[Prescription appropriateness in elderly patients with polypharmacy in primary care: Cluster-randomized controlled trial PHARM-PC].

OBJECTIVES: To assess the effectiveness of a pharmacist-led systematic review of medications on: potentially inappropriate medications (PIM), health outcomes and costs.

DESIGN: Prospective, open, controlled and cluster-randomized clinical trial.

SETTING: Six primary care clinics from Balearic Islands.

PARTICIPANTS: Forty-two clusters (21 per group), and 549 patients aged ≥65 years and ≥5 chronic medications were included; of which 277 were allocated to Intervention Group (IG) and 272 to Control Group (CG). Patients were excluded if they were: institutionalized, temporarily displaced, routinely monitored under private healthcare, or home care.

INTERVENTION: PIM detection by the pharmacist using a combination of explicit and implicit methods; and communication of the most appropriate therapeutic strategies to the physician.

MEASUREMENTS: Proportion of patients with PIM and mean number of PIM/patient (main outcomes); and morbidity, mortality, and costs (secondary outcomes) were assessed.

STATISTICAL PLAN: Following an intention-to-treat approach, quantitative and qualitative outcomes variables were compared by T-Student and Chi-square tests, respectively. Results were providing as difference in proportions for qualitative outcomes and difference in means for quantitative outcomes with respective 95% confidence intervals (95% CI).

RESULTS: After intervention, proportion of patients with PIM decreased by 13.7% (95% CI: 9.3; 18.2) more in IG than CG. Mean number of PIM/patient and mean cost of PIM/patient decreased by 0.43 (95% CI: 0.32; 0.54) and 72.11€ (95% CI: 26.15; 118.06) more in IG than CG, respectively. However, no statistically significant differences were observed in morbidity, mortality or costs of healthcare resources.

CONCLUSIONS: PIM detection and recommendations provided by pharmacist could contribute to reduce significantly PIM and drug expenditure; but without reaching statistically significant differences in morbidity, mortality, and healthcare resources costs.