Can clinical pharmacy services have a positive impact on drug-related problems and health outcomes in community-based older adults?

J.T. Hanlon et al.

The American Journal of Geriatric Pharmacotherapy

Can Clinical Pharmacy Services Have a Positive Impact on Drug-Related Problems and Health Outcomes in Community-Based Older Adults? Joseph T. Hanlon, PharmD, MS, 14 Catherine I. Lindblad, PharmD, 4,5 and Shelly L. Gray, PharmD, MS 6

1Division of Geriatric Medicine, Oepartment of Medicine, School of Medicine, 2Oepartment of Pharmacy and Therapeutics, School of Pharmacy, University of Pittsburgh, and 3Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, Pennsylvania, 4Institute for the Study of Geriatric Pharmacotherapy and Department of Experimental and Clinical Pharmacology, College of Pharmacy, University of Minnesota, 5VA Minneapolis Medical Center, Minneapolis, Minnesota, and °School of Pharmacy, University of Washington, Seattle, Washington ABSTRACT Background: Although pharmacotherapy can be beneficial in the elderly, it can also lead to drug-related problems (DRPs), including untreated indications, drug use without an indication, improper drug selection, subtherapeutic dosage, overdosage, medication error, medication nonadherence, drug interactions, adverse drug reactions, adverse drug withdrawal events, and therapeutic failure. Objective: The goal of this article was to review evidence from randomized controlled studies to determine whether DRPs and the related health outcomes can be modified by providing clinical pharmacy services for the elderly in community-based settings. Methods: Randomized controlled studies that assessed DRPs and health outcomes in persons aged >65 years after pharmacist interventions were identified through searches of MEDLINE (1970-March 2003), the Cochrane Database of Systematic Reviews (through March 2003), and International Pharmaceutical Abstracts (1966-March 2003). The search combined the terms clinical pharmacists, aged, outcomes, and randomized controlled trial. A manual

search of the reference lists of identified articles and the authors' own materials was also conducted. Results: This literature review included 14 randomized controlled studies: 5 involved interventions in the home health setting, 3 were instituted at hospital discharge with home follow-up, 3 were clinic based, 1 was conducted in the community pharmacy setting, and 2 were conducted in long-term care facilities. These studies provided considerable evidence that clinical pharmacy interventions reduced the occurrence of DRPs in the elderly but showed limited evidence that such interventions reduced morbidity, mortality, or health care costs. Conclusions: Future large multicenter studies are necessary to test the cost-effectiveness of clinical pharmacy services for the community-based elderly and the impact of these services on such health outcomes as use of health services, timed functional-status measures, and adverse drug reactions. ( A m J Geriatr Pharmacother. 2004;2:3-13) Copyright © 2004 Excerpta Medica, Inc. Key words: elderly, drug therapy, health care quality, pharmacist, randomized controlled trial.

INTRODUCTION Although pharmacotherapy can be beneficial in the elderly, it can also result in drug-related problems (DRPs), including untreated indications, drug use without an indication, improper drug selection, subtherapeutic dosage, overdosage, medication error, medication nonadherence, drug interactions, adverse drug reactions (ADRs), adverse drug withdrawal events (ADWEs), and therapeutic failure (TF). 1,2 An Institute of Medicine report cited DRPs, particularly

ADRs, as a major safety concern. 3 Studies conducted in the outpatient and nursing home settings have reported an incidence of ADRs ranging from 2.5% to 50.0% in elderly individuals. 4 Applying a structure-process-outcomes model to the listed DRPs, only ADRs, ADWEs, and TF are patient health outcomes, and the remainder are process mea-

AcceptedforpublicationSeptember22,2003. Copyright© 2004 ExcerptaMedica, Inc.

Volume 2 ,* Number 1

1543-5946/04/$19.00

March2004

3

The American Journal of Geriatric Pharmacotherapy ] J.T. Hanlon et al.

sures, s The distinction between process and outcome measures is important, because the ultimate goal of improving drug therapy is to improve health outcomes. Although it is assumed that improving drug-process measures will lead to better health outcomes, this is not always the case. In addition to ADRs, ADWEs, and TF, a number of other important health outcomes are related to drug therapy, including the "D's": death, disease, dollars, disability, discomfort, and dissatisfaction. 6,7 Two cost-of-illness analyses conducted in the United States suggest that morbidity and mortality associated with DRPs cost an estimated $177.4 billion per year for ambulatory patients 8 and an estimated $4 billion per year for nursing home residents. 9 The involvement of a pharmacist in the care of older adults is one of the most common techniques for reducing DRPs. A number of authors have described the provision by clinical pharmacists of specialized pharmaceutical care for elderly patients1°-12; however, to the best of our knowledge, there has been no comprehensive review of the impact of such clinical pharmacy services on DRPs and health outcomes in elderly persons residing in community settings (comprising ambulatory care and long-term care facilities). Moreover, individuals in community settings are important patient populations in geriatrics, representing a large number of consumers of large numbers of medications. These patients' health problems, including DRPs, are often treated by clinicians in the outpatient clinic and nursing home settings. Assuming that clinical pharmacy services have potential benefits, questions remain concerning their effectiveness and magnitude of effect. The objective of this article was to review evidence from randomized controlled studies to determine whether DRPs and the related health outcomes can be modified by providing clinical pharmacy services for the elderly in community-based settings.

