Abstract

Depression and anxiety are prevalent among medical students, impacting their academic performance and well-being. This study aims to predict mental health levels, focusing on depression and anxiety, among medical students using demographic variables such as age, sex, year of study, and health status. A cross-sectional dataset of 886 medical students was analyzed. Stepwise regression with backward selection was employed to identify significant predictors of depression, as measured by the CES-D scale. Independent variables included demographic and health-related factors. Multicollinearity was assessed using the Variance Inflation Factor (VIF), and model diagnostics were conducted using the Omnibus test, Durbin-Watson statistic, and Jarque-Bera test. The stepwise regression model explained 81.6% of the variance in depression scores (R-squared+0.816, adjusted R-squared+0.803). Anxiety (STAI_T), Emotional Exhaustion (MBI_EX), and health satisfaction emerged as significant predictors, with anxiety showing the strongest correlation (r+0.73, p<0.001). The model demonstrated a good fit, as indicated by low AIC (6007) and BIC (6050) values. Regression diagnostics showed slight deviations from normality in residuals but no significant multicollinearity or autocorrelation. Anxiety, emotional exhaustion, and health satisfaction are key predictors of depression among medical students. The findings emphasize the need for targeted mental health interventions in medical education. Addressing limitations such as residual normality can further improve predictive accuracy.

Keywords:Mental health; Medical students; Depression; Anxiety; Predictive modelling