Exploring Hazing Victimization and Its Effects on Mental Health in Medical Schools Using Latent Class Analysis and Structural Equation Modeling

Abstract Hazing, a ritualized, coercive group practice distinct from general mistreatment, remains prevalent in higher education, but its mental-health impact in medical training is understudied. We aimed to quantify the prevalence and patterns of hazing exposure in Brazilian medical students and test whether associations with depression and anxiety are mediated by academic stress and the learning environment. We conducted a cross-sectional online survey of N = 1,024 students from 74 medical schools. Measures included the UNI-Hazing victimization subscale, Johns Hopkins Learning Environment Scale (JHLES), Medical School Stressor Factors Scale (MSSF), Patient Health Questionnaire-9 (PHQ-9), and Generalized Anxiety Disorder-7 (GAD-7). Latent class analysis (LCA) characterized exposure patterns; structural equation modeling (SEM) tested direct and indirect paths. Three exposure groups emerged: Low Hazing (69.0%), Moderate Hazing (24.5%), and High Hazing (6.5%). Relative to the Low group, the High Hazing group reported higher stress (MSSF: 144.2 ± 22.4 vs 127.3 ± 29.2), poorer learning-environment perceptions (JHLES: 76.5 ± 14.4 vs 84.1 ± 12.5), and greater depressive (PHQ-9: 15.8 ± 6.6 vs 12.5 ± 6.8) and anxiety symptoms (GAD-7: 12.7 ± 5.8 vs 9.9 ± 6.3). Mediation analyses indicated significant indirect effects via university stress on depression (β = 0.11, p < .001) and anxiety (β = 0.09, p < .001), with smaller but consistent indirect effects via the learning environment. Hazing victimization harms student well-being by increasing academic stress and damaging the learning environment. Institutions should prevent hazing, improve faculty training, adopt confidential reporting, and support students. Monitoring stress and environment indicators can identify at-risk students, especially in the High Hazing group, and enable timely, evidence-based interventions.