Predictors of suicide attempt in adolescents with a history of self-harm: a machine learning study combining psychosocial and actigraphy measures
Year: 2026
Type: preprint
Study design: Observational Study
Abstract: Youth suicide attempt is an important clinical and public health concern, yet prediction ofsuicide attempt among youth with a history of self-harm remains challenging. Disrupted states ofarousal may play an important role in the etiology of suicide attempt but have rarely been measuredobjectively in prospective studies. In this study, we examined to what extent actigraphy-derived arousalmarkers and questionnaire-based psychosocial measures collected at age 14 could predict self-reportedsuicide attempt at age 17 in a population-based sample of 1,321 adolescents with a history of self-harmfrom a United Kingdom cohort. We conducted both inferential analyses (univariable and multivariablelogistic regressions), and predictive (machine learning) modeling using nested cross-validation.Cannabis use, peer problems, depressive symptoms, higher BMI, lower self-esteem, and sex emergedas significant independent predictors of later reported suicide attempt. Machine learning modelsachieved moderate predictive performance (AUROC = 0.65–0.67), with relatively high sensitivity(0.71–0.81) but lower specificity (0.37–0.53). Peer problems and depressive symptoms wereconsistently among the most important predictors across models. Although some actigraphy-derivedmeasures contributed to prediction in specific models, two-stage analyses suggested little incrementalpredictive value beyond questionnaire-based measures alone. These findings highlight the importanceof cannabis use, interpersonal difficulties and depressive symptoms in predicting suicide attemptamong adolescent who self-harmed and suggest a limited additional contribution of actigraphy-derivedarousal markers for long-term prediction.
Source: PsyArXiv (OSF Preprints)
Language: English
Subfield: Clinical Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7212422311