Abstract
BACKGROUND: Low back pain (LBP), a prevalent disorder among middle-aged and older adults, imposes substantial medical and socioeconomic burdens on individuals and society. Worse still, comorbid depression is frequently present in chronic LBP, which aggravates the functional prognosis. It is of significance to identify relevant predictors associated with the occurrence of depression in LBP. Therefore, this study developed a risk prediction model to estimate depression risk in LBP patients, which may support its early detection and intervention. METHODS: This study used representative data from the China Health and Retirement Longitudinal Study. Thirty-one candidate variables encompassing socio-demographic, pain-related, behavioral, health status, and psychological factors were analyzed. Participants were randomly split into the training and validation cohorts (7:3 ratio). LASSO regression with tenfold cross-validation was used to select predictors. A logistic regression model was constructed, and a nomogram was developed based on the final predictors. Model performance was evaluated by area under the receiver operating characteristic curve (AUC), calibration plots, and decision curve analysis (DCA). RESULTS: A total of 1,693 participants with LBP from 2018 to 2020 were included, with a depression incidence of 29.4%. Multivariable logistic regression identified predictors of depression in LBP, involving gender, education, sleep quality, chronic diseases, life satisfaction, cognitive function, pain severity, use of analgesics, and among others. The nomogram demonstrated good discrimination (AUC 0.736 and 0.718 in the training and validation set, respectively). Hosmer-Lemeshow tests indicated a good model fit (P > 0.05). DCA confirmed favorable clinical utility. CONCLUSION: The developed model provides a practical tool for assessing the risk of depression in middle-aged and older adults with LBP, supporting early identification and targeted preventive strategies in clinical practice.