Error and adverse outcomes on a surgical service: what is the role of the resident?

Abstract

Purpose: The changeover in residents that occurs each July in university-affiliated hospitals has been postulated to result in impaired delivery of patient care (the “July phenomenon”). To assess the impact of this process on patient safety, we examined the incidence and sources of medical error and adverse outcomes on a pediatric surgery service during the final month of an academic year and the first month of the subsequent academic year. Methods: All admissions to two pediatric surgeons during June and July 2002 were prospectively followed. In-patient care was reviewed daily by the attending surgeon, a pediatric surgery fellow and a medical student. Errors committed by doctors, nurses, and allied health workers were identified through daily patient encounters, nursing rounds, medical rounds, and chart audit. Adverse outcomes were evaluated based on type and contributing factors, including involvement of residents. To correct for variations in length of stay, incidence of errors and adverse outcomes were expressed as a percentage of total patient days. Results: The error rate was 46/643 patient days (7.1%) in June, and 58/776 patient days (7.5%) in July (p = 0.9). Total adverse outcome rate was also not significantly different between the two months (5% vs 6.7%, p = 0.21). Resident error accounted for 52.2% of errors in June and 39.7% of errors in July (p = 0.28), and the likelihood of a resident error resulting in an adverse outcome was also similar between months (10.8% vs 22.4%, p = 0.2). The incidence of resident error was not influenced by whether the resident was on-call, post-call or neither. Conclusion: Resident changeover at the completion of an academic year was not associated with an increased number of medical errors or adverse outcomes, indicating that effective systems are in place to prevent the “July phenomenon”. The on-call status of the resident did not appear to affect the likelihood of resident error.