Background: Medical residents are at a high risk of depression. The National Institute for Occupational Safety and Health (NIOSH) job stress model explains that job stressors, buffer factors, and individual factors are related to worker health, including depression. This study investigated which job stressors and buffer factors influence depression in residents, independent of individual factors. Methods: A self-administered questionnaire was distributed to second-year residents at 251 teaching hospitals. The questionnaire contained the Brief Scales for Job Stress (BSJS), the Sense of Coherence (SOC) Scale, the Center for Epidemiologic Studies Depression (CES-D) Scale, and questions on working hours and demographic characteristics. The BSJS is composed of six subscales including workload, mental workload, problems in personal relationships, job control, reward from work, and support from colleagues and superiors. The SOC includes individual factors about life stressors and health. We analyzed the relationship between depressive symptoms and these factors. Results: Among 1,121 second-year residents who responded (64.5% response rate), 274 (24.4%) had depressive symptoms. In the multivariable analysis, depressive symptoms were significantly associated with SOC-13 score (p < 0.001), mental workload (p < 0.01), problems in personal relationships (p < 0.001), and reward from work (p < 0.01). Depressive symptoms were not significantly associated with mean number of hours worked per week. Conclusion: Mental workload and problems in personal relationships are job stressors and reward from work is a buffer factor associated with depressive symptoms in medical residents. These findings can help residency programs support mental health of residents more effectively.The effect of stress-related factors on depressive state among medical residents