Related Experiment Video
Updated: Aug 31, 2026
An Experimental Analysis of Children's Ability to Provide a False Report about a Crime
Published on: May 3, 2016
Predicting repeat self-harm in children--how accurate can we expect to be?
Prathiba Chitsabesan1, Richard Harrington, Valerie Harrington
1The University Department of Child and Adolescent Psychiatry, Central Manchester and Manchester Children's University Hospitals, Hospital Road, Pendlebury, Manchester, M27 4HA, UK.
Insights
Predicting repeat self-harm in children is challenging. Parental mental health and prior attempts are key predictors, but a predictive model showed low sensitivity for identifying at-risk youth.
Area of Science:
- Child and Adolescent Psychiatry
- Public Health
- Clinical Psychology
Background:
- Deliberate self-harm in children is a significant concern.
- Understanding predictors of repetition is crucial for intervention.
- Previous research highlights various risk factors.
Purpose of the Study:
- To identify variables that predict the repetition of deliberate self-harm in children.
- To evaluate the predictive accuracy of a model combining key variables.
- To inform service provision for at-risk youth.
Main Methods:
- Analysis of data from a randomized control trial involving children with deliberate self-poisoning.
- Collection of baseline and follow-up (2 and 6 months) outcome measures.
- Statistical modeling to identify predictors of repeat self-harm.
Main Results:
- Parental mental health and a history of previous self-harm attempts were the strongest predictors.
- A predictive model achieved high positive predictive value (86%) but low sensitivity (28%).
- Twenty-three out of 149 children (15.4%) repeated self-harm within the follow-up period.
Conclusions:
- Predicting repeat self-harm in children remains difficult despite comprehensive assessments.
- Current predictive models have limitations in sensitivity.
- Services need adaptation to account for the complexities of predicting repeat self-harm.
Abstract:
The main objective of the study was to find which variables predict repetition of deliberate self-harm in children. The study is based on a group of children who took part in a randomized control trial investigating the effects of a home-based family intervention for children who had deliberately poisoned themselves. These children had a range of baseline and outcome measures collected on two occasions (two and six months follow-up). Outcome data were collected from 149 (92 %) of the initial 162 children over the six months. Twenty-three children made a further deliberate self-harm attempt within the follow-up period. A number of variables at baseline were found to be significantly associated with repeat self-harm. Parental mental health and a history of previous attempts were the strongest predictors. A model of prediction of further deliberate self-harm combining these significant individual variables produced a high positive predictive value (86 %) but had low sensitivity (28 %). Predicting repeat self-harm in children is difficult, even with a comprehensive series of assessments over multiple time points, and we need to adapt services with this in mind. We propose a model of service provision which takes these findings into account.
