Partial replacement of board-certified specialist-grade physicians with emergency medicine trainees in a busy emergency department: Lack of adverse effect on time to physician

Objectives: Standard emergency department (ED) operation goals include minimization of the time interval between patients' initial ED presentation and initial emergency physician (EP) evaluation. Following up on previous work defining factors influencing the “time to physician” ( tMD ) in a busy ED, the current study was undertaken to evaluate whether tMD was adversely impacted by the ED's partial replacement of specialist-grade EPs with emergency medicine (EM) trainees (at the resident and fellow level). Methods: This retrospective study was conducted for four months (September–December 2015) using an ED administrative database (EDAD) in an urban academic tertiary ED with an annual census of approximately 500,000; during the four study months, the ED census was 165,969. To minimize confounding by time of day and related factors, data analysis focused solely on the “day shift” (0600–1400) of each of the study period's 122 days. EDAD data were combined with EP rostering data to generate a multivariate linear regression model that assessed the dependent variable tMD , for significant changes associated with increasing proportion – not necessarily always the same as increasing the absolute number of trainees (i.e., summed residents and fellows as a total percent of all on-duty EPs). There were trainees in the study ED throughout the study, but the trainee numbers as a proportion of the overall physician staffing fluctuated, thus providing a basis for analysis. The model adjusted for covariates previously demonstrated to impact tMD at the study center. Analyses were conducted with Stata 14MP, with statistical significance defined at p  < 0.05 and confidence intervals (CIs) reported at the 95% level. Results: In an acceptable regression model that adjusted for multiple parameters influencing tMD , the introduction of a covariate representing the proportion of on-duty trainee physicians was very small in magnitude (β estimate 0.07, 95% CI − 0.16 to 0.30) and not statistically significant ( p  = 0.53). Conclusions: A multivariate analysis adjusting for variables contributing to tMD showed no indication of adverse tMD impact from partial replacement of board-certified specialist-grade EPs with EM trainees given adequate supervision by properly trained faculty.