+

Clinical Practice Tools for

Identifying Potential

Medication-Related

Problems in the Elderly

ASCP 45th Annual Meeting and ExhibitionOrlando, FLNovember 5, 2014William Simonson, PharmD, CGP, FASCPIndependent Consultant PharmacistSenior Research Professor (Pharmacy Practice)Oregon State University

+Learning Objectives

After attending this educational presentation the participant

should be able to:

List the names of at least three tools to identify ADEs, MRPs, or PIMs

Identify the relative value of these tools in affecting patient outcomes

Describe how these tools can be incorporated into day-to-day

practice

+Disclosures

The speakers have no financial relationships to disclose

+PIMs and ADEs

What Tools are Available?

Beers List

Medication Appropriateness Index

IPET

Zhan-AHRQ

Medication Regimen Appropriateness Index

STOPP/START

IMAP

NCQA-HEDIS

Beers Criteria

+Definitions and Terminology

MRP- Medication Related Problem: An event or situation involving drug

therapy that negatively interferes with a patient’s health

Polypharmacy – 5 or more medicines?, 6.1? 9 or more medicines?

“Administration of more medicines than are clinically indicated, representing

unnecessary use”

PIM – Potentially Inappropriate Medication

DIM – Definitely Inappropriate Medication

PIP – Potentially Inappropriate Prescription

PIPE – Potentially Inappropriate Prescribing in the Elderly

PPO – Potential Prescribing Omissions

DAE – Drug to be avoided in the elderly

DRP – Drug-Related Problem

ADWE – Adverse Drug Withdrawal Events

Suboptimal prescribing

+Potentially Inappropriate Medications

Emphasis on “Potentially”

PIM ≠ DIM

Consider the individual patient

Experts don’t always agree on “inappropriate”

Delphi technique v. evidence-based methods

Actual harm vs. predicted Harm – High “signal

to noise” ratio

No harm no foul?

PIM identification is only a starting point

+True or False?

The STOPP criteria have been proven to identify

medications that are definitely inappropriate for use by

seniors in nursing facilities.

+Consequences of MRPs and PIMs

Hospitalization

↑ length of stay

ADR

ADE

Inefficient resource use

Financial waste

Polypharmacy

Medication errors

Therapeutic failure

Poor QOL

Morbidity and mortality

↑Illness duration

NF placement

Functional decline

Social decline

+The Beers List

In 1991, Dr Mark Beers published a paper with

explicit criteria to identify potentially inappropriate

medication (PIM) use in nursing home residents.

Delphi technique (also referred to as GOBSAT)

Update published in 1997 to apply to the elderly,

wherever they reside. Updated again in 2003.

Most recent update - 2012 American Geriatrics

Society

+Medication Appropriateness Index

Developed in 1992 by expert team based on clinical experience and background literature Serves as sensitive measure of potential improvement in prescribing quality secondary to

clinical pharmacist intervention May be applicable as quality of care outcome measure in health services research or in

institutional quality assurance programs Measures prescribing appropriateness according to ten criteria for each medication prescribed

Appropriate Marginally appropriate Inappropriate

It does not address under-prescribing. Clinical expertise is required to apply some of the criteria.

Requires at minimum, medical history, problem list, and medication list Barrier: 10 minutes/drug

Hanlon JT. J Clin Epidemiol 1992:45:1045-1051.

+Medication Appropriateness Index

1. Is there an indication for the drug?

2. Is the medication effective for the condition?

3. Is the dosage correct?

4. Are the directions correct?

5. Are the directions practical?

6. Are there clinically significant drug-drug interactions?

7. Are there clinically significant drug-disease interactions?

8. Is there unnecessary duplication with other drugs?

9. Is the duration of therapy acceptable?

10. Is this drug the least expensive alternative compared to

others of equal utility?

Min = 0 = Completely appropriate

Max = 18 = Completely inappropriate

+IPET

Improved Prescribing in the Elderly Tool

1997 consensus-based mail survey of 32 member panel

from Canadian medical centers (included 8 pharmacists)

List of 38 high-risk prescribing situations in an elderly

population, primarily contraindicated drugs and drug-disease

interactions

McLeod PJ et al. Can Med Assoc J 1997;156:385-391.

+IPET Example

Practice

Long-term prescription of long t1/2 benzodiazepines to treat insomnia

Mean clinical significance

3.72

Risk to patient

May cause falls, fractures, confusion, dependence and withdrawal

Alternative therapy

Nondrug therapy or short t1/2 benzodiazepine

% of panel members who agreed with alternative

97%

McLeod PJ et al. Can Med Assoc J 1997;156:385-391.

