Measuring and Evaluating Indicators of Appropriate Prescribing in Older Populations Cahir C., Teeling M., Teljeur C., Bennett K., Fahey T. HRB PhD Scholar.

Measuring and Evaluating Indicators of Appropriate

Prescribing in Older Populations

Cahir C., Teeling M., Teljeur C., Bennett K., Fahey T.

HRB PhD ScholarDivision of Population Health SciencesRCSI

Caitriona CahirHRB PhD ScholarDivision of Population Health SciencesRCSI

Potentially inappropriate prescribing and cost

outcomes for older people: a national population study

C Cahir, T Fahey, M Teeling, C Teljeur, J Feely, K Bennett

Structure of Presentation

• Brief background to study

• Objectives

• Methods

• Results

• Summary

Background

Prescribing for older people is a challenging process

• Multi-morbidity

• Physiological changes

– Pharmacokinetics e.g. digoxin and toxicity

– Pharmacodynamics e.g. benzodiazepines and falls

– Cognitive impairment e.g. dementia

Background

• How do we measure appropriate prescribing in older people?

• Appropriateness of prescribing – assessed by process or outcome measures that are implicit (judgment based) or explicit (criterion based)

• Explicit process measures of potentially inappropriate prescribing (PIP)

– Beers criteria (International)– STOPP screening tool (Irish & UK)

Objectives

1. Estimate the prevalence of PIP in a national Irish older population in 2007 using Beers and STOPP criteria

2. Investigate the association between PIP, number of medications, gender and age

3. Establish the associated cost of PIP

Method

• Retrospective national population study (aged ≥70 years) using the HSE-PCRS for a one year period (2007) (n=338,801)

• Health Services Executive Primary Care Reimbursement Services (HSE-PCRS) – prescription claims, WHO ATC, gender, age

• Beers (independent of diagnosis) and thirty STOPP indicators applied

RESULTS

OVERALL

PIP Overall Prevalence Rates (n=338,801)

Beers STOPP

ONE PIP 25.31% 24.78%

TWO PIP 6.39% 8.08%

THREE PIP 1.27% 2.28%

OVERALL PIP 33.22% 35.84%

PIP and Number of Medications

• Strong association between PIP and the number of different medications

– Beers (10 drugs vs none) OR=9

– STOPP (10 drugs vs none) OR=50

Association between the number of different medications and PIP (STOPP) in 2007

0

10

20

30

40

50

60

1 2 3 4 5 6 7 8 9 10+

No of drug classes (vs none)

Od

ds

rati

o

* Linear and quadratic trend p<0.0001

PIP and Gender and Age

• PIP more likely in females vs males – Beers (1.58, 1.56 to 1.60)– STOPP (1.10, 1.08 to 1.12)

• PIP more likely in those aged >75yrs vs. 70-74yrs– Beers (1.22, 1.20 to 1.24)– STOPP (1.28, 1.26 to 1.30)

• Additionally adjusting for number of medications– Beers Gender (1.46, 1.44 to 1.48) – STOPP Gender (0.91, 0.90 to 0.93) – Beers Age (0.98, 0.97 to 1.00)– STOPP Age (0.95, 0.93 to 0.96)

RESULTS

INDIVIDUAL CRITERIA

Beers Adjusted Analysis 20075 highest prevalence rates per drug class

(n=338,801)

BEERS DESCRIPTION PREV % OR GENDER

(F vs M)

OR AGE

(>75 vs 70-74)

CNS Long acting benzodiazepines

(risk of falls, fractures)

12.93% 1.67

(1.63– 1.70)

0.88

(0.87-0.90)

Cardio Doxazosin

(hypotension, dry mouth, urinary problems)

5.29% 0.99

(0.96-1.02)

0.82

(0.79- 0.85)

CNS Doses of short acting benzodiazepines

(smaller doses more effective)

4.96% 1.45

(1.41- 1.50)

1.19

(1.15-1.23)

Gastro Gastrointestinal antispasmodic drugs

(highly anticholinergic and uncertain effectiveness)

