A predictive model for depression risk in individuals with metabolic dysfunction-associated steatotic liver disease:...

Metabolic dysfunction-associated steatotic liver disease (MASLD), affecting 25-30% of adults globally, is strongly associated with depression, compounding morbidity and mortality. Despite their bidirectional relationship, tools to identify MASLD patients at high depression risk remain limited. This study aimed to develop and validate a predictive model for depression in MASLD using nationally representative data. Using 2017-2023 NHANES data, 6,107 MASLD participants were analyzed. Depression was defined as Patient Health Questionnaire-9 (PHQ-9) scores ≥10. LASSO regression with 10-fold cross-validation identified predictors, followed by multivariable logistic regression to construct a nomogram. Model performance was evaluated via area under the curve (AUC), calibration curves, and decision curve analysis (DCA). Ten predictors were retained: younger age, female gender, never-married status, low family poverty-to-income ratio (PIR), smoking, diabetes, arthritis, cardiovascular disease (CVD), chronic obstructive pulmonary disease (COPD), and platelet count. The model demonstrated robust discrimination (training set AUC = 0.733; testing set AUC = 0.758), excellent calibration (Hosmer-Lemeshow p > 0.05), and clinical utility across threshold probabilities < 40-50%. Socioeconomic factors (low PIR) and comorbidities (arthritis, CVD, COPD) showed strong associations with depression risk. This nomogram-based tool effectively stratifies depression risk in MASLD patients, integrating demographic, socioeconomic, and clinical variables. It offers clinicians a practical screening instrument for early psychological intervention, addressing the intertwined burden of metabolic and mental health disorders. Implementation could enhance holistic care and reduce adverse outcomes in this high-risk population.