Abstract
BACKGROUND: This study aimed to develop and validate a predictive model for major depression risk in adult patients with coronary heart disease (CHD), offering evidence for targeted prevention and intervention. METHODS: Using data from the National Health and Nutrition Examination Survey (NHANES) from 2005 to 2018, 1098 adults with CHD were included. A weighted logistic regression model was applied to construct and validate a nomogram-based prediction tool for major depression in this population. RESULTS: The weighted prevalence of major depression among these patients was 13.95%. Multivariate weighted logistic regression revealed that waist circumference, smoking status, arthritis, sleep disorders, and restricted work capacity were independent risk factors for major depression (odds ratio (OR) >1, p < 0.05). The areas under the receiver operating characteristic (ROC) curve in the nomogram model for both the development and validation cohorts were 0.816 (95% confidence interval (CI): 0.776-0.857) and 0.765 (95% CI: 0.699-0.832), respectively, indicating the model possessed strong discriminative ability. Brier scores in the development and validation cohorts were 0.107 and 0.127, respectively, both well below the 0.25 threshold, demonstrating good calibration. Decision curve analysis (DCA) showed that when the threshold probability for major depression ranged from 0.04 to 0.54 in the development group and from 0.08 to 0.52 in the validation group, the nomogram provided the highest clinical net benefit compared to "Treat All" and "Treat None" strategies, confirming its strong clinical utility. CONCLUSIONS: With a weighted prevalence of 13.95%, this nomogram model shows excellent predictive performance and clinical relevance for predicting major depression risk in patients with CHD. Thus, the model can be applied to aid healthcare professionals in identifying high-risk individuals and implementing targeted preventive strategies, potentially lowering the incidence of major depression in this patient population.