Discrepancies in prescribed medications as reported by patients, general practitioners, and community pharmacists in older patients with polypharmacy in primary care.

Abstract

BACKGROUND: Studies have shown considerable discrepancies between the medication lists reported by patients, physicians, and pharmacists. OBJECTIVE: To investigate discrepancies in prescribed medications as reported by patients, general practitioners (GPs), and community pharmacists (CPs) in older patients with polypharmacy. METHODS: Study design: Cross-sectional analysis. Setting: Twenty-four GP centres and 17 associated community pharmacies in the Netherlands. Data sources: Postal questionnaires, home visits, and the electronic patient record systems of the GPs and CPs. We defined a medication discrepancy as any inconsistency between the patient’s, GP’s, or CP’s medication list. For fixed medications, we included inconsistencies in the medication’s name, dose, or frequency; for variably used medications, we included only inconsistencies in the medication’s name. Logistic regression analyses and GEE were performed to identify discrepancy-related patient and medication characteristics. RESULTS: Data from 751 patients (mean age 72.9 years (SD 7.6), 47.3% female) using 5989 prescribed medications (average 8, range 3–19) were analysed. The overall patient-GP-CP discrepancy was 71.9% at the patient level and 21.8% at the medication level. Statistically significant associations with discrepancy included various medication characteristics, e.g. any additional medication (OR 1.45, 95% CI 1.33–1.59) and variable medication usage (varying ORs 1.61–5.78), and the patient characteristic ‘Living in a residential care home’ (OR 0.42, 95% CI 0.18–0.997). CONCLUSION: We found a high patient-GP-CP discrepancy, with a 72% discrepancy in prescribed medication at the patient level and a 22% discrepancy at the medication level. ‘Living in a residential care home’ reduced the chance of medication discrepancy, whereas medication characteristics, such as variable medication usage, increased it. TRIAL REGISTRATION: NLOMON32535 (https://onderzoekmetmensen.nl/en/trial/32535 registered 15th December 2009).