Impact of pharmaceutical care on adherence, hospitalisations and mortality in elderly patients.

Background: Elderly polypharmacy patients may be more at risk of not adhering to medication. If so, the underlying reasons may be more readily disclosed during private discussions with patients. Hence pharmaceutical care discussions at home might improve treatment adherence.

Objective: The aim of this study was to investigate the impact of pharmaceutical care on medication adherence, hospitalisation and mortality in elderly patients prescribed polypharmacy.

Setting: Pharmaceutical care discussed at home.

Methods: A randomised controlled trial with two arms; pharmaceutical care (n = 315) and controls (n = 315) was designed. It involved patients aged 65+ years living in Aarhus, Denmark who used five drugs or more without assistance. Pharmacists visited the pharmaceuticalcare patients at home, once only, and followed them during the subsequent year with three telephone calls. Non-adherence was measured by a pill-count. Patients were categorised as non-adherent if their mean adherence rate for all drugs consumed was <80 %. The impact of pharmaceutical care on non-adherence and hospitalisation was analysed by 2 × 2 tables, and mortality by Cox regression.

Main Outcome Measure: Medication adherence, hospitalisation and mortality.

Results: The final analyses included 517 patients (median age 74 years; females 52 %). Dropouts were more frequent for the pharmaceutical-care group than for controls. Pharmacists encountered drug-related problems amongst 72 % of pharmaceutical-care patients. Pharmaceutical-care patients (11 %) and control patients (10 %) were similarly nonadherent (Odds ratio 1.14; 95 % confidence interval 0.65-2.00), and similar with respect to hospitalisation frequency (30 vs. 28 %; Odds ratio 1.14; 95 % confidence interval 0.78-1.67) and mortality (7.5 vs. 5 %; Hazard ratio 1.41; 95 % confidence interval 0.71-2.82).

Conclusions: Pharmaceutical care given to our elderly polypharmacy patients made no significant impact on medication adherence, hospitalisation or mortality, when compared to comparable control patients.