已发表论文

基于群体轨迹模型和家庭相关决定因素的决策树分析青少年非自杀自残行为轨迹的异质性

 

Authors Li L , Yang H 

Received 12 July 2023

Accepted for publication 19 August 2023

Published 25 August 2023 Volume 2023:16 Pages 3359—3371

DOI https://doi.org/10.2147/PRBM.S427090

Checked for plagiarism Yes

Review by Single anonymous peer review

Peer reviewer comments 2

Editor who approved publication: Professor Mei-Chun Cheung

Background: Non-suicidal self-injury (NSSI) is a global public health problem. While some studies have noted the importance of behavioural development patterns in NSSI, most NSSI research is based on cross-sectional survey data. Few studies explore the time-series heterogeneity of trajectories of NSSI symptoms and family-related influencing factors of NSSI.
Aim: The purpose of the study was to identify the heterogeneity in trajectories of adolescents’ NSSI behaviour and their family-related influencing factors.
Methods: The group-based trajectory model (GBTM) was used to identify the heterogeneity in the NSSI behaviour trajectories of 208 adolescents in China. Next, the decision tree model (DTM) was used to analyse which family related factors influence the trajectory type.
Results: The GBTM revealed two heterogeneous trajectories of NSSI behaviour: the high–risk and low-risk NSSI behaviour groups. Next, DTM’s average accuracy was 83.2%. A total of seven independent variables were used for the DTM: gender, number of NSSIs in the past month, and family economic, family structure, family conflict, parental psychological control, parental behavior control and family intimacy risks. Family conflict risk was located at the root node and was the most important factor.
Conclusion: Heterogeneity within the population should be considered in the management of adolescents’ NSSI behaviours. Further, from the perspective of family system theory and cumulative risk, focusing on the adverse effect of multiple risk factors on adolescents’ NSSI addiction is more meaningful rather than the impact of single risk factors. Studies that use time series data should focus on the trend of dynamic changes in NSSI addiction characteristics over time.
Keywords: non-suicidal self-injury, group-based, heterogeneity, trajectory, decision tree, influencing factor