Construction of a decision tree algorithm to predict the risk of non-suicidal self-injurious behavior in young and middle-aged patients with major depressive disorder
Objective To construct a predictive model for the risk of non-suicidal self-injury(NSSI)in young and middle-aged patients with major depressive disorder based on the decision tree algorithm.Methods Young and middle-aged patients with major depressive disorder admitted to the Chongqing Mental Health Center from January 2021 to June 2023 were selected as the research subjects and their clinical data were collected.According to whether NSSI occurred or not,the patients were divided into the NSSI group(n=72)and the non-NSSI group(n=82).Univariate and multivariate logistic regression analyses were used to analyze the risk factors of NSSI in young and middle-aged patients with major depressive disorder.A deci-sion tree predictive model was established based on the results of logistic regression analysis and the Chi-square Automatic Interaction Detection(CHAID)algorithm.The predictive value of the model was evaluated using Receiver Operating Characteristic(ROC)curve.Results The incidence of NSSI in 154 young and mid-dle-aged patients with major depressive disorder was 46.8％.The results of single-factor analysis showed that there were no statistically significant differences(P＞0.05)in gender,age,education level,occupation,family composition,only child status,relationship status,fertility status,main place of residence,family economic source,and housing conditions between the two groups.Statistically significant differences(P＜0.05)were observed in living conditions,family economic status,duration of depression,SSRS score,PSS-Fa score,and DS score between the two groups.Multi-factor logistic regression analysis showed that lower family economic lev-el,longer duration of depression,SSRS score＜20 points,PSS-Fa score＜6 points,and DS score ≥47.2 points were independent risk factors for NSSI in young and middle-aged patients with major depressive disorder(P＜0.05).The constructed decision tree model had five layers and ten nodes,with family economic situation,duration of depression,SSRS score,and PSS-Fa score selected as the nodes of the model,among which the PSS-Fa score was the most important predictive factor.ROC curve analysis showed that the AUC of the model was 0.881(95％CI:0.844-0.918).Conclusion There are many influencing factors for the occurrence of NSSI in young and middle-aged patients with major depressive disorder.The decision tree model constructed based on these factors has a high predictive value for the risk of NSSI in young and middle-aged patients with major depressive disorder.
young and middle-agedmajor depressive disordernon-suicidal self-injurious behaviorin-fluencing factorsdecision tree
万方数据
维普