+Zhan-AHRQ

Looked at PIMs in community-dwelling elderly in 1996 using

Medical Expenditure Panel Survey representing 33.2 million lives

Expert panel of 7 members (geriatricians,

pharmacoepidemiologist, pharmacist) classified 33 drugs from

1997 Beers drugs into three categories: Always avoid (used by 2.5% of study population)

Rarely appropriate (used by 9.1% of study population)

Some indications (used by 13.3 % of study population)

Most use considered inappropriate

Zhan C et al. JAMA 2001;286:2823-29.

+ Assessing Care of Vulnerable Elderly (ACOVE):

Quality Indicators for Appropriate Medication Use in

Vulnerable Elders

RAND Corporation, 2001, developed quality indicators to examine the quality of

medical care for the vulnerable elderly in the US

Vulnerable elderly – community-dwelling persons expected to die or become severely

disabled within next 2 years

The most comprehensive examination to date

Combination of clinical evidence and expert opinion

ACOVE Phase 3 added new indicators for: COPD, colorectal cancer, breast cancer,

sleep disorders, BPH

Knight EL. Ann Int Med 2001;135:703-710

+ Assessing Care of Vulnerable Elderly (ACOVE):

Quality Indicators for Appropriate Medication Use in

Vulnerable Elders

ACOVE Quality Indicators

Drug indication—clearly defined in record

Patient education—purpose, how to take, expected side effects, important ADEs

Medication list—up-to-date, in record

Response to therapy—documented within six months

Periodic drug regimen review—at least annually

Monitoring warfarin therapy—INR w/in 4 days and at least every 6 weeks

Monitoring of diuretic therapy—electrolytes w/in 1 week and yearly

Avoid use of chlorpropamide as hypoglycemic agent, due to long half-life, serious

hypoglycemia

Avoid drugs with strong anticholinergic properties when possible

Avoid barbiturates—potent CNS depressants, low therapeutic index, highly

addictive, multiple drug interactions, increase risk for falls/fractures

Avoid meperidine—increased risk for delirium

Monitor renal function and potassium in patients prescribed ACE inhibitors w/in 1

week

Knight EL. Ann Int Med 2001;135:703-710

+Medication Regimen

Complexity Index (MRCI)

Developed under the assumption that complexity of drug therapy involves more than number and types of

medications

Developed by researchers and expert panel

Tool consists of three sections

Dosage form

Dosing frequency

Additional directions

MRCI is a sum of the 3 sections -- higher scores, more complex regimen

Drugs include Rx, OTC, nutritional supplements, health products, dermatologicals, short-term medications

(e.g. antibiotics)

Requires 2-8 minutes per regimen, depending on complexity

Possible use

Risk assessment tool

To predict health outcomes

To identify patients who would benefit from additional services

Reporting drug regimen data

Research tool

George J. Ann Pharmacother 2004;38:1369-76.

+NCQA-HEDIS (2006)

2002, Secretary of HHS called for national action plan to ensure appropriate use of therapeutic

agents in the elderly population

NCQA convened expert consensus panel using modified delphi technique to identify rates of

inappropriate prescribing in the elderly

Panel classified the 2003 beers drugs as follows

Always avoid

Rarely appropriate

Some indications

Drugs in the always avoid and rarely appropriate composed the 2006 Health Plan Employer

Data and Information set (HEDIS) measure to assess quality of care of older Americans

Percent of persons receiving at least 1 HEDIS criteria drug

Male 19.2%

Female 23.3%

Pugh MH et al. J Manag Care Pharm. 2006;12:537-45.

+NCQA-HEDIS (2014)

National Committee for Quality Assurance, Health Care

Effectiveness Data and Information Set (HEDIS)

Continues to assess % of Medicare members ≥age 66 who receive

high-risk medications

Based on 2012 Beers Criteria

↓ use of high-risk medications is an opportunity to reduce costs and

encourage clinicians to prescribe safer alternative medications

Many other HEDIS measures are reported

+STOPP/START

Screening Tool of Older Persons Potentially Inappropriate

Prescriptions

Identifies commission errors

Comprehensive list of geriatric PIMs

Screening Tool to Alert Doctors to the Right Treatment

Identifies omission errors

Recommends beneficial medications for specific conditions

Developed in 2008 by European geriatricians using Delphi consensus

technique and clinical evidence

Inter-rater reliability: proportion of positive agreement

STOPP 87%

START 84%

Gallagher P et al. Int J Clin Pharmacol Ther 2008;46:72-83

+Selected STOPP Items

Thiazide diuretic with diagnosis of gout

Calcium channel blocker with constipation

Tricyclic antidepressants with dementia

PPI for PUD @ full dose for >8 weeks

Regular opiates >2 weeks with chronic constipation without

laxative

High risk drugs in fallers (psychoactive Rx, vasodilators,

diphenhydramine, etc.)

