Clinical Question

Which interventions improve medication adherence in patients with multiple medical conditions?

Evidence-Based Answer

Interventions that involve pharmacists to review, simplify, and monitor medication use significantly improve adherence in patients with multiple medical conditions. (Strength of Recommendation: B, based on heterogeneous randomized controlled trials [RCTs].) The effect of these interventions on patient-oriented outcomes is not clear. There are no data on physician- or nurse-based interventions for improving medication adherence.

Evidence Summary

A 2008 systematic review evaluating interventions to improve medication adherence found increased adherence in five of seven trials.¹ The authors could not perform a meta-analysis because of the heterogeneous interventions and outcome measures. The first trial was an RCT of a pharmacist-based intervention in seven European countries with 2,454 patients (mean age = 74 years; mean number of medications per patient = 7).² Patients were presumed to have multiple medical conditions because of their age and the number of medications. Investigators supplied pharmacists in the intervention group with an instruction manual, and provided five training sessions on how to perform medication reviews, simplify regimens, create reminder charts, and educate patients. The pharmacists in the control group were not trained. Investigators measured adherence using a patient self-report questionnaire. At 18 months, more patients in the intervention group adhered to their prescribed medication regimen than in the control group (15.2 versus 12.2 percent; P = .03). There was no effect on quality-of-life scores or total health care costs. Forty-two percent of patients entering the trial were lost to follow-up.

An RCT from the United States evaluated a pharmacist-based intervention involving 200 patients with hypertension and hyperlipidemia (mean age = 78 years; mean number of medications per patient = 9).³ Pharmacists provided patient education, performed follow-up to assess and support medication adherence, and dispensed medication in blister packs. After six months of pharmacist intervention, the remaining patients were randomized to continued intervention (n = 83) or usual care (n = 76) for another six months. Investigators measured adherence using pill counts, and tracked blood pressure and low-density lipoprotein cholesterol levels. Adherence rates increased from 61.2 percent at baseline to 96.9 percent at six months (P < .001); however, at 12 months the intervention group continued to have high adherence, whereas adherence in the usual care group declined almost to baseline (95.5 versus 69.1 percent; P < .001). Systolic blood pressure was lower in the intervention group than in the control group at 12 months (124 versus 133 mm Hg; P = .001), although the diastolic blood pressure and low-density lipoprotein cholesterol levels were comparable.

An RCT from Hong Kong evaluating patient adherence randomized 442 patients (mean age = 71 years; mean number of medications per patient = 6) to receive usual care or a pharmacist-based intervention that included medication consultations and reminder telephone calls.⁴ Investigators documented multiple medical conditions among the patients. Pharmacists made 10- to 15-minute reminder calls to patients at the midpoint between scheduled physician visits. At the end of the study, more patients in the intervention group reported adherence than in the usual care group (81 versus 58 percent; P = .001). Additionally, mortality rates were lower in the intervention group than in the control group (11 versus 17 percent; relative risk of death = 0.59; 95% confidence interval, 0.35 to 0.97; number needed to treat = 16 after adjustment for potential confounders). However, more patients in the intervention group were taking lipid-lowering and antiplatelet medications, and cardiovascular events were the major cause of death in both groups.

A study from India randomized 90 patients in a 1:2 ratio to receive pharmacist counseling or usual care.⁵ Most patients were between 40 and 60 years of age, and approximately 90 percent had more than one medical condition. Pharmacists at the patients' own clinics provided a single counseling session. One month later, patient adherence (as measured by pill count) was better in the intervention group than in the usual care group (92 versus 85 percent; P value not given).

Two RCTs in the 2008 systematic review found no improvement in treatment adherence with pharmacist intervention.¹ The first was a six-month British study of 362 patients that compared usual care with pharmacist counseling in which pharmacists developed and revised medication plans with patients during follow-up visits. The second was a six-month U.S. study of 180 patients with diabetes mellitus and hyperlipidemia that compared usual care with pharmacist assessment of medication use by telephone.

Recommendations from Others

The Royal College of General Practitioners recommends that physicians involve patients in decisions about their medications; understand patients' knowledge, beliefs, and concerns about medications; provide medication-specific information; and assess adherence.⁶ To improve treatment adherence in older patients, the University of Iowa Gerontological Nursing Interventions Research Center recommends using memory aids such as pill boxes and automatic dispensers, performing frequent physician review to reduce polypharmacy, and monitoring for changes in functional status that might impact adherence.⁷