MATERIALS AND METHODS Articles that assessed DRPs and health outcomes in elderly individuals after receipt of clinical pharmacist interventions were identified through searches of MEDLINE (1970-March 2003), the Cochrane Database of Systematic Reviews (through March 2003), and International Pharmaceutical Abstracts (1966-March 2003). The search combined the terms clinical pharmacists, aged, outcomes, and randomized controlled trial. Additional publications were identified by a manual search of the reference lists of identified articles and the authors' own materials. 2,4,1°-23 The

,

I

authors then reviewed the identified studies and included those that (1) had a randomized controlled design; (2) used _>1 DRP process measure (eg, medication adherence problems); (3) measured >1 patient health outcome (eg, ADRs, death, disability); and (4) involved community-based elderly individuals.

RESULTS The initial literature search identified 20 publications describing 18 studies. Four were excluded: 1 for having no control group, 24 another for having a quasiexperimental study design, 25 and 2 because they included only DRP process measures. 26,27 Thus, this review included 16 articles describing 14 randomized controlled trials of clinical pharmacist interventions designed to improve medication-use process measures and health outcomes in the elderly. Five studies were conducted in the home health setting (Table 1),28-32 3 were instituted at hospital discharge with home followup (Table II), 33-36 3 were conducted in clinic-based settings (Table I11),37-40 1 was conducted in the community pharmacy setting, 41 and 2 were conducted in long-term care facilities (Table IV). 42,43 The control groups received usual care in all studies. Interventions in the Home Health Setting Sidel et al28 conducted a randomized controlled trial of an 11-month clinical pharmacist intervention in 284 older adults living at home and at high risk for DRPs. Clinical pharmacists paid home visits and provided telephone follow-up as needed to patients receiving the intervention. Pharmacists developed patient-specific medication information packets, cleaned patients' medicine cabinets, counseled patients on good medicationtaking practices, and stressed good communication with health care providers. The physician was contacted if the patient experienced an ADR or the pharmacist needed additional information about the patient to make decisions regarding therapy. There was no between-group difference in DRP process measures. At the end of the study, patients receiving the intervention had significantly fewer outpatient physician visits compared with those in the control group (-0.69 vs 0.22, respectively; P < 0.05). The published report did not mention whether measurement of outcomes was blinded. This study was limited by its use of unconventional medication-use measures and the lack of concordance between improvement in a single health outcome and any of the DRP process measures. Nonetheless, this clinical intervention was associated with a reduction in 1 aspect of health services use.

J.T. Hanlon et al. I The AmericanJournal of Geriatric Pharmacotherapy

Table I. R a n d o m i z e d c o n t r o l l e d t r i a l s of clinical p h a r m a c y i n t e r v e n t i o n s (vs usual care) f o r e l d e r l y indiv i d u a l s in the h o m e health setting.

Setting

No. Randomized

Sidel et at28

NewYork

284

Patient-specific information about drugs provided, patients' medicine cabinets cleaned, good medication-taking practices and good communication with health care providers encouraged; in-home visit at least twice in 6-11 mo, with telephone follow-up as needed

11 mo

P: no consistent betweengroup differences in change in knowledge, attitudes, or practices with regard to medications; Q: intervention patients had significantly fewer outpatient physician visits (P < 0.05)

Begley et a129

England

190

5 Home visits at which patient counseling was given

12 mo

P: improved compliance (P < 0.05); Q: reduced outpatient visits (P < 0.05)

Krska et al 3°

Scotland

332

Pharmaceutical care plan completed and implemented

3 mo

P: resolution of more DRPs in intervention group vs control group (P < 0.05); O: no between-group differences in SF-36 scores or health services utilization

Meredith et a131 NewYork, Los Angeles

317

DRR to identify problems; structured templates to address problems were presented to physicians

From 6 wk to 90 d

P: less therapeutic duplication (P < 0.05); more appropriate use of cardiovascular drugs (P < 0.05); O: no betweengroup differences in deterioration in health, falls, or health services utilization

Goodyer et a132 Patients with stable chronic heart failure in London

100

3 Home visits at which intensive counseling and compliance aids were provided

3 mo

P: intervention group had improved compliance scores and medication knowledge (both, P < 0.001); O: more patients in intervention group were free of peripheral edema (P < 0.01) and had better exercise tolerance (P < 0.05)

Reference

Intervention

Duration/ Follow-up

Results

P = medication-use process measures; 0 = health-outcome measures; DRPs = drug-related problems; SF-36 = Medical Outcomes Study 36-Item Short Form Health Survey; DRR = drug regimen review.