+Selected START Items

Warfarin in chronic atrial fibrillation

ACE inhibitor with chronic heart failure

Antidepressants in severe depression >3 months

Bisphosphonates when taking chronic corticosteriod Rx

Ca++/Vit D in osteoporosis

+IMAP

Individualized Medication Assessment and Planning tool.

Developed in 2011 for use by ambulatory care pharmacists and for

research

IMAP based on the best of existing tools

Easy to use

Applicable to ambulatory care

Intuitive (easy identification of MRP category and recommendation)

MRP clearly defined and distinctive

Reliable and valid

Crisp GD et al. Am J Geriatr Pharmacotherapy 2011;9:451-460.

+IMAP

Developed for ambulatory care pharmacists

Guide RPhs’ comprehensive assessment of a pt’s medication use to

identify MRPs

Provide RPh with mechanism for classifying:

Clinically meaningful information to describe each MRP

Their plan to address and resolve each MRP

Crisp GD et al. Am J Geriatr Pharmacotherapy 2011;9:451-460.

+The Beers “List” – What Is It?

Beers Criteria, not Beers List

Most recent update - 2012 American Geriatrics Society Updated

Beers Criteria published on-line (americangeriatrics.org) Evidence-based with recommendations, based on risk v benefit assessment

Strength of evidence: strong, weak, insufficient

Quality of evidence: high, moderate, low

Well-known and respected, but not necessarily well-understood

+The Beer’s List – What Is It Not?

A list

A tool to identify “forbidden” drugs in the elderly

A resource that everyone agrees on

A resource that always improves clinical outcomes

+Take away Points

Many different tools to identify MRPs, PIMs, etc. etc.

Consider how they were developed

Consider strengths v. weaknesses

Consider what they are designed to do

“Potential” problem vs. “actual”problem

The tool doesn’t rule - never lose sight of the individual

patient

+PIMs and ADEs: What’s the Evidence of Harm

ASCP 45th Annual Meeting and Exhibition

Orlando, FL

November 5, 2014

H. Edward Davidson, PharmD, MPH

Assistant Professor of Internal Medicine

Eastern Virginia Medical School

Partner, Insight Therapeutics, LLC

Norfolk, VA

+Are we looking in the right places?

Does the evidence build a strong case that

PIMs are contributing to increased

hospitalization or death in older individuals?

Or

Are there others areas that we should be focusing on

in order to reduce ADEs in older individuals?

+Hill’s Criteria of Causation (1965)

Strength of Association: The larger the relative effect, the more likely

the causal role of the factor.

Dose-response: If the risk increases with increasing dose of the risk

factor, the more likely the causal role of the factor.

Consistency: If similar associations are found in different studies in

different populations, the more likely the causal role of the factor.

Temporality: Risk factor exposure must precede the outcome.

Intervention: Reduction or removal of the risk factor must reduce the

risk of the outcome.

Biological Plausibility: The association agrees with currently accepted

understanding of pathological processes.

Coherence: Associations between the risk factor and the outcome must

be consistent with existing knowledge.

31

+The Evidence

Lau DT et al. Arch Intern Med 2005;165:68-74. Sample: nursing home 65 and over, MEPS NHC, PIM = Beers

2003 Design: Retrospective cohort study, N=3,372 Measures: use of PIM during 1 year period PIMs increased risk of hospitalization: OR 1.28 (1.10-1.50) PIMs increased risk of death: OR 1.21 (1.00-1.46) Most frequent PIMs: narcotics, antihistamines,

sedative/hypnotics, GI antispasmotics, antidepressants, platelet inhibitors, iron supplements

Limitations: did not find dose-response/duration effect, no causality assessment

32

+The Evidence – cont.

Hamilton H et al. Arch Intern Med 2011;171:1013-19. Sample: hospitalized 65 and over, PIM = STOPP/Beers 2003

Design: Prospective cohort study, N=600

Measures: WHO-UMC ADE causality and expert panel consensus

STOPP PIMs contributed to hospitalization; OR 1.85 (1.51-2.26)

Beers PIMS did not; OR 1.28 (0.95-1.72)

Most common PIMs: benzodiazepines, antihypertensives, opiates

Limitations: did not include OTC meds, duration of use not

determined

33

+The Evidence – cont.

