Nomogram for predicting major depression and suicidal ideation in medical residents: A multi-center cross-sectional study

Abstract Background: Few studies have addressed major depression and suicidal ideation in medical residents, yet the high incidence rate and low attendance rate highlight suggest the need for a greater focus on mental health. To our knowledge, there is no model for predicting major depression and suicidal ideation in medical residents. Thus, we developed and validated a model for predicting major depression and suicidal ideation in this specific subpopulation of medical professionals.Design: The development cohort included 938 medical residents from six centers between January 1, 2017 and December 31, 2018. A total of 405 consecutive medical residents from two other centers met the inclusion and exclusion criteria and participated in the validation cohort. Lasso regression was utilized for data dimension reduction and feature selection. Multivariable logistic regression was then used to develop a predictive model. The efficacy of this predictive model was assessed with respect to its clinical usefulness, calibration, and discrimination.Results: We identified four shared predictive factors for both major depression and suicidal ideation: sleep quality, Masrah Burnout Inventory personal accomplishment, Masrah Burnout Inventory depersonalization, and optimism of Psychological Capital. Furthermore, we identified two predictors that affect major depression exclusively: working duration and Masrah Burnout Inventory emotional exhaustion. Of the study participants, 44.90% (603/1343) experienced depressive symptoms, 12.90% (173/1343) experienced major depressive symptoms, and 9.70% (130/1343) experienced suicidal ideation. In the validation cohort, our model showed good discrimination, with an AUROC (Area under the receiver operating characteristic curves) of 0.906 (95% CI: 0.872–0.940) and good calibration (unreliability test, P = 0.836). Decision curve analysis showed that the model was clinically useful.Conclusion: This study provides a reliable nomogram to facilitate the individualized prediction of major depression and suicidal ideation among medical residents, allowing for the early diagnosis and treatment of mental disorders in this specific subpopulation of medical professionals. The findings of our study are conducive to expanding the knowledge of mental disorders and improving the development of public health.