In a randomized controlled trial from England, Begley et al29 evaluated an intervention in which a clinical pharmacist paid 5 home visits to 190 elderly persons over 12 months to counsel them about compliance and medication management. The initial assessments were blinded, but the study report did not mention the use o f blinding at follow-up. Patients in the intervention group had significantly better compliance than those in the control group (86% vs 69%, respectively; P < 0.05)

and required significantly fewer outpatient visits (54% vs 79%; P < 0.05). This study was limited by its small sample size, which prevented examination o f the impact o f the intervention on other health outcomes, such as costeffectiveness. Nonetheless, this home-based pharmacy intervention was effective in reducing certain DRPs and improving certain related health outcomes. Krska et al30 reported on a randomized controlled trial from Scotland that evaluated the effect o f a clini-

5

The American Journal of Geriatric Pharmacotherapy I J.T. Hanlon et al.

Table II. R a n d o m i z e d c o n t r o l l e d trials o f clinical p h a r m a c y i n t e r v e n t i o n s (vs usual care) f o r e l d e r l y indiv i d u a l s at hospital d i s c h a r g e w i t h h o m e f o l l o w - u p .

Reference Lipton et a133 Lipton and Bird 34

Setting Greater San Francisco Bay area

Nazareth et a135 England

AI-Rashed et a136

England

No. Randomized

Intervention

Duration/ Follow-up

Results

736

Consultation before hospital discharge and at 1, 2-4, 8, and 12 weeks after discharge; physician contacted about problems

3 and 6 mo

P: at 3 months, patients receiving intervention were receiving more appropriate regimens (P = 0.01), fewer medications (P < 0.05), and less complex regimens (P < 0.05), and had better drug knowledge and compliance (both, P < 0.05); O: at 6 months, no impact on health services utilization or costs

362

Consultation at hospital discharge and within 2 wk after discharge

3 and 6 mo

P: no between-group difference in medication adherence or knowledge; O: no between-group difference in health services utilization, general well-being, or satisfaction

Consultation before hospital discharge and at 1 and 3 wk after discharge

2-3 wk and 3 mo

P: better compliance in intervention group (P < 0.001); O: fewer outpatient physician visits and hospital readmissions (both, P < 0.05)

83

P = medication-use process measures; O = health-outcome measures.

cal pharmacist medication review of DRPs, healthrelated quality of life (HRQOL), and health services utilization in 332 older adults who had >2 chronic diseases and regularly took >4 prescribed medications. Clinical pharmacists conducted in-home interviews and developed a pharmaceutical care plan. The pharmacists implemented all actions agreed on by the patients' physicians. At the 3-month reassessment, significantly more DRPs were resolved in the intervention group compared with the control group (82.7% vs 41.2%, respectively; P < 0.05). There were no significant between-group differences in HRQOL (measured in terms of Medical Outcomes Study 36-Item ShortForm Health Survey [SF-36] score) or health services utilization. The study report provided no information on whether measurement of outcomes was blinded. Although the study findings may be limited by potential contamination (participating physicians had patients in both the intervention and control groups)

,

I

and the short follow-up period, the clinical pharmacy intervention did result in greater resolution of DRPs. A randomized controlled trial by Meredith et a131 studied the effect on DRPs (as defined by an expert panel's consensus explicit criteria) of an intervention in which a clinical pharmacist and home health care nurse provided services to 317 elderly patients from 2 home health care programs. The study focused on 4 DRPs: unnecessary duplication of therapy, problems involving cardiovascular medications, use of psychotropic drugs, and use of nonsteroidal anti-inflammatory drugs. A research assistant blinded to the study groups collected data on outcome measures. Significant improvements were seen in the intervention group compared with the control group in terms of therapeutic duplication (71% vs 24%, rcspectively; P < 0.05) and appropriate use of cardiovascular medications (55% vs 18%; P < 0.05). There were no between-group differences in the use of psychotropic drugs or nonsteroidal anti-inflammatory

J.T. Hanlon et al. ] The AmericanJournal of Geriatric Pharmacotherapy

Table III. R a n d o m i z e d c o n t r o l l e d t r i a l s of clinical p h a r m a c y i n t e r v e n t i o n s (vs usual care) f o r e l d e r l y indiv i d u a l s in the clinic-based setting.

Reference

Setting

Hanlon e t a [ , 37 Veterans Cowper et a138 Affairs general medicine clinic

Zermansky et a139

General medicine practices in the United Kingdom

Gattis et al 4°

Heart failure patients seen in outpatient clinics

No. Randomized 208

1188

181

Intervention

Duration/ Follow-up

Results

DRR and written drug-therapy recommendations to physician; patient counseling at each clinic visit

12 mo

P: less inappropriate prescribing at 3 and 12 mo (P < 0.01); no between-group differences in number of medications taken, compliance, or knowledge; O: no between-group differences in ADEs, SF-36 scores, health care satisfaction, or health services utilization and related costs

DRR followed by treatment recommendations to general practitioner

12 mo

P: more drug changes (P = 0.02) and fewer repeat prescriptions (P = 0.01)in intervention group; Q: drug costs higher in control group vs intervention group (P < 0.001 )

DRR; recommendations to physician and discussions with patient concerning purpose of medication and compliance at 2, 12, and 24 wk

6 mo

P: patients receiving intervention were closer to target ACEinhibitor dose (P < 0.01); O: fewer heart failure events in intervention patients (P = 0.005)

