Reappraisal of Mental Illness-Related Beliefs During Psychiatric Clinical Exposure: Perceived Dangerousness, Ruminative Thought Style, and Cognitive–Metacognitive Characteristics

Stigma-related beliefs toward mental illness may change during psychiatric clinical exposure, but individual differences in this process remain insufficiently understood. This single-center observational pre–post study examined observed changes in mental illness-related beliefs among 185 medical interns who completed an approximately 30-day psychiatry clerkship between May 2025 and May 2026, with perceived dangerousness defined as the primary outcome. Beliefs were assessed before and after the exposure period using the Beliefs Toward Mental Illness Scale. Associations between baseline cognitive–metacognitive characteristics and observed change were evaluated using paired comparisons, linear mixed-effects models, and sensitivity analyses. Dangerousness scores decreased by the end of the exposure period, mean change = −2.66, 95% CI [−3.48, −1.81], q < 0.001, dz = −0.47. Smaller decreases were observed in Helplessness and Deterioration in Interpersonal Relationships, whereas no significant change was found in the two-item Shame subscale. Higher baseline ruminative thought style was associated with a more limited decrease in dangerousness scores; B = 1.70, SE = 0.55, 95% CI [0.62, 2.77], q = 0.023. Although this association was consistent across sensitivity analyses, its incremental contribution was small (ΔR2marginal = 0.017; f2 = 0.018 in the focused model). The finding should therefore be interpreted as a quantitatively modest association with observed belief change, not as a clinically meaningful predictor. Because the study included no concurrent control group or experimental manipulation and was conducted at one center, it does not establish a causal clerkship effect or support educational screening based on ruminative thought style.