Development and validation of a machine learning-based screening tool for early detection of adolescent suicide risk.

IF 7 1区医学 Q1 PSYCHIATRY

Journal of Child Psychology and Psychiatry Pub Date : 2026-05-03 DOI:10.1111/jcpp.70160

Weijia Li, Yongtian Cheng, Zeming Zhang, Yingying Yang, Yanpeng Jin, Zhenhua Cui, Juan Wang, Yinan Duan, Runsen Chen

{"title":"Development and validation of a machine learning-based screening tool for early detection of adolescent suicide risk.","authors":"Weijia Li, Yongtian Cheng, Zeming Zhang, Yingying Yang, Yanpeng Jin, Zhenhua Cui, Juan Wang, Yinan Duan, Runsen Chen","doi":"10.1111/jcpp.70160","DOIUrl":null,"url":null,"abstract":"Background: Adolescent suicide remains a significant public health concern, yet existing suicide screening instruments primarily focus on already manifested suicidal phenomena, underscoring the need for reliable and practical tools to enable early identification and intervention.Methods: Based on a large-scale school-based cohort study conducted in Southern China in 2022, this study aimed to develop and preliminarily validate two machine learning-based tools (a 51-item full version and an 11-item abbreviated version) designed to help identify adolescents at risk of developing suicide risk. The dataset was divided into two samples for tool development and longitudinal interview validation. During the tool development phase, LASSO regression was employed to select items with optimal contributions for recent suicide attempts from a multidimensional set of risk factors, followed by model training with multiple machine learning algorithms. The developed models were subsequently evaluated for their ability to predict suicide risk as assessed by the follow-up interview in the longitudinal validation phase.Results: Both versions of the screening tool demonstrated adequate discriminative ability, with the CatBoost algorithm outperforming others (AUROC ≥ 0.87). The abbreviated tool showed a slight trade-off between model precision and practicality, with a 0.02 reduction in AUROC, while still maintaining appropriate discrimination. Longitudinal validation using follow-up interview outcomes supported the predictive validity of both tools. These findings provide preliminary evidence for the utility of machine learning-based suicide risk screening tools among adolescents.Conclusions: This study provides evidence supporting the machine learning-based screening tools for early suicide risk detection in adolescents that integrates multidimensional vulnerabilities. The tools show promise in facilitating early identification and targeted interventions in school settings, addressing a critical need in adolescent mental health care. Nonetheless, further research is warranted to confirm their efficacy and support broader implementation.","PeriodicalId":187,"journal":{"name":"Journal of Child Psychology and Psychiatry","volume":" ","pages":""},"PeriodicalIF":7.0000,"publicationDate":"2026-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Child Psychology and Psychiatry","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1111/jcpp.70160","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PSYCHIATRY","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Adolescent suicide remains a significant public health concern, yet existing suicide screening instruments primarily focus on already manifested suicidal phenomena, underscoring the need for reliable and practical tools to enable early identification and intervention.

Methods: Based on a large-scale school-based cohort study conducted in Southern China in 2022, this study aimed to develop and preliminarily validate two machine learning-based tools (a 51-item full version and an 11-item abbreviated version) designed to help identify adolescents at risk of developing suicide risk. The dataset was divided into two samples for tool development and longitudinal interview validation. During the tool development phase, LASSO regression was employed to select items with optimal contributions for recent suicide attempts from a multidimensional set of risk factors, followed by model training with multiple machine learning algorithms. The developed models were subsequently evaluated for their ability to predict suicide risk as assessed by the follow-up interview in the longitudinal validation phase.

Results: Both versions of the screening tool demonstrated adequate discriminative ability, with the CatBoost algorithm outperforming others (AUROC ≥ 0.87). The abbreviated tool showed a slight trade-off between model precision and practicality, with a 0.02 reduction in AUROC, while still maintaining appropriate discrimination. Longitudinal validation using follow-up interview outcomes supported the predictive validity of both tools. These findings provide preliminary evidence for the utility of machine learning-based suicide risk screening tools among adolescents.

Conclusions: This study provides evidence supporting the machine learning-based screening tools for early suicide risk detection in adolescents that integrates multidimensional vulnerabilities. The tools show promise in facilitating early identification and targeted interventions in school settings, addressing a critical need in adolescent mental health care. Nonetheless, further research is warranted to confirm their efficacy and support broader implementation.

查看原文本刊更多论文

基于机器学习的青少年自杀风险早期检测筛查工具的开发和验证。

背景：青少年自杀仍然是一个重大的公共卫生问题，但现有的自杀筛查工具主要侧重于已经表现出来的自杀现象，强调需要可靠和实用的工具来实现早期识别和干预。方法：基于2022年在中国南方进行的一项大规模学校队列研究，本研究旨在开发并初步验证两种基于机器学习的工具（51项完整版本和11项简化版本），旨在帮助识别有自杀风险的青少年。数据集分为两个样本，用于工具开发和纵向访谈验证。在工具开发阶段，使用LASSO回归从多维风险因素集中选择对近期自杀企图有最佳贡献的项目，然后使用多种机器学习算法进行模型训练。随后，通过纵向验证阶段的随访访谈评估了开发的模型预测自杀风险的能力。结果：两种版本的筛选工具都表现出足够的判别能力，其中CatBoost算法优于其他算法（AUROC≥0.87）。简化的工具在模型精度和实用性之间表现出轻微的权衡，AUROC降低了0.02，同时仍然保持适当的区分。采用随访访谈结果的纵向验证支持这两种工具的预测有效性。这些发现为在青少年中使用基于机器学习的自杀风险筛查工具提供了初步证据。结论：本研究为基于机器学习的青少年早期自杀风险检测筛查工具提供了证据，该工具整合了多维脆弱性。这些工具有望促进学校环境中的早期识别和有针对性的干预，解决青少年精神卫生保健方面的关键需求。尽管如此，仍有必要进一步研究以证实其有效性并支持更广泛的实施。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Child Psychology and Psychiatry 医学-精神病学

CiteScore

13.80

自引率

5.30%

发文量

169

审稿时长

1 months

期刊介绍： The Journal of Child Psychology and Psychiatry (JCPP) is a highly regarded international publication that focuses on the fields of child and adolescent psychology and psychiatry. It is recognized for publishing top-tier, clinically relevant research across various disciplines related to these areas. JCPP has a broad global readership and covers a diverse range of topics, including: Epidemiology: Studies on the prevalence and distribution of mental health issues in children and adolescents. Diagnosis: Research on the identification and classification of childhood disorders. Treatments: Psychotherapeutic and psychopharmacological interventions for child and adolescent mental health. Behavior and Cognition: Studies on the behavioral and cognitive aspects of childhood disorders. Neuroscience and Neurobiology: Research on the neural and biological underpinnings of child mental health. Genetics: Genetic factors contributing to the development of childhood disorders. JCPP serves as a platform for integrating empirical research, clinical studies, and high-quality reviews from diverse perspectives, theoretical viewpoints, and disciplines. This interdisciplinary approach is a key feature of the journal, as it fosters a comprehensive understanding of child and adolescent mental health. The Journal of Child Psychology and Psychiatry is published 12 times a year and is affiliated with the Association for Child and Adolescent Mental Health (ACAMH), which supports the journal's mission to advance knowledge and practice in the field of child and adolescent mental health.