DRR = drug regimen review; P = medication-use process measures; O = health-outcome measures;ADEs = adverse drug events; SF-36 = Medical Outcomes Study 36-1tern Short-Form Health Survey;ACE = angiotensin-convertingenzyme.

drugs, deterioration in health, falls, agency nurse visits, or duration of home health care. The results of this study were limited by the use ofunvalidated medicationuse measures and short foUow-up (from 6 weeks to 90 days). Nevertheless, the study intervention improved suboptimal drug prescribing. Goodyer et a132 conducted a randomized controlled study of a disease-specific intervention in which a clinical pharmacist provided intensive in-home medication counseling to improve compliance in 100 elderly patients with chronic stable heart failure. Study physicians were blinded to study groups when making clinical assessments. Compliance scores (as determined based on pill counts) and medication knowledge had improved significantly in intervention patients compared with control patients at 6- to 12-week follow-up (both, P < 0.001). Moreover, patients in the intervention group were able to walk a significantly longer distance on the 6-minute exercise test and reported diminished

breathlessness compared with the control group (P < 0.05). At the end of the study, a significantly greater proportion of patients in the intervention group had no peripheral edema compared with the control group (81% vs 49%, respectively; P < 0.01). Although the study had a high dropout rate (18%) and short follow-up period, patients benefited from medication counseling by a clinical pharmacist.

Interventions at Hospital Discharge with Home Follow-Up A randomized controlled trial reported by Lipton et a133 and Lipton and Bird 34 evaluated a clinical pharmacist intervention involving 736 elderly patients discharged from community hospitals and taking >3 medications for chronic conditions. The clinical pharmacist reviewed the charts and drug regimens of each patient in the intervention group to identify DRPs; if any were found, the pharmacist made recommendations to the patient's

I

?

The AmericanJournal of Geriatric Pharmacotherapy ] J.T. Hanlon et al.

Table IV. R a n d o m i z e d c o n t r o l l e d trials of clinical p h a r m a c y i n t e r v e n t i o n s (vs usual care) f o r e l d e r l y indiv i d u a l s in l o n g - t e r m care facilities.

Reference Avorn et a142

Setting 12 Massachusetts nursing homes

Roberts et a143 52 Australian nursing homes

No. Randomized 823

3230

Intervention

Duration/ Follow-up

Results

Comprehensive educational outreach program ("academic detailing"), including mailings and literature summaries

Two 30-d periods (before and after intervention)

P: greater reductions in mean scores for inappropriate psychoactive drug use (P = 0.02) and more discontinuations of psychoactive drug use in intervention patients (P < 0.05); O: disruptive behavior and staff distress similar between groups; no change in level of care, hospitalizations, or mortality

Educational session for nurses, DRR placed in charts and made available to physicians

12 mo

P: intervention reduced drug use (P = NS), particularly use of psychoactive drugs; O: no between-group changes in hospitalization, adverse events, disability indices, or survival; cost savings with intervention due to reduction in drugs

P = medication-use process measures; O = health-outcome measures; DRR = drug regimen review.

physician. The clinical pharmacist consuked with the patient before discharge and by telephone or at hospitalbased clinic visits 1, 2 to 4, 8, and 12 weeks after discharge. The study report did not describe what was done to improve medication adherence/knowledge. Using a standardized tool previously tested for reliability, a physician-pharmacist panel that was blinded to studygroup assignment evaluated the appropriateness of prescribing in a random sample of 236 patients. Based on a summary score of the appropriateness of patients' overall drug regimens, the regimens of the intervention group were significantly more appropriate than those of the control group at 12 weeks after discharge (P = 0.01). The intervention group was receiving fewer medications than the control group (5.2 vs 6.8, respectively) and less complex regimens (8.3 vs 12.0), were more knowledgeable about the purpose of their medications (95.7% vs 85.9%), and were more compliant with their regimens (96.3 vs 91.2) (all, P< 0.05). There were no between-group differences in health services utilization or costs at 6 months. The findings of this study were limited by the short follow-up period and the use of statistical methods that did not take baseline values on DRP measures into account. Nonetheless, the

8

I

intervention was associated with improvements in suboptimal prescribing and medication adherence. Nazareth et al35 conducted a randomized controlled trial of a pharmacist intervention involving 362 hospitalized patients aged >75 years who were taking >4 medications. At hospital discharge, a hospital pharmacist assessed patients' medication-management skills and provided written and verbal information to enhance adherence. This was followed within 2 weeks of discharge by a home visit by a community pharmacist, who again stressed adherence and medication knowledge. A research assistant collected information about outcome measures at 3 and 6 months. There were no significant between-group differences in hospital readmissions, use of other health services, general well-being, satisfaction, medication adherence, or knowledge at 3 or 6 months. This study was limited by its use of an intervention that did not aim to optimize the quality of medications, high dropout rates (26%), and lack of power to detect differences in hospital readmission rates (only 25% were likely to be drug related4). It illustrates a type of intervention that is unlikely to improve medication adherence and knowledge.

