A nomogram model to predict the risk for etomidate use disorder with antisocial personality disorder

The co-occurrence of etomidate use disorder (EUD) and antisocial personality disorder (ASPD) poses significant challenges in clinical assessment and management. This study aims to develop a predictive model to estimate the risk of ASPD in EUD patients.

Male patients with EUD were recruited from a drug rehabilitation center between March and December 2024. Behavioral, health, psychological, and sociodemographic variables were collected, and diagnoses of ASPD were established using the Mini-International Neuropsychiatric Interview (MINI). The baseline variables were screened to identify potential predictors associated with EUD co-occurring with ASPD, via least absolute shrinkage and selection operator (LASSO) regression and Boruta algorithm. A predictive nomogram model was constructed based on multivariate logistic regression analysis. The performance of the predictive model was comprehensively evaluated through the ROC curve and calibration curve. Internal validation was performed using the Bootstrap method.

A total of 122 patients with EUD were included in the analysis, of whom 60 (49.2%) presented with ASPD. The final predictive model incorporated five variables: Craving (OR = 1.52; 95% CI: 1.08-2.15), Attentional impulsiveness (OR = 1.06; 95% CI: 1.03-1.09), Physical Aggression (OR = 1.04; 95% CI: 1.02-1.07), Trauma (OR = 3.08; 95% CI: 1.12-8.48), and Crime (OR = 3.48; 95% CI: 1.40-8.67). The ROC curve indicated AUC of 0.840 (p < 0.001, 95% CI: 0.770-0.909), and calibration curves indicated excellent agreement between predicted and observed outcomes in predicting EUD with comorbid ASPD.

We have successfully developed, evaluated and interpreted a nomogram model for predicting the co-occurrence of EUD and ASPD, which demonstrates favorable predictive performance and clinical utility.

Not applicable.