Pasina L et al. Clin Pharm Ther 2014;39:511-515. Sample: hospitalized 65 and over, PIM = Beers 2003/2012

Design: Cross-sectional study, N=844

Measures: use of PIM at hospital discharge, re-hospitalization or

death within 3 months

No significant association with re-hospitalization: OR 0.77 (0.48-

1.19)

No significant association with mortality: OR 0.84 (0.44-1.52)

Most frequent PIMs (2012): ticlopidine, antiarrhythmic drugs,

alpha blockers, benzodiazepines

Limitations: conducted in Italy, did not assess adherence, no

causality assessment

34

+The Evidence – cont.

Fick DM et al. Res Nursing Health 2008.;31:42-51.

Sample: MCO 65 and older; PIM: Beers 2003

Design: Retrospective cohort study, N=17,971

Measure: health care utilization over 6 months, PIM use

PIMs increased risk for hospitalization: OR 1.99 (1.76-2.26)

Most frequent PIMs: estrogen only, propoxyphene,

benzodiazepines, digoxin, NSAIDs

Limitations: did not consider diagnosis or condition criteria

35

+The Evidence – cont.

Dedhiya et al. Am J Geriatr Pharmacother 2010.

Sample: Medicaid 65 and older; PIM: Beers 2003

Design: Retrospective cohort study, N=7,594

Measure: PIM use

PIMs increased risk for hospitalization: OR 1.27 (1.10-1.46)

PIMs increased risk of death: OR 1.46 (1.31-1.62)

Most frequent PIMs: inappropriate drug choice category

Limitations: retrospective, no causality assessment

36

+The Evidence – cont.

Reich O et al. PLos ONE 2014;9. Sample: Health Claims (Swiss) 65 and older; PIM: Beers

2012, PRISCUS (German)

Design: Retrospective cohort study, N=49,668

Measure: PIM use and hospitalization at 1 year

PIMs increased risk for hospitalization: One PIM: OR 1.13

(1.07-1.19), 3 or more PIMs: 1.63 (1.40-1.90)

Most frequent PIMs: not reported

Limitations: did not consider diagnosis/condition criteria,

retrospective, no causality assessment

37

+PIMs, Pharmacist Intervention, and Hospitalization

Cochrane Collaboration review – 2012 4 studies addressed PIM use (Beers 2003, MAI), pharmacist

intervention, and hospitalization rate

Overall, a significant reduction in MAI score post

intervention noted

One of 4 studies …

reported significant reduction in hospitalization rate in

intervention group (22% reduction)

limitations: significant differences in comorbidities

between groups and small sample size (N=69)

38

+Quick Review

PIMs and hospitalization – low to moderate

level data suggest relationship, but….

PIMs and death – as above

STOPP vs Beers – STOPP appears to be more

sensitive for harm (one study)

+True or False?

When referring to causation, temporality describes

the increase in risk of an adverse drug event with

increasing dose of medication.

+ Adverse Drug Events and the Elderly

Individuals > 65 yrs more likely than younger to suffer an ADE; RR 2.4 (95% CI 1.8-3.0)

Budnitz DS et al. JAMA 2006:296:1858-66

Budnitz et al. New Engl J Med 2011;365:2002-12.

Estimated Rates of Emergency Hospitalizations for Adverse Drug Events in Older Adults, 2007-2009

+

+Executive Summary

653 Medicare beneficiaries discharged from hospitals

to SNF for post-acute care (35 days or less)

Assessed for adverse events (AE) (SNF Trigger Tool)

and temporary harm and if preventable

2 stage attribution process; screener, MD panel

22% experienced an AE during SNF stay

59% were deemed preventable

11% experienced harm (60% hospitalized)

44

+

+Independent Risk Factors for Having a Preventable ADE in NFs

Risk Factor Odds Ratio 95% CI

Male 0.55 0.30 - 0.99

No. regularly scheduled meds

0-4

5-6

7-8

>=9

1.0

1.7

3.2

2.9

Referent

0.83 - 3.5

1.4 - 6.9

1.3 - 6.8

New resident+ 2.9 1.5 -5.7

+within 60 days of admission

Field TS, Gurwitz JH et al. Arch Intern Med 2001;161:1629-34.

+What about ….

Therapeutic failure

Adverse drug withdrawal events

Contribution of declining kidney function

Medication reconciliation

+Take Away

PIM tools do . . .

Educate health care team

Raise awareness

Make you think

PIM tools do not . . .

Always prevent harm

Work well to identify those at risk for harm