View in EDS HTML Full Text PDF Full Text

Predicting criminal offence in adolescents who exhibit antisocial behaviour: a machine learning study using data from a large randomised controlled trial of multisystemic therapy.

Saved in:
Bibliographic Details
Title: Predicting criminal offence in adolescents who exhibit antisocial behaviour: a machine learning study using data from a large randomised controlled trial of multisystemic therapy.
Authors: Suh, Jae Won, Saunders, Rob, Simes, Elizabeth, Delamain, Henry, Butler, Stephen, Cottrell, David, Kraam, Abdullah, Scott, Stephen, Goodyer, Ian M, Wason, James, Pilling, Stephen, Fonagy, Peter
Source: European Child & Adolescent Psychiatry. May2025, Vol. 34 Issue 5, p1579-1588. 10p.
Subjects: Juvenile delinquency, Risk assessment, Boosting algorithms, Human services programs, Prediction models, Research funding, Logistic regression analysis, Decision making in clinical medicine, Severity of illness index, Descriptive statistics, Recidivism, Machine learning, Confidence intervals, Antisocial personality disorders, Algorithms, Psychosocial factors, Adolescence
Geographic Terms: England
Abstract: Introduction: Accurate prediction of short-term offending in young people exhibiting antisocial behaviour could support targeted interventions. Here we develop a set of machine learning (ML) models that predict offending status with good accuracy; furthermore, we show interpretable ML analyses can complement models to inform clinical decision-making. Methods: This study included 679 individuals aged 11–17 years who displayed moderate-to-severe antisocial behaviour, from a controlled trial of Multisystemic-therapy in England. The outcome was any criminal offence in the 18 months after study baseline. Four types of ML algorithms were trained: logistic regression, elastic net regression, random forest, and gradient boosting machine (GBM). Prediction models were developed (1) using predictors readily available to clinicians (e.g. sociodemographics, previous convictions), and (2) with additional information (e.g. parenting). Model agnostic feature importance values were calculated and the most important predictors identified. Nested cross-validation with 100 iterations of random data splits and 10-fold cross-validation within each iteration was employed, and the average predictive performance was reported. Results: Among the ML models using readily available predictors, the GBM is the strongest model (AUC 0.85, 95% CI 0.85–0.86); the other models have average AUCs of 0.82. This performance was better than using only the total number of previous offences as the predictor (0.67, 0.66–0.68), and the model simply assuming past offending status as the prediction (0.81, 0.80–0.81). Additional predictors slightly increased the performance of logistic regression and random forest models but decreased the performance of elastic net regression and gradient boosting machine-based models. Conclusion: The potential utility of ML approaches for accurately predicting criminal offences in high-risk youth is demonstrated. Interpretable ML-based predictive models could be utilised in youth services or research to help develop and deliver effective interventions. [ABSTRACT FROM AUTHOR]
Copyright of European Child & Adolescent Psychiatry is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Psychology and Behavioral Sciences Collection
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: pbh
DbLabel: Psychology and Behavioral Sciences Collection
An: 185520000
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Predicting criminal offence in adolescents who exhibit antisocial behaviour: a machine learning study using data from a large randomised controlled trial of multisystemic therapy.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Suh%2C+Jae+Won%22">Suh, Jae Won</searchLink><br /><searchLink fieldCode="AR" term="%22Saunders%2C+Rob%22">Saunders, Rob</searchLink><br /><searchLink fieldCode="AR" term="%22Simes%2C+Elizabeth%22">Simes, Elizabeth</searchLink><br /><searchLink fieldCode="AR" term="%22Delamain%2C+Henry%22">Delamain, Henry</searchLink><br /><searchLink fieldCode="AR" term="%22Butler%2C+Stephen%22">Butler, Stephen</searchLink><br /><searchLink fieldCode="AR" term="%22Cottrell%2C+David%22">Cottrell, David</searchLink><br /><searchLink fieldCode="AR" term="%22Kraam%2C+Abdullah%22">Kraam, Abdullah</searchLink><br /><searchLink fieldCode="AR" term="%22Scott%2C+Stephen%22">Scott, Stephen</searchLink><br /><searchLink fieldCode="AR" term="%22Goodyer%2C+Ian+M%22">Goodyer, Ian M</searchLink><br /><searchLink fieldCode="AR" term="%22Wason%2C+James%22">Wason, James</searchLink><br /><searchLink fieldCode="AR" term="%22Pilling%2C+Stephen%22">Pilling, Stephen</searchLink><br /><searchLink fieldCode="AR" term="%22Fonagy%2C+Peter%22">Fonagy, Peter</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22European+Child+%26+Adolescent+Psychiatry%22">European Child & Adolescent Psychiatry</searchLink>. May2025, Vol. 34 Issue 5, p1579-1588. 10p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Juvenile+delinquency%22">Juvenile delinquency</searchLink><br /><searchLink fieldCode="DE" term="%22Risk+assessment%22">Risk assessment</searchLink><br /><searchLink fieldCode="DE" term="%22Boosting+algorithms%22">Boosting algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22Human+services+programs%22">Human services programs</searchLink><br /><searchLink fieldCode="DE" term="%22Prediction+models%22">Prediction models</searchLink><br /><searchLink fieldCode="DE" term="%22Research+funding%22">Research funding</searchLink><br /><searchLink fieldCode="DE" term="%22Logistic+regression+analysis%22">Logistic regression analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Decision+making+in+clinical+medicine%22">Decision making in clinical medicine</searchLink><br /><searchLink fieldCode="DE" term="%22Severity+of+illness+index%22">Severity of illness index</searchLink><br /><searchLink fieldCode="DE" term="%22Descriptive+statistics%22">Descriptive statistics</searchLink><br /><searchLink fieldCode="DE" term="%22Recidivism%22">Recidivism</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br /><searchLink fieldCode="DE" term="%22Confidence+intervals%22">Confidence intervals</searchLink><br /><searchLink fieldCode="DE" term="%22Antisocial+personality+disorders%22">Antisocial personality disorders</searchLink><br /><searchLink fieldCode="DE" term="%22Algorithms%22">Algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22Psychosocial+factors%22">Psychosocial factors</searchLink><br /><searchLink fieldCode="DE" term="%22Adolescence%22">Adolescence</searchLink>
