Resident attrition negatively affects residents and programs. The incidence of attrition in emergency medicine (EM) and program-specific factors associated with attrition remain unclear. We quantified the percentage of EM residencies affected by attrition between 2007 and 2016 and identified program-specific factors associated with attrition. We performed a retrospective analysis of data derived from the American Medical Association National Graduate Medical Education Census. We defined attrition as any postgraduate who left their residency training program prior to completion. We calculated the percentage of residency programs that experienced attrition and the overall incidence of attrition. We used Fisher's exact tests, Wilcoxon rank sum tests, and t tests, as well as multivariable logistic regression, to identify program-specific factors associated with attrition. Between 2007 and 2016, 139 EM residency programs (82%) experienced attrition of at least 1 resident. An average of 23% of EM training programs experienced attrition annually. The incidence of EM resident attrition averaged 0.85% per year. Program-specific factors associated with attrition include 4-year residencies (P = .031), programs with medium class size (P = .0003), more female residents (P = .002), and more female faculty (P = .003). After analysis, only medium class size (compared to small) was associated with attrition (odds ratio = 4.96, 95% confidence interval 1.65-14.91). Between 2007 and 2016, while the incidence of resident attrition in EM was low (< 1%), the majority of programs experienced resident attrition. Medium class size (7 to 12 residents) was the only program-specific factor associated with increased attrition.