Burnout Phenotypes Among U.S. General Surgery Residents

OBJECTIVE: Although well-established metrics exist to measure workplace burnout, researchers disagree about how to categorize individuals based on assessed symptoms. Using a person-centered approach, this study identifies classes of burnout symptomatology in a large sample of general surgery residents in the United States. DESIGN, SETTING, PARTICIPANTS: A survey was administered following the 2018 American Board of Surgery In-Training Examination (ABSITE) to study wellness among U.S. general surgery residents. Latent class models identified distinct classes of residents based on their responses to the emotional exhaustion and depersonalization questions of the modified abbreviated Maslach Burnout Inventory (aMBI). Classes were assigned representative names, and the characteristics of their members and residency programs were compared. RESULTS: The survey was completed by 7415 surgery residents from 263 residency programs nationwide (99.3% response rate). Five burnout classes were found: Burned Out (unfavorable score on all six items, 9.8% of total), Fully Engaged (favorable score on all six items, 23.1%), Fatigued (favorable on all items except frequent fatigue, 32.2%), Overextended (frequent fatigue and burnout from work, 16.7%), and Disengaged (weekly symptoms of fatigue and callousness, 18.1%). Within the more symptomatic classes (Burned Out, Overextended, and Disengaged), men manifested more depersonalization symptoms, whereas women reported more emotional exhaustion symptoms. Burned Out residents were characterized by reports of mistreatment (abuse, sexual harassment, and gender-, racial-, or pregnancy and/or childcare-based discrimination), duty hour violations, dissatisfaction with duty hour regulations or time for rest, and low ABSITE scores. CONCLUSIONS: Burnout is multifaceted, with complex and variable presentations. Latent class modeling categorizes general surgery residents based on their burnout symptomatology. Organizations should tailor their efforts to address the unique manifestations of each class as well as shared drivers.