j.T. Hanlon et al. I The American Journal of Geriatric Pharmacotherapy

Al-Rashed et a136 studied the effect of the use of medication summaries, counseling, and a simple medicine reminder card on compliance, medication knowledge, and health services utilization in 83 hospitalized elderly patients who were prescribed ->4 drugs at hospital discharge. A pharmacist counseled the intervention group about their medications and compliance before discharge, and another pharmacist paid 2 home visits to patients (at -2-3 weeks and 3 months after discharge). The study report did not mention whether the assessment of outcomes was blinded. Compliance was significantly better at the 2 home visits in the intervention group compared with the control group (P < 0.001). At visit 2, the intervention group had significandy fewer outpatient physician visits compared with the control group (24 vs 32, respectively; P < 0.05) and significantly fewer hospital readmissions (3 vs 15; P < 0.05). Thus, the use of inpatient pharmacist counseling linked to a medication list and outpatient reminders appeared to result in better compliance, as well as a reduction in unplanned physician visits and hospital readmissions.

Interventions in the Clinic-Based Setting A randomized controlled trial by Hanlon et a137evaluated the effect of a sustained clinical pharmacist intervention involving 208 elderly outpatients at a Veterans Affairs Medical Center general medicine clinic who were taking ->5 chronic medications. A clinical pharmacist with specialty training in geriatrics met with intervention patients during all scheduled visits over a 12month period to evaluate their drug regimens and make written recommendations to patients and their physicians. Using the Medication Appropriateness Index (MAI), a reliable and valid tool, the clinical pharmacist evaluated the appropriateness of prescribing in a blinded fashion. All other outcome measures were collected by another clinical pharmacist blinded to study-group assignment. Physicians were receptive to the intervention, enacting the clinical pharmacist's recommendations for change in the intervention group more often than they instituted change independently in the control group. Based on the MAI, inappropriate prescribing scores had declined significantly more in the intervention group compared with the control group (24% vs 6%, respectively; P< 0.01) by 3 months, a change that was sustained at 12 months. Compared with the control group, fewer intervention patients experienced adverse drug events (30% vs 40%), although this difference was not statistically significant. At 12 months, there was no between-group difference

in HRQOL, as measured using the SF-36. Measures of most other outcomes remained unchanged in both groups. This group then conducted an investigation of the cost-effectiveness of the intervention in the previous trial. 38 The mean fixed costs (eg, training time for the study pharmacist, clinic orientation for the pharmacist, beeper) were $36 per patient receiving the intervention. The variable costs (eg, pharmacist's time spent delivering the intervention, physician- and patient-education supplies) were $84 per patient. Health services use and the associated costs were similar in the 2 groups. Annual drug costs were not significantly lower in the intervention group compared with the control group ($1006 vs $1096 per patient, respectively). The costeffectiveness ratio for the intervention was $7.50 per 1-unit change in prescribing appropriateness, as measured on the MAI. Despite the small, selected sample and limited generalizability to settings other than a Veterans Affairs Medical Center, this intervention improved suboptimal prescribing and was relatively inexpensive. A randomized controlled study by Zermansky et a139 examined the impact of a clinical pharmacist intervention on prescribing for elderly outpatients in the United Kingdom. The sample contained 1188 patients from 4 general medicine practices who were receiving >1 repeat prescription. No information was provided on whether measurement of outcomes was blinded. Patients seen by the pharmacist for medication review had significantly more drug changes (2.2 vs 1.9 per patient, respectively; P = 0.02) and significantly fewer repeat prescriptions (0.2 vs 0.4; P = 0.01) than those in the control group, resulting in significant cost savings (equivalent to -$100/patient per year; P < 0.001). The intervention and control groups had similar numbers of outpatient visits and hospitalizations. This study was limited by its lack of measurement of medication adherence or appropriateness or of other humanistic patient outcomes. Nonetheless, the intervention reduced medication use and the associated costs. Gattis et al40 reported on a disease-specific clinical pharmacist intervention in 181 clinic outpatients with heart failure. The clinical pharmacist reviewed each intervention patient's chart and drug regimen and consuited with the physician and patient to optimize medication use at baseline and at 2, 12, and 24 weeks, either in person or via telephone. Clinical events were evaluated by a blinded physician. Patients in the intervention group were closer to target angiotensinconverting enzyme (ACE)-inhibitor doses (P < 0.01);

9

The American Journal of Geriatric Pharmacotherapy I J.T. Hanlon et al.

of those not receiving an ACE inhibitor, 75% were receiving alternative vasodilator therapy, compared with 26% of controls ( P - - 0.02). The intervention group experienced 4 heart failure events (3 deaths, 1 emergency department visit or hospitalization for heart failure), compared with 16 such events in the control group (5 deaths, 11 emergency department visits or hospitalization for heart failure) (P = 0.005). This study was limited by its use of a hybrid outcome measure (health services use and death) and its failure to measure adherence, HRQOL, or heart failure-specific physiologic measures. Nevertheless, the intervention improved drug prescribing and reduced mortality and hospital admissions in older patients with heart failure.

