Construction and validation of a risk prediction model for suicidal ideation in first-episode schizophrenia patients based on glucose metabolism disorders

LING Liyuan; LANG Xiao'e; WANG Meixia; DU Weinan

doi:10.12102/j.issn.1009-6493.2026.03.003

Abstract：Objective:To develop a risk prediction model for suicidal ideation in first-episode schizophrenia patients based on glucose metabolism disorders.Methods:A total of 333 patients with schizophrenia were randomly divided into a training group(n=233)and a validation group(n=100)in a 7∶3 ratio.Positive and Negative Syndrome Scale(PANSS)were used to evaluate psychiatric symptoms.Depressive and cognitive factors were evaluated based on the five-factor model.Fasting blood glucose,glycated hemoglobin,and fasting insulin were collected.The insulin resistance index was calculated.Patients were categorized into groups with and without suicidal ideation by using the Beck Suicide Ideation Scale.In the training group,relevant variables were identified through LASSO and Logistic regression analysis.A Nomogram was constructed to predict the risk of suicidal ideation.The area under the ROC curve(AUC)was used to evaluate the predictive ability of the model.Calibration curve was used to evaluate accuracy.Decision curve analysis(DCA)was used to evaluate clinical practicality.Results:A Nomogram was constructed based on positive symptoms scores,depressive factors scores,fasting blood glucose,and fasting insulin.The AUC of the training group and validation group were 0.776(95%CI 0.715-0.837)and 0.763(95%CI 0.665-0.860),respectively,indicating good discriminative ability of the model.Calibration analysis and DCA confirmed the clinical applicability of the nomogram.Conclusions:The Nomogram constructed in this study for predicting the risk of suicidal ideation in patients with schizophrenia could help identify high-risk suicide patients in clinical practice,provide reference for timely implementation of prevention strategies,and reduce suicide risk.