– Name: SubjectGeographic
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22England%22">England</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Introduction: Accurate prediction of short-term offending in young people exhibiting antisocial behaviour could support targeted interventions. Here we develop a set of machine learning (ML) models that predict offending status with good accuracy; furthermore, we show interpretable ML analyses can complement models to inform clinical decision-making. Methods: This study included 679 individuals aged 11–17 years who displayed moderate-to-severe antisocial behaviour, from a controlled trial of Multisystemic-therapy in England. The outcome was any criminal offence in the 18 months after study baseline. Four types of ML algorithms were trained: logistic regression, elastic net regression, random forest, and gradient boosting machine (GBM). Prediction models were developed (1) using predictors readily available to clinicians (e.g. sociodemographics, previous convictions), and (2) with additional information (e.g. parenting). Model agnostic feature importance values were calculated and the most important predictors identified. Nested cross-validation with 100 iterations of random data splits and 10-fold cross-validation within each iteration was employed, and the average predictive performance was reported. Results: Among the ML models using readily available predictors, the GBM is the strongest model (AUC 0.85, 95% CI 0.85–0.86); the other models have average AUCs of 0.82. This performance was better than using only the total number of previous offences as the predictor (0.67, 0.66–0.68), and the model simply assuming past offending status as the prediction (0.81, 0.80–0.81). Additional predictors slightly increased the performance of logistic regression and random forest models but decreased the performance of elastic net regression and gradient boosting machine-based models. Conclusion: The potential utility of ML approaches for accurately predicting criminal offences in high-risk youth is demonstrated. Interpretable ML-based predictive models could be utilised in youth services or research to help develop and deliver effective interventions. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of European Child & Adolescent Psychiatry is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=pbh&AN=185520000
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s00787-024-02592-7
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 10
        StartPage: 1579
    Subjects:
      – SubjectFull: Juvenile delinquency
        Type: general
      – SubjectFull: Risk assessment
        Type: general
      – SubjectFull: Boosting algorithms
        Type: general
      – SubjectFull: Human services programs
        Type: general
      – SubjectFull: Prediction models
        Type: general
      – SubjectFull: Research funding
        Type: general
      – SubjectFull: Logistic regression analysis
        Type: general
      – SubjectFull: Decision making in clinical medicine
        Type: general
      – SubjectFull: Severity of illness index
        Type: general
      – SubjectFull: Descriptive statistics
        Type: general
      – SubjectFull: Recidivism
        Type: general
      – SubjectFull: Machine learning
        Type: general
      – SubjectFull: Confidence intervals
        Type: general
      – SubjectFull: Antisocial personality disorders
        Type: general
      – SubjectFull: Algorithms
        Type: general
      – SubjectFull: Psychosocial factors
        Type: general
      – SubjectFull: Adolescence
        Type: general
      – SubjectFull: England
        Type: general
    Titles:
      – TitleFull: Predicting criminal offence in adolescents who exhibit antisocial behaviour: a machine learning study using data from a large randomised controlled trial of multisystemic therapy.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Suh, Jae Won
      – PersonEntity:
          Name:
            NameFull: Saunders, Rob
      – PersonEntity:
          Name:
            NameFull: Simes, Elizabeth
      – PersonEntity:
          Name:
            NameFull: Delamain, Henry
      – PersonEntity:
          Name:
            NameFull: Butler, Stephen
      – PersonEntity:
          Name:
            NameFull: Cottrell, David
      – PersonEntity:
          Name:
            NameFull: Kraam, Abdullah
      – PersonEntity:
          Name:
            NameFull: Scott, Stephen
      – PersonEntity:
          Name:
            NameFull: Goodyer, Ian M
      – PersonEntity:
          Name:
            NameFull: Wason, James
      – PersonEntity:
          Name:
            NameFull: Pilling, Stephen
      – PersonEntity:
          Name:
            NameFull: Fonagy, Peter
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 05
              Text: May2025
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 10188827
          Numbering:
            – Type: volume
              Value: 34
            – Type: issue
              Value: 5
          Titles:
            – TitleFull: European Child & Adolescent Psychiatry
              Type: main
ResultId 1