An Intervention in the Community Pharmacy Setting Bernsten et al4~ measured the outcomes of a structured pharmaceutical care program provided by community pharmacists to elderly patients taking _>4 medications. This study was performed in 7 European countries and involved 1290 intervention patients and 1164 control patients. The intervention pharmacist received 1 day of training, as well as a training manual. Follow-up was at 6, 12, and 18 months. The primary outcome was H R Q O L (measured using the SF-36), on which power calculations were based. It appears that hospitalizations and the associated costs were also a priori primary outcomes. Contact with general practitioners was another health outcome measured. Process measures included knowledge, compliance, number of prescription medications, nonprescription drug use, and changes in therapy. The mode of assessment of the outcome measures was patient self-report. It is not clear whether appropriate blinding was carried out. Overall, no differences in the primary outcomes were found. However, hospitalization data were available for only 4 of the 7 participating countries. There were some positive but statistically nonsignificant changes in certain process measures in the intervention group compared with the control group. There were a number of problems with this study. First, it is possible that the training of intervention pharmacists was insufficient, either before and/or during the study. Patients who were cognitively impaired were not screened out of the study (although including them would have been acceptable if they had a proxy). In addition, the dropout rate was high; data were collected in different ways across sites; and different time points were used (some sites employed baseline to 18 months, others 6-month intervals). Finally, cost analyses did not take

,0

I

into account outpatient laboratory and other procedures, emergency department visits, or nursing home admissions.

Interventions in Long-Term Care Facilities A study by Avorn et a142 examined the impact of a clinical pharmacist intervention on the use of psychotropic medications in elderly nursing home patients. The study included 823 elderly patients from 6 randomized matched pairs of nursing homes. The clinical pharmacist intervention, which was based on "academic detailing," consisted of educating physicians, nurses, and nurses' aides in the principles of geriatric psychopharmacology. A blinded research assistant evaluated patients' functioning. The intervention was associated with significant improvements in the use of psychotropic agents in patients from the intervention nursing homes compared with those from the control nursing homes; specifically, the intervention was associated with significant decreases in the use of antipsychotic drugs (32% vs 14%, respectively), hypnotics (45% vs 21%), and long half-life benzodiazepines (20% vs 9%) (all, P < 0.05). Overall, mean scores for inappropriate psychoactive drug use declined significantly in the intervention nursing homes compared with the control nursing homes (P = 0.02). Most measures of clinical status remained unchanged in both groups. This study was limited by its use of an unvalidated measure ofpsychotropic prescribing quality and its low power to detect differences in health outcomes given the short follow-up period (30 days). Nonetheless, the intervention was effective in improving suboptimal psychotropic prescribing for elderly nursing home patients. A more recent randomized controlled trial from Australia by Roberts et al43 examined the effects of a 12-month clinical pharmacy intervention on drug use, mortality, and morbidity in nursing home residents. The intervention, which included an educational session for nurses and drug regimen reviews, involved 52 nursing homes (13 intervention homes matched with 39 control homes) and 3230 residents. No information was provided about whether measurement of outcomes was blinded. Drug use was reduced by 14.8% in the intervention homes compared with the control homes (P = NS), with no change in morbidity indices (hospitalization rates, adverse events, or disability measures). There was a nonsignificant trend toward improved survival in the intervention group (adjusted hazard ratio, 0.85). The cost savings associated with the reduction in drug use was -$40 per resident per year.

J.T. Hanlon et al. I The American Journal of Geriatric Pharmacotherapy

This study was limited suboptimal prescribing errors. Nonetheless, the sary drug use in elderly

by its lack o f measurement o f or medication-administration intervention reduced unnecesnursing home patients.

DISCUSSION AND CONCLUSIONS There are some potential limitations to this review and to the studies included. Publication bias may have been a factor, as there may have been unpublished studies (usually those with negative findings) that could not be included in the review. In addition, due to the heterogeneity o f the interventions, their settings, and the outcome measures used, we were unable to perform quantitative synthesis o f existing data (ie, meta-analysis). Finally, a number o f studies measured some form o f global H R Q O L , several o f them using the SF-36. 3°,37,41 It is important to note that the SF-36 was designed as a discriminative health status index and may not be sufficiently responsive to longitudinal change. 44 Despite these potential limitations, the results o f numerous well-conducted studies in diverse care settings indicate that interventions by clinical pharmacists improved measures o f the quality o f drug use (DRP process measures) in elderly persons, including improvements in medication adherence and suboptimal prescribing. Less consistent was the effect o f clinical pharmacist interventions on important patient outcomes. Only i study each found an improvement in health status, s2 a reduction in hospital admissions or mortality, 4° and a lesser increase in drug costs. 39 Two studies reported a decrease in outpatient visits, 2s,86 and the latter study also reported fewer readmissions. The impact o f clinical pharmacy services may be better studied in patients with a specific problem (eg, congestive heart failure) rather than looking at patients' overall problems and medications. Future large-scale, multicenter studies arc necessary to determine the impact that clinical pharmacists with specialty training in geriatrics can have on the patient outcomes that are likely to be sensitive to such interventions (eg, health services use, timed functionalstatus measures, and ADRs). For example, a minimum o f 800 to 1400 elderly patients would need to bc randomizcd for detection o f a 25% reduction in ADRs due to a clinical pharmacy intervention. Even larger sample sizes would be necessary to detect differences in hospital admissions, as only up to 25% are drug related. 4 Future studies should also examine the effectiveness o f clinical pharmacist interventions in patients with specific conditions that are highly sensitive to medication optimization (eg, congestive heart failure) and allow

use o f disease-specific H R Q O L measures. Finally, the economic implications o f such pharmaceutical care programs should be studied using formal costeffectiveness or cost-benefit methods. Until then, this review provides some evidence for the benefit o f clinical pharmacist interventions among community-based elderly patients.