2.53% 1.30

(1.24-1.36)

0.91

(0.87-0.95)

CNS Amitriptyline

(anticholingeric and sedation properties)

2.40% 1.42

(1.35-1.49)

0.74

(0.70-0.77)

STOPP Adjusted Analysis 20075 highest prevalence rates per drug class

(n=338,801)

STOPP DESCRIPTION PREV % OR GENDER

(F vs M)

OR AGE

(>75 vs 70-74)

Gastro PPI > 8 weeks

full therapeutic dose

(dose reduction, discontinuation)

16.69% 0.80

(0.78-0.81)

1.05

(1.02-1.07)

Musco NSAID >3M

(simple analgesics preferable)

8.76% 1.25

(1.22-1.28)

0.78

(0.76-0.81)

CNS >1M Long-acting benzodiazepines

(risk of falls, fractures)

5.22% 1.72

(1.65- 1.78)

0.89

(0.87-0.92)

Duplicates NSAIDs, SSRIs, Antidep, ACE, Loop diuretics, opioids

(optimisation of monotherapy)

4.78% 1.19

(1.15-1.23)

0.74

(0.71-0.76)

Cardio Beta-blocker with COPD(risk of increased bronchospasm)

2.34% 0.53

(0.51-0.56)

0.84

(0.80-0.89)

RESULTS

COSTS

Overall Costs

• Beers maximum costs

– Net ingredient cost €10,712,129

– Total expenditure €15,478,526

• STOPP maximum costs

– Net ingredient cost €38,664,640

– Total expenditure €45,631,319

Beers Cost Analysis 2007 Highest costs as a proportion of overall costs (NIC= €10,712,129 and Total Exp=€15,478,526)

BEERS DESCRIPTION NIC € TOTAL EXP €

Doxazosin 5,400,793 (50%) 6,196,854 (40%)

Long-acting benzodiazepines 890,252 (8%) 2,397,549 (15%)

Nifedipine 849,995 (8%) 1,151,294 (7%)

Fluoxetine 800,637 (7%) 958,362 (6%)

Muscle relaxants and antispasmodics 738,213 (7%) 916,362 (6%)

Total of other criteria 2,032,239 (19%) 3,858,104 (25%)

STOPP Cost Analysis 2007 Highest costs as a proportion of overall costs (NIC=€38,664,640 and Total Exp= €45,631,319)

STOPP DESCRIPTION NIC € TOTAL EXP €

PPI > 8 weeks full therapeutic dose 22,352,240 (58%) 24,715,010 (54%)

Neuroleptics >1M

Neuroleptics >1M with Parkinsonism

Anticholingerics for side-effects

5,612,192* (15%) 6,079,905 (13%)

Duplicate drugs 4,531,160 (12%) 5,499,118 (12%)

NSAID >3M

Warfarin and NSAIDs

3,969,629* (11%) 5,050,640* (11%)

Long-acting benzodiazepines >1M 572,009 (1%) 1,352,209 (3%)

Total of other criteria 1,627,410 (4%) 2,934,437 (6%)

* Adjusted for claimants receiving the same medication per more than one criteria

Summary

• High prevalence of PIP in older populations in Ireland per Beers and STOPP criteria

• The greater the number of medications the more likely PIP

• Differences in prevalence rates across gender and age groups

• Significant costs

Acknowledgements

• Dr Stephen Byrne, School of Pharmacy and Dr Denis O’Mahony,Department of Medicine, UCC for developing the STOPP criteria

• Health Services Executive Primary Care Reimbursement Services (HSE-PCRS)

• The Health Services Research Institute - the Irish Health Research Board’s (HRB) Cross-Institution PhD Scholars Programme in Health Services Research

• HRB Centre for Primary Care Research

Duplicates and PPIs

• Duplicates – 4.78%– Two concurrent NSAIDs (2.22%)

Diclofenac with glucosamine or nimesulide– Two concurrent opioids (1.24%)

Tramadol with codeine and morphine

• PPIs– Co-prescriptions (asp, warfarin, NSAID) – Duration and dosage