Psychological risk factors for non-suicidal self-injury in sexually traumatized adolescents: a cross-sectional analysis with random forest classification

Non-suicidal self-injury (NSSI) is a significant mental health issue among adolescents, especially those with histories of sexual trauma. This study examined the individual and synergistic roles of posttraumatic stress symptoms (PTSS), depression, anxiety, and sleep quality in predicting NSSI among Iranian adolescents. We conducted a cross-sectional study with 448 Iranian adolescents (aged 14–18 years; 60.7% female) who had confirmed sexual abuse histories, recruited from Tehran psychology clinics. Participants completed Persian-language self-report measures to assess NSSI, PTSS, depression, anxiety, and sleep quality. Hierarchical logistic regression examined the incremental contributions of each predictor to NSSI. Additionally, a Random Forest classifier with a 70/30 train-test split (train n = 313, test n = 135) was implemented to evaluate predictive accuracy and identify key risk factors. Model interpretability was enhanced using Permutation Importance and SHAP analyses. Among 448 adolescents, 39.7% ( n = 179) reported NSSI. All predictors showed significant positive associations with NSSI (0.18 > r < 0.36, p < 0.01). Hierarchical logistic regression revealed PTSS as the primary predictor (OR = 1.07; p < 0.001), accounting for 17.9% of variance, with other factors becoming nonsignificant in its presence. Random Forest classification achieved 74.07% accuracy on the held-out test set. Permutation Importance analysis identified overall PTSS (mean accuracy decrease = 0.103) and avoidance/numbing symptoms (0.080) as the leading contributors to classification, whereas hyperarousal symptoms showed negligible importance (–0.002). SHAP analysis further indicated that sleep difficulties exerted a strong influence on the magnitude of model predictions (mean |SHAP| = 0.076) despite a modest contribution to overall accuracy. Findings emphasize the strong association of PTSS in NSSI among sexually traumatized Iranian adolescents, supporting trauma-focused interventions supplemented by sleep-related therapies. Future longitudinal research is needed to clarify temporal relationships and optimize culturally sensitive treatments.

Authors

Institutions

Publication Details

Journal
Child and Adolescent Psychiatry and Mental Health
Published
2026-09-06
DOI
https://doi.org/10.1186/s13034-026-01175-5
Primary Topic
Suicide and Self-Harm Studies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Psychological risk factors for non-suicidal self-injury in sexually traumatized adolescents: a cross-sectional analysis with random forest classification

Banafsheh Mohajerin, Maryam Raissi, Amirhossein Rasouli, Erfan Salehi
Child and Adolescent Psychiatry and Mental Health
Suicide and Self-Harm Studies
article

Psychological risk factors for non-suicidal self-injury in sexually traumatized adolescents: a cross-sectional analysis with random forest classification

Banafsheh Mohajerin, Maryam Raissi, Amirhossein Rasouli, Erfan Salehi
article en

Abstract

Non-suicidal self-injury (NSSI) is a significant mental health issue among adolescents, especially those with histories of sexual trauma. This study examined the individual and synergistic roles of posttraumatic stress symptoms (PTSS), depression, anxiety, and sleep quality in predicting NSSI among Iranian adolescents. We conducted a cross-sectional study with 448 Iranian adolescents (aged 14–18 years; 60.7% female) who had confirmed sexual abuse histories, recruited from Tehran psychology clinics. Participants completed Persian-language self-report measures to assess NSSI, PTSS, depression, anxiety, and sleep quality. Hierarchical logistic regression examined the incremental contributions of each predictor to NSSI. Additionally, a Random Forest classifier with a 70/30 train-test split (train n = 313, test n = 135) was implemented to evaluate predictive accuracy and identify key risk factors. Model interpretability was enhanced using Permutation Importance and SHAP analyses. Among 448 adolescents, 39.7% ( n = 179) reported NSSI. All predictors showed significant positive associations with NSSI (0.18 > r < 0.36, p < 0.01). Hierarchical logistic regression revealed PTSS as the primary predictor (OR = 1.07; p < 0.001), accounting for 17.9% of variance, with other factors becoming nonsignificant in its presence. Random Forest classification achieved 74.07% accuracy on the held-out test set. Permutation Importance analysis identified overall PTSS (mean accuracy decrease = 0.103) and avoidance/numbing symptoms (0.080) as the leading contributors to classification, whereas hyperarousal symptoms showed negligible importance (–0.002). SHAP analysis further indicated that sleep difficulties exerted a strong influence on the magnitude of model predictions (mean |SHAP| = 0.076) despite a modest contribution to overall accuracy. Findings emphasize the strong association of PTSS in NSSI among sexually traumatized Iranian adolescents, supporting trauma-focused interventions supplemented by sleep-related therapies. Future longitudinal research is needed to clarify temporal relationships and optimize culturally sensitive treatments.

Child and Adolescent Psychiatry and Mental Health
Islamic Azad University, Science and Research Branch (IR), Islamic Azad University of Hamedan (IR), Hamedan University of Technology (IR), Shahid Beheshti University of Medical Sciences (IR)
Gender equality
Openalex Percentile: Top 6%
Suicide and Self-Harm Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.