ACKNOWLEDGMENTS Financial support was provided from the Veterans o f Foreign Wars Endowed Chair in Pharmacotherapy for the Elderly, College o f Pharmacy, University o f Minnesota (Drs. Hanlon and Lindblad). Dr. Gray was supported by grant K 0 8 A G 0 0 8 0 8 - 0 1 from the National Institute on Aging.

REFERENCES 1. Strand LM, Morley PC, Cipolle RJ, et al. Drug-related problems: Their structure and function. Drug Intell Clin Pharm. 1990;24:1093-1097. 2. Hanlon JT, Lindblad CI, Maher R, Schmader KE. Geriatric pharmacotherapy. In: Tallis R, Fillit HM, eds. Brocklehurst's Textbook of Geriatric Medicine. 6th ed. London: Churchill Livingstone; 2003:1289-1296. 3. Kohn L, Corrigan J, Donaldson M, eds, for the Committee on Quality of Health Care in America, Institute of Medicine. To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press; 1999. 4. Hanlon JT, Schmader K, Gray S. Adverse drug reactions. In: Delafuente JC, Stewart RB, eds. Therapeutics in the Elderly. 3rd ed. Cincinnati, Ohio: Harvey Whitney Books; 2000:289-314. 5. Farris KB, Kirking DM. Assessing the quality of pharmaceutical care. II. Application of concepts of quality assessment from medical care. Ann Pharmacother. 1993;27: 215-223. 6. Lohr KN. Outcome measurement: Concepts and questions. Inquiry. 1988;25:37-50. 7. Branch LG. Community long-term care services: What works and what doesn't? Gerontologist. 2001;41:305-306. 8. Ernst FR, Grizzle AJ. Drug-related morbidity and mortality: Updating the cost-of-illness model. J Am Pharm Assoc (Wash). 2001;41:192-199. 9. Bootman JL, Harrison DL, Cox E. The health care cost of drug-related morbidity and mortality in nursing facilities. Arch Intern Med. 1997;157:2089-2096. 10. Adarncik BA, Rhodes RS. The pharmacist's role in rational drug therapy of the aged. Drugs Aging. 1993;3:481-486. 11. Dyer CC, Oles KS, Davis SW. The role of the pharmacist in a geriatric nursing home: A literature review. Drug Intell Clin Pharm. 1984;18:428-433.

11

The American Journal of Geriatric Pharmacotherapy I J"T. Hanlon et al.

12. Owens NJ, Silliman RA, Fretwell MD. The relationship between comprehensive functional assessment and optimal pharmacotherapy in the older patient. Drug Intell Clin Pharm. 1989;23:847-854. 13. Soumerai SB, Majumdar SR, Lipton HL. Evaluating and improving physician prescribing. In: Strom BL, ed. Pharmaeoepidemiology. 3rd ed. Toronto: John Wiley & Sons; 2000:483-503. 14. McDonald HP, Garg AX, Haynes RB. Interventions to enhance patient adherence to medication prescriptions: Scientific review. JAMA. 2002;288:2868-2879. 15. Kidder SW. Cost benefit of pharmacist-conducted medication reviews. Consult Pharm. 1987;2:394-398. 16. McGhan WF, Einarson TR, Sabers DL, Gardner ME. A meta-analysis of the impact of pharmacist drug regimen reviews in long term care facilities. J Geriatr Drug Ther. 1987;1:23-34. 17. Gurwitz JH, Soumerai SB, Avorn J. Improving medication prescribing and utilization in the nursing home. J A m Geriatr Soc. 1990;38:542-552. 18. Hanlon JT, Schmader KE, Ruby CM, Weinberger M. Suboptimal prescribing in older inpatients and outpatients. J A m Geriatr Soc. 2001;49:200-209. 19. Tett SE, Higgins GM, Armour CL. Impact of pharmacist interventions on medication management by the elderly: Review of the literature. Ann Pharmacother. 1993;27: 80-86. 20. Haynes RB, McDonald H, Garg AX, Montague P. Interventions for helping patients to follow prescriptions for medications (Cochrane Methodology Review). In: The Cochrane Library. Chichester, UK: John Wiley & Sons; Issue 4, 2003. 21. Singhal PK, Raisch DW, Gupchup GV. The impact of pharmaceutical services in community and ambulatory care settings: Evidence and recommendations for future research. Ann Pharmacother. 1999;33:1336-1355. 22. Beney J, Bero LA, Bond C. Expanding the roles of outpatient pharmacists: Effects on health services utilisation, costs, and patient outcomes (Cochrane Methodology Review). In: The Cochrane Library. Chichester, UK: John Wiley & Sons; Issue 4, 2003. 23. Lipton HL, Byrns PJ, Soumerai SB, Chrischilles EA. Pharmacists as agents of change for rational drug therapy. Int J Technol Assess Health Care. 1995;11:484-508. 24. Monane M, Matthias DM, Nagle BA, Kelly MA. Improving prescribing patterns for the elderly through an online drug utilization review intervention: A system linking the physician, pharmacist, and computer. JAMA. 1998;280:1249-1252. 25. Thompson JF, McGhan WF, Ruffalo RL, et al. Clinical pharmacists prescribing drug therapy in a geriatric set-

