Construction of a suicide risk prediction model for adolescent patients with depression based on psychosocial risk factors

LUO Jingyi; HONG Gang; REN Xiaoqun

doi:10.12104/j.issn.1674-4748.2024.23.003

Abstract：Objective:To construct a suicide risk prediction model for adolescent patients with depression based on psychosocial risk factors,and verify its predictive value.Methods:A total of 160 adolescent patients with depression admitted to the hospital from January 2019 to October 2023 were selected as the research objects,and were divided into the modeling group(112 cases)and the verification group(48 cases)according to the allocation ratio of 7∶3.Patients in the modeling group were divided into the short-sighted group(51 cases)and the normal group(61 cases)according to whether they had suicidal behavior.Clinical baseline data of patients in the modeling group were collected,and inter-group factor analysis was conducted between the short-sighted group and the normal group to determine the risk factors affecting suicide tendency of adolescents with depression in the collected information.Based on independent influencing factors and data collection of verification group,a nomogram prediction model was constructed and verified.Factor analysis Results:suggest that teenagers depression patients'Social Support rating Scale(Social Support Rat-ing Scale,SSRS)of objective Support,subjective Support,Support utilization rating,family function Scale(family assessment device,The role,affective intervention,general function score,internal and external orientation,psychosis and neurotic personality trait score were all related factors affecting the suicidal tendency of patients(all P＜0.05).SSRS subjective support and support utilization score were independent protective factors affecting suicide tendency(all P＜0.05).FAD emotional intervention score and neurotic personality trait score were independent risk factors for patients'suicidal tendency(both P＜0.05).A nomogram prediction model was constructed based on the data of the verification group.The model verification results showed that the Area Under Curve(AUC)of the receiver operating characteristic(ROC)curve of the modeling group was 0.922.The AUC of ROC in the verification group was 0.968,and the calibration curve fit was good(P＜0.05 for all).Conclusions:The prediction model based on the social psychological risk factors can provide data reference for the formulation of psychological intervention and treatment,or avoid the occurrence of suicide in adolescents with depression.