A prediction model for non-suicidal self-injury behavior risk among children aged 10 to 13 years
10.19485/j.cnki.issn2096-5087.2025.06.002
- Author:
GONG Liangliang
;
RONG Zhidong
- Publication Type:Journal Article
- Keywords:
non-suicidal self-injury;
nomogram;
prediction model
- From:
Journal of Preventive Medicine
2025;37(6):546-550
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To establish a prediction model for non-suicidal self-injury (NSSI) behavior risk among children aged 10 to 13 years, so as to provide the basis for early prevention and intervention of NSSI behavior in children.
Method:Students aged 10 to 13 years from four primary schools in Cixi City, Zhejiang Province from 2023 to 2024 were selected using a stratified cluster sampling method. Basic information, parents' marital status, family functioning, peer relationships, and academic satisfaction were collected through a questionnaire survey. The NSSI behavior in the past year was investigated by the Adolescent NSSI Behavior Rating Scale. The influencing factors for NSSI behavior were screened using a multivariable logistic regression model. A nomogram was established and verified with receiver operating characteristic (ROC) curve. Hosmer-Lemeshow goodness of fit test was used to evaluate the fitting effect and Bootstrap method was used for internal verification.
Results:A total of 986 students were surveyed, including 516 boys (52.33%) and 470 girls (47.67%). In the past year, 173 students (17.55%) engaged in NSSI behavior. The multivariable logistic regression model identified four influencing factors of NSSI behavior, and the prediction model was established as follows: ln[p/(1-p)]=-2.930+0.747×recent unpleasant events in the past week+0.441×having new media accounts+0.972×self-rated poor health status+0.557×self-rated general health status+0.745×family functioning. The area under the ROC curve was 0.791 (95%CI: 0.754-0.827), with a sensitivity of 0.792 and a specificity of 0.636.
Conclusions:The constructed nomogram model demonstrates a certain degree of discrimination, with good calibration and stability. It can be used to roughly predict the risk of NSSI behavior occurrence among children aged 10 to 13 years.
- Full text:202511051635224202310~13岁儿童非自杀性自伤行为风险预测模型研究.pdf
