Assessing the impact of multi-compartment compliance aids on clinical outcomes in the elderly: a pilot study

Background Medication non-adherence is a major problem for elderly people. Multicompartment compliance aids (MCAs) have been advocated as a solution for this problem. Objective To assess the impact of using MCAs in self-reported adherence and clinical biomarkers of elderly patients followed in a community pharmacy. Setting One community pharmacy at Sabugal (Portugal). Methods A four-month prospective, non-randomised, controlled study was performed. Autonomous patients aged 65 or more using 3 or more medicines and under follow-up in the pharmacy were invited to participate. All patients were offered to receive their medication in MCAs prepared in the pharmacy. Patients refusing the MCA were used as control. The intervention consisted of providing 4 weekly MCAs during the monthly visit. All patients received regular pharmacy counselling. Blood pressure (BP), lipid profile and glycaemia were assessed at baseline and monthly for all the patients. Morisky self-reported scale was applied at baseline and at the end of the study. Bivariate analysis and generalized estimation equations (GEE) were used. Main Outcome Measure: Self-reported medication adherence, clinical biomarkers: BP, lipid profile, glycaemia. Results 54 patients between 65 and 90 years were under follow-up. 44 patients accepted the MCA, constituting the intervention group. No difference in the baseline biomarkers between both groups was found. The bivariate pre-post analysis yielded significant improvements in the intervention groups, but not in the control, for glycaemia (p < 0.001), HDL-c (p = 0.018), and systolic (p < 0.001) and diastolic (p = 0.012) BP. However, when introducing the ‘time in follow-up’ in the GEE model, all the differences became non-significant, except systolic BP, but the time remained significant for all the biomarkers. Conclusion MCAs apparently improve several clinical biomarkers in a cohort of patients under pharmacist’s follow-up. When including the time in pharmacist’s followup in a GEE, the effect of the MCA disappeared, remaining only the time as a significant variable. Not considering the time in follow-up may be overestimating the effect of MCAs.