patients sometimes altered between saying generic and brand medication names, and on occasions they mentioned no medication names at all, even though patients were being discharged on high risk medications. Some patients asked clarifying questions in response, but many did not. Pharmacist also typically imparted medication information before asking patients if they had any questions at the end of the encounter. However, nonspecific and close-ended queries posed by pharmacists are not always perceived by patients as a question or an invitation to open discussion. [13] When patients neither ask questions about their medications, nor engage in a discussion with pharmacists, they are unlikely to develop their medication knowledge.

25

The origins of pharmacists’ disinclination to explore patients’ understandings of their medications appear to be multifactorial. Exploring understandings is a complex and time-consuming task for pharmacists. In all settings, our results identified multiple job demands, and pressures to discharge patients to prevent bed-flow obstructions, as pervasive influences on pharmacist-patient communication. Babalola and Erhun [17] and Nair et al [29] reported that the scope of pharmacists’ work activities was restricted because of pharmacists’ limited time and heavy workloads. It is also possible that miscommunication between pharmacists and medical staff incited succinct pharmacist-patient communication. [30] Limited patient communication with pharmacists occurred when medical and nursing staff neglected to forewarn pharmacists of a patient’s impending discharge. Pharmacists subsequently shortened patient interactions so not to delay hospital discharge processes. Short consultations have been shown to reduce patient participation and satisfaction with health care professionals. [31]

Of concern, if patients lack detailed understandings of their medications, their ability to effectively communicate problems or concerns to pharmacists and other health care professionals could be impaired. As such, opportunities for pharmacists to institute appropriate and timely medication aids or support services for patients might be overlooked. Patients might subsequently be placed at risk of experiencing preventable medication incidents. Patients are best positioned to make informed and collaborative decisions with pharmacists and other health care professionals if they possess comprehensive understandings of their medications. [10] Although pharmacists detected medication incidents in some admission and discharge interactions, the potential to identify more medication issues exists if pharmacists test and explore patients’ understandings of their medicines during communication interactions.

Hospital pharmacists worked professionally and efficiently under restrictive timeframes and in situations of high patient throughput. They interviewed and educated many patients, and detected numerous medication discrepancies, particular at medication admission interactions. Hospital pharmacists also regularly completed extensive documentation. These actions were accomplished in an organizational environment that highly valued rapid and continuous patient flow. Hospital pharmacists also functioned in noisy conditions and endured multiple interruptions to their interactions with patients. In addition, pharmacists were required to work within the confines of hospital-mandated documentation and communicate information without easy access to interpreter services. Accordingly, from our findings, we have identified several implications for fostering effective communication in practice.

Pharmacists could divide the medication discharge process into multiple interactions to stage discharge information delivery and prevent overloading patients with information. Repeated contact with patients would also enable pharmacists to explore patients’ understandings of information conveyed and their preferred learning style. Pharmacists could then customize education to patients’ specific needs. It is likely that patients who are well informed on their medication details could in turn contribute thorough and accurate medication information during any future admission communication with pharmacists. Pharmacists need to enable patient engagement in communication interactions through the application of patient-centered principles. Information sharing between pharmacists and patients about medications needs to

26

expand beyond what is prescribed in structured communication documents. Opportunities for active patient involvement in communication with pharmacists could be invited through agreed agendas for discussion, private and interruption free environments, and encouraging words from pharmacists that foster patient contributions. Further, hospital managers need to ensure medical interpreters are readily available for pharmacists to utilize when required. Additionally, pharmacists may benefit from training on when to employ interpreter services.

We conducted this study at a single public metropolitan hospital and therefore our findings might not be transferrable to other metropolitan or rural or private hospital settings. In these settings, medication admission and discharge processes might vary, altering pharmacist-patient communication interactions. Our presence in the hospital environment might have caused pharmacists or patients to change their communication. However, because we collected more than 200 hours of observations, after a short time our presence drew very little attention.

Conclusion Effective pharmacist-patient communication is vital to successfully and safely navigate the complexities of medication management across admission and discharge transition points. In this article, we highlight the need for pharmacists in their communication with patients to transcend the restriction of structured communication tools used to frame admission and discharge interactions. Pharmacists need to apply patient-centered communication principles to equalize the imbalance in the pharmacist-patient relationship and facilitate patient engagement in conversations. Active patient involvement in communication with pharmacists enables opportunities for shared medication understandings to be attained, and for medication issues to be identified and addressed in a timely manner. Thus, open and effective communication is a central component of pharmacist-patient interactions for endorsing safe medication management in specialty hospital settings. Acknowledgments and Disclosures The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The authors disclose receipt of the following financial support for the research, authorship, and or publication of this article: This study was supported by the Australian Research Council Discovery Project [DP1093038]. The authors wish to thank the staff and patients from the hospital who participated in this project.

