Development and Validation of a Predictive Model for Circadian Syndrome in Middle-aged and Elderly Adults

Objective: To construct and verify a predictive model for circadian syndrome (CircS) among middle-aged and elderly adults. Methods: A total of 5,977 individuals from the China Health and Retirement Longitudinal Study were enrolled. The least absolute shrinkage and selection operator method was used to identify potential features for multivariable logistic regression analyses. A nomogram was created utilizing the chosen predictors. The nomogram's discrimination, predictive performance, and clinical applicability were evaluated using the area under the curve (AUC), calibration curves, and decision curve analysis (DCA), respectively. Results: The prevalence of CircS among middle-aged and elderly adults was 30.11%. The nomogram comprised 11 predictors: gender, age, triglycerides, high-density lipoproteins, glucose, uric acid, sleep duration, depression score, systolic blood pressure, waist circumference, and body mass index. The AUC values for the model were 0.922 and 0.930 in the training and validation sets, respectively, indicating excellent discriminatory capability. The calibration curve of the nomogram showed strong concordance between the predicted and actual probabilities. The DCA suggested that the nomogram has clinical utility for CircS in middle-aged and elderly individuals. Conclusion: This study successfully developed a reliable nomogram model to predict CircS in middle-aged and elderly adults with high clinical predictive value. Clinicians can use it to identify high-risk populations and provide references for early prevention.