Potential drug-drug interactions in a Canadian tertiary-care hospital

Abstract

Drug-Drug interactions (DDIs) are an important focus of patient safety because they account for a substantial number of adverse drug events and are preventable. Objective. To study DDIs in a Canadian hospital, a retrospective observational study was completed using the Ottawa Hospital Data Warehouse. Study cohort. Admissions to the Ottawa hospital between January 1, 1999 and September 30, 2005. Methods. Potential drug-drug interactions were identified by examining all co-administered medications for combinations of drugs previously reported to have potential interactions. Poisson regression was used to examine potential patient and hospital factors associated with drug-drug interactions. Results. Between 1999 and 2005, we found at least one DDI in 19.3% of all hospitalizations and 18.8% of hospitalization time. Category 1 (drug combinations to be avoided) and Category 2 (drug combinations usually avoided) interactions were rare, accounting for only 0.022% and 1.4% of hospitalization time, respectively. Category 3 interactions (drug combinations requiring alteration) occurred with 5.7% of all drug orders and were present for 17.4% of hospitalization time. Poisson regression analysis found that DDIs were significantly more likely to occur in patients who were: older; admitted to a surgical service; had a greater number of comorbidities; and were prescribed a greater number of drugs. Conclusion. Drug-Drug interactions occurred frequently during hospitalization. Future study is required to determine if the interactions identified are associated with important clinical outcomes.