Depressive symptoms among medical students: comparing three measures in nine countries

OBJECTIVE: Prevalence estimates of depression among medical students are heterogeneous. To investigate this, we evaluated the cross-cultural measurement invariance and case-finding concordance of three depression screening tools across nine countries. This enabled standardized prevalence comparisons and the identification of sociodemographic factors associated with depressive symptoms across multiple countries. METHODS: A cross-sectional online survey assessed medical students from ten institutions across nine countries. Measurement invariance across countries was assessed for each instrument. We estimated pooled apparent prevalence via random-effects meta-analyses, evaluated diagnostic concordance between instruments, and identified associated sociodemographic factors using logistic regression. RESULTS: The sample included 2,733 students (mean age 22.3; 60.7% female). The Patient Health Questionnaire-9 (PHQ-9) and Depression, Anxiety and Stress Scale-Depression subscale (DASS-D) achieved strict measurement invariance, whereas the Center for Epidemiologic Studies Depression Scale (CES-D) was structurally unstable. Pooled prevalence varied markedly by tool: 36.1% (95% CI: 26.9-46.4%, I²=96.1%) for PHQ-9, 43.8% (95% CI: 36.6-51.4%, I²=93.0%) for DASS-D, and 56.4% (95%CI: 49.2-63.3%, I²=92.2%) for CES-D. Only 40.1% of positive screens were concurrently identified by all three instruments. Female sex and lower subjective social status were consistently associated with higher odds of depressive symptoms. CONCLUSIONS: Depression prevalence estimates are highly tool-dependent. The PHQ-9 and DASS-D offer robust frameworks for global comparisons, while the CES-D should be used cautiously. The cross-cultural consistency of female sex and lower social status as risk factors highlights the need for targeted institutional mental health strategies.