Why Residents Quit

National Rates of and Reasons for Attrition Among Emergency Medicine Physicians in Training

Dave W. Lu, MD, MSCI, MBE; Nicholas D. Hartman, MD, MPH; Jeffrey Druck, MD; Jennifer Mitzman, MD; Tania D. Strout, PhD, RN, MS


Western J Emerg Med. 2019;20(2):351-356. 

In This Article


There were a total of 51,882 unique EM residents in the AMA National GME Census database during this 10-year period. The annual number of active EM residents enrolled in an ACGME-accredited program ranged from a low of 4,389 in 2006–2007 to 5,865 in 2015–2016. When comparing overall rates of attrition between EM and the other top 10 largest specialties, EM had the lowest rate of attrition (0.8%, 95% CI [0.7–0.9]), or approximately 51.6 (95% CI [44.7-58.5]) residents per year (Table 1). The majority of EM residents graduated from allopathic medical schools (82.4%, 95% CI [81.4–83.4]), followed by those from osteopathic (11.2%, 95% CI [10.5–11.9]) and international (6.4%, 95% CI [6.0–6.8]) medical schools. There were no significant differences in attrition by type of medical school graduate (χ 2=7.150, df=2, p=0.028).

From 2006 to 2016, women comprised 38.8% (95% CI [37.9–39.7]) of EM residents, with no significant changes in gender composition noted during the study period (F=0.607, p=0.436). In the attrition population, 44.3% (95% CI [40.0–48.4]) of the residents were women, a significantly higher proportion when compared to the proportion of female EM residents overall (z=-2.544, p=0.011).

When examining attrition status, almost half of the residents who left their programs "withdrew" (47.0%, 95% CI [42.8–51.4]) (Table 2). There were no differences in attrition status by gender except for those who were "dismissed," with a significantly greater proportion of men receiving this status than women (16.0% vs 8.3%; χ 2=9.852, df=4, p=0.043). When examining the primary listed reason for attrition, the majority reported a "change in career plans" (57.4%, 95% CI [50.9–63.3]) (Table 3). A significantly greater proportion of women than men reported "health/family" reasons for attrition (21.5% vs 9.6%; χ 2=9.923, df=3, p=0.019). All other queried reasons for attrition were similar between women and men.

Race/ethnicity responses to the AMA National GME Census were reported alone or in combination with any other race/ethnicity response. "Alone" indicated those who selected only one race/ethnicity response, whereas "in combination" indicated those who selected more than one race/ethnicity response. An individual could therefore be represented in more than one race/ethnicity category if that individual reported more than one race/ethnicity response. As such, there were 52,490 subjects in this analysis with 1.2% of the subjects reporting more than one racial/ethnic category. Whites comprised the largest group of EM residents (71.3%), followed by Asians (13.0%), Hispanics/Latinos (6.3%), Blacks/African Americans (5.0%), other (3.3%), American Indians/Alaskan Natives (0.8%), Native Hawaiians/other Pacific Islanders (0.2%), and unknown (0.1%). When comparing attrition across race/ethnicity categories, White (0.9%) residents experienced significantly less attrition than their Hispanic/Latino (1.8%) counterparts (χ 2=32.243, df=7, p<0.001) (Table 4). In addition, the proportion of Hispanic/Latino residents who were "dismissed" from their programs (39.3%), was significantly greater than Asian (10.5%) and White (7.5%) residents experiencing dismissal (χ 2=67.516, df=24, p<0.001) (Table 5).