Machine learning for prediction of childhood mental health problems in social care

Background Rates of childhood mental health problems are increasing in the UK. Early identification of childhood mental health problems is challenging but critical to children’s future psychosocial development. This is particularly important for children with social care contact because earlier id...

Full description

Saved in:
Bibliographic Details
Main Authors: Ryan Crowley, Katherine Parkin, Emma Rocheteau, Efthalia Massou, Yasmin Friedmann, Ann John, Rachel Sippy, Pietro Liò, Anna Moore
Format: Article
Language:English
Published: Cambridge University Press 2025-05-01
Series:BJPsych Open
Subjects:
Online Access:https://www.cambridge.org/core/product/identifier/S2056472425000328/type/journal_article
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850185668045570048
author Ryan Crowley
Katherine Parkin
Emma Rocheteau
Efthalia Massou
Yasmin Friedmann
Ann John
Rachel Sippy
Pietro Liò
Anna Moore
author_facet Ryan Crowley
Katherine Parkin
Emma Rocheteau
Efthalia Massou
Yasmin Friedmann
Ann John
Rachel Sippy
Pietro Liò
Anna Moore
author_sort Ryan Crowley
collection DOAJ
description Background Rates of childhood mental health problems are increasing in the UK. Early identification of childhood mental health problems is challenging but critical to children’s future psychosocial development. This is particularly important for children with social care contact because earlier identification can facilitate earlier intervention. Clinical prediction tools could improve these early intervention efforts. Aims Characterise a novel cohort consisting of children in social care and develop effective machine learning models for prediction of childhood mental health problems. Method We used linked, de-identified data from the Secure Anonymised Information Linkage Databank to create a cohort of 26 820 children in Wales, UK, receiving social care services. Integrating health, social care and education data, we developed several machine learning models aimed at predicting childhood mental health problems. We assessed the performance, interpretability and fairness of these models. Results Risk factors strongly associated with childhood mental health problems included age, substance misuse and being a looked after child. The best-performing model, a gradient boosting classifier, achieved an area under the receiver operating characteristic curve of 0.75 (95% CI 0.73–0.78). Assessments of algorithmic fairness showed potential biases within these models. Conclusions Machine learning performance on this prediction task was promising. Predictive performance in social care settings can be bolstered by linking diverse routinely collected data-sets, making available a range of heterogenous risk factors relating to clinical, social and environmental exposures.
format Article
id doaj-art-27dcd198d9a64fdfb5bc4082ff9ece7a
institution OA Journals
issn 2056-4724
language English
publishDate 2025-05-01
publisher Cambridge University Press
record_format Article
series BJPsych Open
spelling doaj-art-27dcd198d9a64fdfb5bc4082ff9ece7a2025-08-20T02:16:39ZengCambridge University PressBJPsych Open2056-47242025-05-011110.1192/bjo.2025.32Machine learning for prediction of childhood mental health problems in social careRyan Crowley0https://orcid.org/0000-0002-1482-5631Katherine Parkin1https://orcid.org/0000-0001-7338-5667Emma Rocheteau2https://orcid.org/0000-0002-6450-0878Efthalia Massou3Yasmin Friedmann4Ann John5https://orcid.org/0000-0002-5657-6995Rachel Sippy6Pietro Liò7Anna Moore8https://orcid.org/0000-0001-9614-3812New York University Grossman School of Medicine, New York, USDepartment of Public Health and Primary Care, University of Cambridge, Cambridge, UK Department of Psychiatry, University of Cambridge, Cambridge, UK Cambridge Public Health, University of Cambridge, Cambridge, UKDepartment of Computer Science, University of Cambridge, Cambridge, UKDepartment of Public Health and Primary Care, University of Cambridge, Cambridge, UKNeath Port Talbot County Borough Council, Port Talbot, UKPopulation Psychiatry, Suicide and Informatics, Swansea University Medical School, Swansea, UKDepartment of Psychiatry, University of Cambridge, Cambridge, UKDepartment of Computer Science, University of Cambridge, Cambridge, UKDepartment of Psychiatry, University of Cambridge, Cambridge, UK Anna Freud, London, UK Cambridgeshire and Peterborough NHS Foundation Trust, Cambridge, UK Background Rates of childhood mental health problems are increasing in the UK. Early identification of childhood mental health problems is challenging but critical to children’s future psychosocial development. This is particularly important for children with social care contact because earlier identification can facilitate earlier intervention. Clinical prediction tools could improve these early intervention efforts. Aims Characterise a novel cohort consisting of children in social care and develop effective machine learning models for prediction of childhood mental health problems. Method We used linked, de-identified data from the Secure Anonymised Information Linkage Databank to create a cohort of 26 820 children in Wales, UK, receiving social care services. Integrating health, social care and education data, we developed several machine learning models aimed at predicting childhood mental health problems. We assessed the performance, interpretability and fairness of these models. Results Risk factors strongly associated with childhood mental health problems included age, substance misuse and being a looked after child. The best-performing model, a gradient boosting classifier, achieved an area under the receiver operating characteristic curve of 0.75 (95% CI 0.73–0.78). Assessments of algorithmic fairness showed potential biases within these models. Conclusions Machine learning performance on this prediction task was promising. Predictive performance in social care settings can be bolstered by linking diverse routinely collected data-sets, making available a range of heterogenous risk factors relating to clinical, social and environmental exposures. https://www.cambridge.org/core/product/identifier/S2056472425000328/type/journal_articleMental health servicesmedical technologycommunity mental health teamsmachine learning methodsprecision medicine
spellingShingle Ryan Crowley
Katherine Parkin
Emma Rocheteau
Efthalia Massou
Yasmin Friedmann
Ann John
Rachel Sippy
Pietro Liò
Anna Moore
Machine learning for prediction of childhood mental health problems in social care
BJPsych Open
Mental health services
medical technology
community mental health teams
machine learning methods
precision medicine
title Machine learning for prediction of childhood mental health problems in social care
title_full Machine learning for prediction of childhood mental health problems in social care
title_fullStr Machine learning for prediction of childhood mental health problems in social care
title_full_unstemmed Machine learning for prediction of childhood mental health problems in social care
title_short Machine learning for prediction of childhood mental health problems in social care
title_sort machine learning for prediction of childhood mental health problems in social care
topic Mental health services
medical technology
community mental health teams
machine learning methods
precision medicine
url https://www.cambridge.org/core/product/identifier/S2056472425000328/type/journal_article
work_keys_str_mv AT ryancrowley machinelearningforpredictionofchildhoodmentalhealthproblemsinsocialcare
AT katherineparkin machinelearningforpredictionofchildhoodmentalhealthproblemsinsocialcare
AT emmarocheteau machinelearningforpredictionofchildhoodmentalhealthproblemsinsocialcare
AT efthaliamassou machinelearningforpredictionofchildhoodmentalhealthproblemsinsocialcare
AT yasminfriedmann machinelearningforpredictionofchildhoodmentalhealthproblemsinsocialcare
AT annjohn machinelearningforpredictionofchildhoodmentalhealthproblemsinsocialcare
AT rachelsippy machinelearningforpredictionofchildhoodmentalhealthproblemsinsocialcare
AT pietrolio machinelearningforpredictionofchildhoodmentalhealthproblemsinsocialcare
AT annamoore machinelearningforpredictionofchildhoodmentalhealthproblemsinsocialcare