27

References 1. Spinewine, A., Claeys, C., Foulon, V., Chevalier, P. (2013). Approaches for improving continuity of care in medication management: a systematic review. International Journal for Quality in Health Care 25:403-417. 2. Vira, T., Colquhoun, M., Etchells, E. (2006). Reconcilable differences: correcting medication errors at hospital admission and discharge. Quality & Safety in Health Care 15:122-126. 3. Tam, V.C., Knowles, S.R., Cornish, P.L., Fine, N., Marchesano, R., Etchells, E.E. (2005). Frequency, type and clinical importance of medication history errors at admission to hospital: a systematic review. Canadian Medical Association Journal 173:510-515. 4. Department of Health. (2013) Quality Use of Medicines. Available at: http://www.health.vic.gov.au/qum/hrm.htm. (last accessed October 4 2013]. 5. Department of Health. (2012) Supporting patient safety. Sentinel event program annual report 2010-11. Available at: http://docs.health.vic.gov.au/docs/doc/F55ED6726B0250DECA257A690079AAFC/$FILE/AnnualReport010-11%5BBookmarkedweb version%5D.pdf. (last accessed October 4 2013]. 6. Nichols, P., Copeland, T., Craib, I., Hopkins, P., Bruce, D. (2008). Learning from error: identifying contributory causes of medication errors in an Australian hospital. The Medical Journal of Australia 188:276-279. 7. Pilnick, A. (1999). Patient counseling by pharmacists. Sociological Quarterly 40:613-622. 8. Bauman, A.E., Fardy, H.J., Harris, P.G. (2003). Getting it right: why bother with patient-centred care? Medical Journal of Australia 179:253-256. 9. Murad, M.S., Chatterley, T., Guirguis, L.M. (2013). A meta-narrative review of recorded patient–pharmacist interactions: Exploring biomedical or patient-centered communication? Research in Social and Administrative Pharmacy: in press. 10. Chewning, B., Sleath, B., L. (1996). Medication decision-making and management: A client centered model. Social Science & Medicine 42:389-398. 11. de Oliveira, D., Brummel, A., Miller, D. (2010). Medication therapy management: 10 years of experience in a large integrated health care system. Journal of Managed Care Pharmacy 16:185-195. 12. Hargie, O.D., Morrow, N.C., Woodman, C. (2000). Pharmacists’ evaluation of key communication skills in practice. Patient Education and Counseling 39:61-70. 13. Shah, B.K., Chewning, B. (2011). Concordance between observer reports and patient survey reports of pharmacists’ communication behaviors. Research in Social and Administrative Pharmacy 7:272-280. 14. LeBlanc, J., Dasta, J. (2005). Scope of international hospital pharmacy practice. Annals of Pharmacotherapy 39:183-191. 15. Mueller, S., Sponsler, K., Schnipper, J. (2012). Hospital-based medication reconciliation practices: a systematic review. Archives of Internal Medicine 172:1057-1069. 16. Vogelsmeier, A., Pepper, G.A., Oderda, L., Weir, C. (2013). Medication reconciliation: A qualitative analysis of clinicians' perceptions. Research in Social and Administrative Pharmacy 9:419-430. 17. Babalola, O., Erhun, W. (2001). Pharmacist medication history interview of elderly patients in a teaching hospital in Nigeria. International Journal of Pharmacy Practice 9:97-103.

28

18. Pilnick, A. (2003). "Patient counselling" by pharmacists: Four approaches to the delivery of counselling sequences and their interactional reception. Social Science & Medicine 56:835-849. 19. Polit, D., Beck, C. (2006). Essentials of Nursing Research. 6th ed. Philadelphia: Lippincott Williams & Wilkins. 20. Lincoln, Y., Guba, E. (1985). Naturalistic Inquiry. California: Sage Publications. 21. Denzin, N. (1989). The Research Act: A Theoretical Introduction to Sociological Methods. Englewood Cliffs, New Jersey: Prentice Hall. 22. Polit, D., Beck, C. (2010). Generalization in quantitative and qualitative research: Myths and strategies. International Journal of Nursing Studies 47:1451-1458. 23. Ritchie, J., Spencer, L. (1994) Qualitative data analysis for applied policy research. In: Analyzing Qualitative Data (ed. Bryman, A., Burgess, R.G.). pp. 173-194. London: Routledge. 24. Greenhill, N., Anderson, C., Avery, A., Pilnick, A. (2011). Analysis of pharmacist–patient communication using the Calgary-Cambridge guide. Patient Education and Counseling 83:423-431. 25. Morecroft, C., Thornton, D., Caldwell, N. (2013). Inpatients' expectations and experiences of hospital pharmacy services: qualitative study. Health Expectations. 26. Worley, M.M., Schommer, J.C., Brown, L.M., Hadsall, R.S., Ranelli, P.L., Stratton, T.P., Uden, D.L. (2007). Pharmacists' and patients' roles in the pharmacy-patient relationship: Are pharmacists and patients reading from the same script? Research in Social and Administrative Pharmacy 3:47-69. 27. Dutton, K., Hedger, N., Wills, S., Brown, D., Davies, P. (2003). Prevent medication errors on admission. Clinical Governance: An International Journal 8:128-137. 28. Hume, A., Kirwin, J., Bieber, H., et al. (2012). Improving care transitions: Current practice and future opportunities for pharmacists. Pharmacotherapy 32:e326-e337. 29. Nair, K., Dolovich, L., Cassels, A., McCormack, J., Levine, M., Gray, J., Mann, K., Burns, S. (2002). What patients want to know about their medications. Focus group study of patient and clinician perspectives. Canadian Family Physician 48:104-110. 30. Gallagher, R.M., Gallagher, H.C. (2012). Improving the working relationship between doctors and pharmacists: Is inter-professional education the answer? Advances in Health Sciences Education 17:247-257. 31. Freeman, G.K., Horder, J.P., Howie, J.G.R., Hungin, A.P., Hill, A.P., Shah, N.C., Wilson, A. (2002). Evolving general practice consultation in Britain: issues of length and context. British Medical Journal 324:880-882.