Incidence of adverse drug events in an academic hospital: a prospective cohort study.

OBJECTIVE To determine the incidence of adverse drug events (ADEs) and assess their severity and preventability. DESIGN A prospective cohort study. SETTING A 900-bed tertiary academic hospital. PARTICIPANTS A total of 977 patients admitted to two medical, one surgical and two intensive care units over four months. MAIN OUTCOME MEASURES The primary outcomes were the incidence of ADEs, preventability of ADEs, potential ADEs and medication errors. A physician and a clinical pharmacist independently determined the likelihood that incidents were caused by medications and judged severity and preventability. RESULTS Pharmacists reviewed the medical records of the 977 patients. Pharmacists identified 361 incidents, of which 281 (78%) were considered to be an ADE, potential ADE or medication error by reviewers. The incidence of ADEs was 8.5 per 100 admissions (95% confidence interval (CI) 6.8-10.4), with the highest rate found in the intensive care unit (21.1 per 100 admissions) (95% CI 15.1-28.8). Of all ADEs, 59% were rated as significant, 35% as serious and 6% as life threatening. Thirty percent of ADEs were preventable and 96% of these occurred in the ordering stage. The incidence of potential ADEs was 13.8 per 100 admissions (95% CI 11.5-16.2). Overall, 223 medication errors were identified, 66 (30%) were harmless, 132 (59%) had the potential to cause harm and 25 (11%) resulted in harm. CONCLUSIONS The incidence of ADEs in a Saudi Hospital was 8.5 per 100 admissions. Preventable ADEs most commonly occurred in the ordering stage; therefore, interventions to reduce ADEs should target the ordering stage.