No Cover Image

Journal article 188 views 16 downloads

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

Ryan Crowley Orcid Logo, Katherine Parkin Orcid Logo, Emma Rocheteau Orcid Logo, Efthalia Massou, Yasmin Friedmann, Ann John Orcid Logo, Rachel Sippy, Pietro Liò, Anna Moore Orcid Logo

BJPsych Open, Volume: 11, Issue: 3, Start page: e86

Swansea University Author: Ann John Orcid Logo

  • 69347.VOR.pdf

    PDF | Version of Record

    © The Author(s), 2025. Published by Cambridge University Press on behalf of Royal College of Psychiatrists. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (CC BY).

    Download (875.2KB)

Check full text

DOI (Published version): 10.1192/bjo.2025.32

Abstract

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 ide...

Full description

Published in: BJPsych Open
ISSN: 2056-4724
Published: Royal College of Psychiatrists 2025
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa69347
first_indexed 2025-04-24T11:55:53Z
last_indexed 2025-05-17T07:24:20Z
id cronfa69347
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2025-05-16T13:49:46.1227241</datestamp><bib-version>v2</bib-version><id>69347</id><entry>2025-04-24</entry><title>Machine learning for prediction of childhood mental health problems in social care</title><swanseaauthors><author><sid>ed8a9c37bd7b7235b762d941ef18ee55</sid><ORCID>0000-0002-5657-6995</ORCID><firstname>Ann</firstname><surname>John</surname><name>Ann John</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2025-04-24</date><deptcode>MEDS</deptcode><abstract>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&#x2019;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&#x2013;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.</abstract><type>Journal Article</type><journal>BJPsych Open</journal><volume>11</volume><journalNumber>3</journalNumber><paginationStart>e86</paginationStart><paginationEnd/><publisher>Royal College of Psychiatrists</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic>2056-4724</issnElectronic><keywords>Mental health services, medical technology, community mental health teams, machine learning methods, precision medicine</keywords><publishedDay>1</publishedDay><publishedMonth>5</publishedMonth><publishedYear>2025</publishedYear><publishedDate>2025-05-01</publishedDate><doi>10.1192/bjo.2025.32</doi><url/><notes/><college>COLLEGE NANME</college><department>Medical School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MEDS</DepartmentCode><institution>Swansea University</institution><apcterm>Another institution paid the OA fee</apcterm><funders>All research at the Department of Psychiatry in the University of Cambridge is supported by the National Institute for Health and Care Research (NIHR) Cambridge Biomedical Research Centre (grant number BRC-1215-20014) and NIHR Applied Research Collaboration East of England. A.M. is funded through a UK Research and Innovation Future Leaders Fellowship and an Anna Freud fellowship. The Delphi study was funded by Medical Research Council Adolescent Engagement Awards (number MR/T046430/1). Data access and data linkage were funded by What Works for Children&#x2019;s Social Care and Cambridgeshire and Peterborough NHS Foundation Trust. K.P. is funded by the NIHR School for Public Health Research (grant number PD-SPH-2015) and the NIHR Applied Research Collaboration East of England.</funders><projectreference/><lastEdited>2025-05-16T13:49:46.1227241</lastEdited><Created>2025-04-24T12:46:47.0243441</Created><path><level id="1">Faculty of Medicine, Health and Life Sciences</level><level id="2">Swansea University Medical School - Health Data Science</level></path><authors><author><firstname>Ryan</firstname><surname>Crowley</surname><orcid>0000-0002-1482-5631</orcid><order>1</order></author><author><firstname>Katherine</firstname><surname>Parkin</surname><orcid>0000-0001-7338-5667</orcid><order>2</order></author><author><firstname>Emma</firstname><surname>Rocheteau</surname><orcid>0000-0002-6450-0878</orcid><order>3</order></author><author><firstname>Efthalia</firstname><surname>Massou</surname><order>4</order></author><author><firstname>Yasmin</firstname><surname>Friedmann</surname><order>5</order></author><author><firstname>Ann</firstname><surname>John</surname><orcid>0000-0002-5657-6995</orcid><order>6</order></author><author><firstname>Rachel</firstname><surname>Sippy</surname><order>7</order></author><author><firstname>Pietro</firstname><surname>Li&#xF2;</surname><order>8</order></author><author><firstname>Anna</firstname><surname>Moore</surname><orcid>0000-0001-9614-3812</orcid><order>9</order></author></authors><documents><document><filename>69347__34094__ad7b55da57f544dca65e7053c2ac88e0.pdf</filename><originalFilename>69347.VOR.pdf</originalFilename><uploaded>2025-04-24T12:53:06.3940186</uploaded><type>Output</type><contentLength>896207</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; The Author(s), 2025. Published by Cambridge University Press on behalf of Royal College of Psychiatrists. