Sleep, Diet, and Environment: A Nomogram for Early Anxiety Detection in 11,654 Health Examination Adults

<title>Abstract</title> <p> Background Anxiety symptom is an increasingly important public health concern among adults undergoing routine health examinations, particularly in large-scale preventive healthcare settings without specialty mental health services. This study aimed to develop a nomogram model that integrates demographic, behavioral, dietary, environmental, and clinical factors to improve early screening of anxiety symptoms in a general adult population. Methods This study involved 11,654 adults undergoing routine health examinations. Anxiety symptoms were defined using the Self-Rating Anxiety Scale (SAS ≥ 50). Least absolute shrinkage and selection operator (LASSO) regression with 10-fold cross-validation was applied for feature selection, followed by multivariable logistic regression to construct the prediction <italic>model</italic> . Model performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves, Hosmer–Lemeshow goodness-of-fit tests, and decision curve analysis (DCA). The cohort was randomly divided into a training set (n = 8,157) and a validation set (n = 3,497). Results Fourteen predictors were identified, including sleep quality, binge eating behavior, educational level, dietary patterns, environmental exposures, and clinical history. The model showed good discrimination, with an area under the curve (AUC) of 0.748 in the training set and 0.732 in the validation set. Calibration curves demonstrated good agreement between predicted and observed outcomes. The Hosmer-Lemeshow test indicated no significant lack of fit (P &gt; 0.05). DCA showed consistent positive net benefit across threshold probabilities ranging from 0 to 44%, with the highest clinical utility observed at a 5% threshold. Conclusions This study developed and validated a nomogram for early screening of anxiety symptoms in a large health examination population. The model integrates multidimensional risk factors and demonstrates good discrimination, calibration, and clinical utility. The findings support its potential application in routine health screening for early anxiety detection. </p>