Machine learning for violence prediction: a systematic review and critical appraisal

TL;DR

This systematic review evaluates machine learning models for violence prediction, highlighting their limited clinical utility due to bias, overfitting, and poor generalizability, while suggesting future directions for improving model trustworthiness and applicability.

Contribution

The paper provides a comprehensive synthesis and critical appraisal of existing violence prediction models, emphasizing methodological quality and proposing key considerations for future research.

Findings

Most models report high AUCs (0.68-0.99) but lack calibration and external validation.

High risk of bias was found in 31 out of 38 studies, mainly in analysis.

Current models have limited clinical utility due to overfitting and low generalizability.

Abstract

Purpose To conduct a systematic review of machine learning models for predicting violent behaviour by synthesising and appraising their validity, usefulness, and performance. Methods We systematically searched nine bibliographic databases and Google Scholar up to September 2025 for development and/or validation studies on machine learning methods for predicting all forms of violent behaviour. We synthesised the results by summarising discrimination and calibration performance statistics and evaluated study quality by examining risk of bias and clinical utility. Results We identified 38 studies reporting the development and validation of 40 models. Most studies reported Area Under the Curve (AUC) as the discrimination statistic with a range of 0.68-0.99. Only eight studies reported calibration performance, and three studies reported external validation. 31 studies had a high risk of bias, mainly in the analysis domain, and three studies had low risk of bias. The overall clinical utility of violence prediction models is poor, as indicated by risks of overfitting due to small samples, lack of transparent reporting, and low generalisability. Conclusion Although black box machine learning models currently have limited applicability in clinical settings, they may show promise for identifying high-risk individuals. We recommend five key considerations for violence prediction modelling: (i) ensuring methodological quality (e.g. following guidelines) and interdisciplinary collaborations; (ii) using black box algorithms only for highly complex data; (iii) incorporating dynamic predictions to allow for risk monitoring; (iv) developing more trustworthy algorithms using explainable methods; and (v) applying causal machine learning approaches where appropriate.