Impact of Physicians' Characteristics On the Risk of Admission Among Children Visiting a Pediatric Emergency Department

Abstract

Introduction: Children seeking care in pediatric emergency departments (ED) can be treated by physicians trained in various specialties; their characteristics may impact patient management. Objective: To assess the effects of physicians' training, work experience and gender on hospitalization rates among children visiting a pediatric ED. Methods: This retrospective cohort study used the computerized database of a tertiary care pediatric ED staffed by pediatric emergency physicians, pediatricians and general emergency physicians. Participants were all children aged less than 19 years old who were seen by a physician in the ED between April 1st 2008 and March 31st 2009. The physician's gender, years of work experience and specialty training were assessed as predictors of admission. The primary outcome was hospitalization. The secondary outcome was the proportion of unscheduled return visits to the ED within 48 hours of discharge. Multivariate logistic regression was used to evaluate associations between physicians' characteristics and the risk of admission, adjusting for triage level and referral status. Sample size calculation was based on the fact that 10 physicians are needed for each risk factor. We estimated that studying all patients visiting the ED for one year would provide 50,000 patient visits and 5,000 admissions. Results: During the study, 49,500 patients were evaluated by 45 physicians; each physician treated at least 20 patients. Of the patients, 6,087 (12.3%) were admitted and 3,040 (5.6%) returned to the ED within 48 hours of discharge. Physicians' individual admission and return rates were 0%-29% and 0%-11%, respectively. On multiple logistic regression, the physician's years of work experience was associated with admission rates (OR for 5 years work: 1.20 (95% CI 1.10, 1.29); OR for 5-10 years work: 0.82 (95% CI 0.76, 0.90); reference >10 years work) and with unscheduled return visits (OR 1.16 95% CI 1.04, 1.30 for <5 years and 5-10 years worked, as compared to >10 years work). Physicians' specialty and gender were not determinants of hospitalization. Conclusion: This study demonstrates that, in addition to patient characteristics, a physician's work experience has an impact on admission rates, independent of specialty training.