OBJECTIVE: To investigate the prevalence of concomitant use and coprescribing of cisapride with potentially interacting drugs to evaluate the impact of these warnings from 1994 to 1998. DESIGN: Retrospective follow-up study of patients using cisaprde. SETTING: Data for this study were obtained from the pharmacy database of the Dutch PHARMO record linkage system (n = 834,000). RESULTS: From 1994 to 1998, the prevalence rate of the observed versus expected use of any potentially interacting drug decreased significantly over time (p < 0.01). However, the number of days-at-risk and number of coprescriptions of potentially interacting drugs among cisapride users increased on average by 13% and 9% per year, respectively. This increase was almost exclusively explained by a large increase in concomitant prescribing of clarithromycin, the most commonly used potentially interacting drug. Decreases in prevalence rates were observed for all individual potentially interacting drugs, except for concomitant use of fluconazole and miconazole. CONCLUSIONS: Over the last few years, healthcare professionals have refrained from dispensing potentially interacting drugs to patients who use cisapride, probably as a result of drug warnings implemented during this period. The limited absolute effects result from an increase of coprescription and concomitant use of clarithromycin and fluconazole among cisapride users. Because therapeutically equivalent alternatives were available for both drugs, such combinations wereavoidable. Communicating information on these drug-drug interactions to prescribers and pharmacists and inclusion of cardiovascular morbidity as a relative contraindication for prescribing cisapride with these drugs may substantially decrease the risk ofpotentially adverse events to cisapride.