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

ring: Outcome of a trial. J A m Geriatr Soc. 1984;32: 154-159. Kimberlin CL, Berardo DH, Pendergast JF, McKenzie LC. Effects of an education program for community pharmacists on detecting drug~related problems in elderly patients. Med Care. 1993;31:451-468. Lowe CJ, Raynor DK, Purvis J, et al. Effects of a medicine review and education programme for older people in general practice. BrJ Clin Pharmacol. 2000;50:172-175. Sidel VW, Beizer JL, Lisi-Fazio D, et al. Controlled study of the impact of educational home visits by pharmacists to high-risk older patients. J Community Health. 1990; 15:163-174. Begley S, Livingstone C, Hodges N, Williamson V. Impact of domiciliary pharmacy visits on medication management in an elderly population. IntJPharm Pract. 1997;5:111-121. Krska J, Cromarty JA, Arris F, et al. Pharmacist-led medication review in patients over 65: A randomized, controlled trial in primary care. Age Ageing. 2001;30:205-211. Meredith S, Feldman P, Frey D, et al. Improving medication use in newly admitted home healthcare patients: A randomized controlled trial. J A m Geriatr Soc. 2002;50:1484-1491. Goodyer LI, Miskelly F, Milligan P. Does encouraging good compliance improve patients' clinical condition in heart failure? Br J Clin Pract. 1995;49:173-176. Lipton HL, Bero LA, Bird JA, McPhee SJ. The impact of clinical pharmacists' consultations on physicians' geriatric drug prescribing. A randomized controlled trial. Med Care. 1992;30:646-658. Lipton HL, Bird JA. The impact of clinical pharmacists' consultations on geriatric patients' compliance and medical care use: A randomized controlled trial. Gerontologist. 1994;34:307-315. Nazareth I, Burton A, Shulman S, et al. A pharmacy discharge plan for hospitalized elderly patients--a randomized controlled trial. Age Ageing. 2001;30:33--40. AI-Rashed SA, Wright DJ, Roebuck N, et al. The value of inpatient pharmaceutical counselling to elderly patients prior to discharge. Br J Clin Pharmacol. 2002; 54:657-664. Hanlon JT, Weinberger M, Samsa GP, et al. A randomized controlled trial of a clinical pharmacist intervention to improve inappropriate prescribing in elderly outpatients with polypharmacy. A m J Med. 1996;100:428437. Cowper P, Weinberger M, Hanlon JT, et al. The costeffectiveness of a clinical pharmacist intervention among elderly outpatients. Pharmacotherapy. 1998;18:327332.

J.T. Hanlon et al. I The American Journal of Geriatric Pharmacotherapy

39. Zermansky AG, Petty DR, Raynor DK, et al. Randomised controlled trial of clinical medication review by a pharmacist of elderly patients receiving repeat prescriptionsin general practice. BM~. 2001;323:1340-1343. 40. Gattis WA, Hasselblad V, Whellan DJ, O'Connor CM. Reduction in heart failure events by the addition of a clinical pharmacist to the heart failure management team: Results of the Pharmacist in Heart Failure Assessment Recommendation and Monitoring (PHARM) study. Arch Intern Med. 1999;159:1939-1945. 41. Bernsten C, Bjorkman I, Caramona M, et al. Improving the well-being of elderly patients via community pharmacybased provision of pharmaceutical care: A multicentre

study in seven European countries. Drugs Aging. 2001; 18:63-77. 42. Avorn J, Soumeral SB, Everitt DE, et al. A randomized trial of a program to reduce the use of psychoactive drugs in nursing homes. NEnglJMed. 1992;327:168-173. 43. Roberts MS, Stokes JA, King MA, et al. Outcomes of a randomized controlled trial of a clinical pharmacy intervention in 52 nursing homes. Br J Clin Pharmacol. 2001;51:257-265. 44. Stadnyk K, Calder J, Rockwood K. Testing the measurement properties of the Short Form-36 Health Survey in a flail elderly population. J Clin Epidemiol. 1998;51:827835.

Address c o r r e s p o n d e n c e to: Joseph T. Hanlon, PharmD, MS, Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Kaufman Medical Building, Suite 514, 3471 5th Avenue, Pittsburgh, PA 15213. E-mail: [email protected]

J

13