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (CC BY).</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2025-05-16T13:49:46.1227241 v2 69347 2025-04-24 Machine learning for prediction of childhood mental health problems in social care ed8a9c37bd7b7235b762d941ef18ee55 0000-0002-5657-6995 Ann John Ann John true false 2025-04-24 MEDS 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. Journal Article BJPsych Open 11 3 e86 Royal College of Psychiatrists 2056-4724 Mental health services, medical technology, community mental health teams, machine learning methods, precision medicine 1 5 2025 2025-05-01 10.1192/bjo.2025.32 COLLEGE NANME Medical School COLLEGE CODE MEDS Swansea University Another institution paid the OA fee All research at the Department of Psychiatry in the University of Cambridge is supported by the National Institute for Health and Care Research (NIHR) Cambridge Biomedical Research Centre (grant number BRC-1215-20014) and NIHR Applied Research Collaboration East of England. A.M. is funded through a UK Research and Innovation Future Leaders Fellowship and an Anna Freud fellowship. The Delphi study was funded by Medical Research Council Adolescent Engagement Awards (number MR/T046430/1). Data access and data linkage were funded by What Works for Children’s Social Care and Cambridgeshire and Peterborough NHS Foundation Trust. K.P. is funded by the NIHR School for Public Health Research (grant number PD-SPH-2015) and the NIHR Applied Research Collaboration East of England. 2025-05-16T13:49:46.1227241 2025-04-24T12:46:47.0243441 Faculty of Medicine, Health and Life Sciences Swansea University Medical School - Health Data Science Ryan Crowley 0000-0002-1482-5631 1 Katherine Parkin 0000-0001-7338-5667 2 Emma Rocheteau 0000-0002-6450-0878 3 Efthalia Massou 4 Yasmin Friedmann 5 Ann John 0000-0002-5657-6995 6 Rachel Sippy 7 Pietro Liò 8 Anna Moore 0000-0001-9614-3812 9 69347__34094__ad7b55da57f544dca65e7053c2ac88e0.pdf 69347.VOR.pdf 2025-04-24T12:53:06.3940186 Output 896207 application/pdf Version of Record true © The Author(s), 2025. Published by Cambridge University Press on behalf of Royal College of Psychiatrists. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (CC BY). true eng https://creativecommons.org/licenses/by/4.0/
title Machine learning for prediction of childhood mental health problems in social care
spellingShingle Machine learning for prediction of childhood mental health problems in social care
Ann John
title_short 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_sort Machine learning for prediction of childhood mental health problems in social care
author_id_str_mv ed8a9c37bd7b7235b762d941ef18ee55
author_id_fullname_str_mv ed8a9c37bd7b7235b762d941ef18ee55_***_Ann John
author Ann John
author2 Ryan Crowley
Katherine Parkin
Emma Rocheteau
Efthalia Massou
Yasmin Friedmann
Ann John
Rachel Sippy
Pietro Liò
Anna Moore
format Journal article
container_title BJPsych Open
container_volume 11
container_issue 3
container_start_page e86
publishDate 2025
institution Swansea University
issn 2056-4724
doi_str_mv 10.1192/bjo.2025.32
publisher Royal College of Psychiatrists
college_str Faculty of Medicine, Health and Life Sciences
hierarchytype
hierarchy_top_id facultyofmedicinehealthandlifesciences
hierarchy_top_title Faculty of Medicine, Health and Life Sciences
hierarchy_parent_id facultyofmedicinehealthandlifesciences
hierarchy_parent_title Faculty of Medicine, Health and Life Sciences
department_str Swansea University Medical School - Health Data Science{{{_:::_}}}Faculty of Medicine, Health and Life Sciences{{{_:::_}}}Swansea University Medical School - Health Data Science
document_store_str 1
active_str 0
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.
published_date 2025-05-01T05:58:06Z
_version_ 1837234637317865472
score 11.068938

Similar Items