Stress, Not Stress–Body Image Interaction, Linked to Adolescent Depressive Symptoms

Researchers analyzed Korea Youth Risk Behavior Survey data from 201,895 adolescents in 2022–2025. Data from 2022–2023 (n=99,235) were used to develop models, with 2024–2025 data (n=102,660) providing temporal validation. The prespecified analysis used multivariable logistic regression to assess perceived stress, body image and their interaction in relation to self-reported depressive symptoms; additional analyses tested alternative body-image specifications and model performance.

Perceived stress was associated with depressive symptoms (OR 2.715, 95% CI 2.555–2.885). The prespecified stress-by-body-image interaction was not significant (OR 0.996, 95% CI 0.977–1.014; p=.640), and survey-weighted analysis was similar. Post hoc categorical modeling found an overall interaction, but effects varied across categories and no individual interaction remained significant after multiplicity correction in the weighted analysis. In temporal validation, logistic regression had an AUC of 0.736, versus 0.712 for a balanced random forest.

The findings do not support the proposed stress-related amplification of body-image vulnerability. The study identifies stress as a consistent correlate, not a demonstrated cause. Results concern self-reported symptoms, and the exploratory category-specific findings need independent confirmation. The authors caution that the random-forest probabilities overestimated absolute risk and are not calibrated clinical estimates.