Compliance with national guidelines for the management of drug-drug interactions in dutch community pharmacies

BACKGROUND: Pharmacists contribute to the detection and prevention of drug therapy-related problems, including drug-drug interactions. Little is known about compliance with pharmacy practice guidelines for the management of drug-drug interaction alerts. OBJECTIVE: To measure the compliance of community pharmacists with Dutch guidelines for the management of drug-drug interactions and to determine patient- and prescriber-related determinants for noncompliance. METHODS: Sixteen clinically relevant drug-drug interactions were included in the study based on certain described criteria. From June to August 2005, Dutch pharmacists (N = 149) collected alerts occurring in daily patient care for these interactions as well as information related to the patient, the alert itself, the prescriber, and the management of the alert. Noncompliance was measured by comparing the management executed by the pharmacy with the national guidelines. RESULTS: Overall compliance with the guidelines was 69.3% (n = 423), with large differences between the various drug-drug interactions. Male sex (OR 2.25; 95% CI 1.52 to 3.31), oldest age (> 75 y; ORA.97; 95% CI 1.03 to 3.75), and polypharmacy (> 7 medications; OR 2.35; 95% CI 1.46 to 3.80) were associated with a higher probability for noncompliance with the guidelines. Prescriber-related variables had no significant influence on guideline compliance. Substitution of one of the involved agents, recommended for most of the drug-drug interactions, was executed in a small minority of cases. The outcome of interaction management, such as substitution, dose reduction, or temporary stop of one of the agents, was frequently inconsistent with the guidelines. Compliance rates were partly influenced by the ultimate decision made by the prescriber. In that way, pharmacies' compliance was not solely assessed. However, in only 22.5% of the cases was the drug-drug interaction presented to the prescriber. CONCLUSIONS: Noncompliance with Dutch guidelines for the management of drug-drug interaction alerts is common in community pharmacies. Further research into underlying reasons for noncompliance is warranted, such as the relation between pharmacist and prescriber in this context.