Latent classes of emotional and behavioural problems in epidemiological and referred samples and their relations to DSM-IV diagnoses

Valentina Bianchi, Paolo Brambilla, Marco Garzitto, Paola Colombo, Livia Fornasari, Monica Bellina, Carolina Bonivento, Alessandra Tesei, Sara Piccin, Stefania Conte, Giampaolo Perna, Alessandra Frigerio, Isabella Castiglioni, Franco Fabbro, Massimo Molteni, Maria Nobile

Research output: Contribution to journalArticle

Abstract

Researchers’ interest have recently moved toward the identification of recurrent psychopathological profiles characterized by concurrent elevations on different behavioural and emotional traits. This new strategy turned to be useful in terms of diagnosis and outcome prediction. We used a person-centred statistical approach to examine whether different groups could be identified in a referred sample and in a general-population sample of children and adolescents, and we investigated their relation to DSM-IV diagnoses. A latent class analysis (LCA) was performed on the Child Behaviour Checklist (CBCL) syndrome scales of the referred sample (N = 1225), of the general-population sample (N = 3418), and of the total sample. Models estimating 1-class through 5-class solutions were compared and agreement in the classification of subjects was evaluated. Chi square analyses, a logistic regression, and a multinomial logistic regression analysis were used to investigate the relations between classes and diagnoses. In the two samples and in the total sample, the best-fitting models were 4-class solutions. The identified classes were Internalizing Problems (15.68%), Severe Dysregulated (7.82%), Attention/Hyperactivity (10.19%), and Low Problems (66.32%). Subsequent analyses indicated a significant relationship between diagnoses and classes as well as a main association between the severe dysregulated class and comorbidity. Our data suggested the presence of four different psychopathological profiles related to different outcomes in terms of psychopathological diagnoses. In particular, our results underline the presence of a profile characterized by severe emotional and behavioural dysregulation that is mostly associated with the presence of multiple diagnosis.

Original languageEnglish
Pages (from-to)549-557
Number of pages9
JournalEuropean Child and Adolescent Psychiatry
Volume26
Issue number5
DOIs
Publication statusPublished - 2017

Fingerprint

Diagnostic and Statistical Manual of Mental Disorders
Logistic Models
Child Behavior
Checklist
Population
Problem Behavior
Emotion
Comorbidity
Regression Analysis
Research Personnel

Keywords

  • Adolescence
  • Child Behaviour Checklist
  • Childhood
  • Dysregulation
  • Latent class analysis (LCA)
  • Psychopathology

ASJC Scopus subject areas

  • Pediatrics, Perinatology, and Child Health
  • Developmental and Educational Psychology
  • Philosophy
  • Psychiatry and Mental health

Cite this

Latent classes of emotional and behavioural problems in epidemiological and referred samples and their relations to DSM-IV diagnoses. / Bianchi, Valentina; Brambilla, Paolo; Garzitto, Marco; Colombo, Paola; Fornasari, Livia; Bellina, Monica; Bonivento, Carolina; Tesei, Alessandra; Piccin, Sara; Conte, Stefania; Perna, Giampaolo; Frigerio, Alessandra; Castiglioni, Isabella; Fabbro, Franco; Molteni, Massimo; Nobile, Maria.

In: European Child and Adolescent Psychiatry, Vol. 26, No. 5, 2017, p. 549-557.

Research output: Contribution to journalArticle

Bianchi, V, Brambilla, P, Garzitto, M, Colombo, P, Fornasari, L, Bellina, M, Bonivento, C, Tesei, A, Piccin, S, Conte, S, Perna, G, Frigerio, A, Castiglioni, I, Fabbro, F, Molteni, M & Nobile, M 2017, 'Latent classes of emotional and behavioural problems in epidemiological and referred samples and their relations to DSM-IV diagnoses', European Child and Adolescent Psychiatry, vol. 26, no. 5, pp. 549-557. https://doi.org/10.1007/s00787-016-0918-2
Bianchi, Valentina ; Brambilla, Paolo ; Garzitto, Marco ; Colombo, Paola ; Fornasari, Livia ; Bellina, Monica ; Bonivento, Carolina ; Tesei, Alessandra ; Piccin, Sara ; Conte, Stefania ; Perna, Giampaolo ; Frigerio, Alessandra ; Castiglioni, Isabella ; Fabbro, Franco ; Molteni, Massimo ; Nobile, Maria. / Latent classes of emotional and behavioural problems in epidemiological and referred samples and their relations to DSM-IV diagnoses. In: European Child and Adolescent Psychiatry. 2017 ; Vol. 26, No. 5. pp. 549-557.
@article{f7d58eb6d56043c396e2a33fbf58a578,
title = "Latent classes of emotional and behavioural problems in epidemiological and referred samples and their relations to DSM-IV diagnoses",
abstract = "Researchers’ interest have recently moved toward the identification of recurrent psychopathological profiles characterized by concurrent elevations on different behavioural and emotional traits. This new strategy turned to be useful in terms of diagnosis and outcome prediction. We used a person-centred statistical approach to examine whether different groups could be identified in a referred sample and in a general-population sample of children and adolescents, and we investigated their relation to DSM-IV diagnoses. A latent class analysis (LCA) was performed on the Child Behaviour Checklist (CBCL) syndrome scales of the referred sample (N = 1225), of the general-population sample (N = 3418), and of the total sample. Models estimating 1-class through 5-class solutions were compared and agreement in the classification of subjects was evaluated. Chi square analyses, a logistic regression, and a multinomial logistic regression analysis were used to investigate the relations between classes and diagnoses. In the two samples and in the total sample, the best-fitting models were 4-class solutions. The identified classes were Internalizing Problems (15.68{\%}), Severe Dysregulated (7.82{\%}), Attention/Hyperactivity (10.19{\%}), and Low Problems (66.32{\%}). Subsequent analyses indicated a significant relationship between diagnoses and classes as well as a main association between the severe dysregulated class and comorbidity. Our data suggested the presence of four different psychopathological profiles related to different outcomes in terms of psychopathological diagnoses. In particular, our results underline the presence of a profile characterized by severe emotional and behavioural dysregulation that is mostly associated with the presence of multiple diagnosis.",
keywords = "Adolescence, Child Behaviour Checklist, Childhood, Dysregulation, Latent class analysis (LCA), Psychopathology",
author = "Valentina Bianchi and Paolo Brambilla and Marco Garzitto and Paola Colombo and Livia Fornasari and Monica Bellina and Carolina Bonivento and Alessandra Tesei and Sara Piccin and Stefania Conte and Giampaolo Perna and Alessandra Frigerio and Isabella Castiglioni and Franco Fabbro and Massimo Molteni and Maria Nobile",
year = "2017",
doi = "10.1007/s00787-016-0918-2",
language = "English",
volume = "26",
pages = "549--557",
journal = "European Child and Adolescent Psychiatry",
issn = "1018-8827",
publisher = "Dr. Dietrich Steinkopff Verlag GmbH and Co. KG",
number = "5",

}

TY - JOUR

T1 - Latent classes of emotional and behavioural problems in epidemiological and referred samples and their relations to DSM-IV diagnoses

AU - Bianchi, Valentina

AU - Brambilla, Paolo

AU - Garzitto, Marco

AU - Colombo, Paola

AU - Fornasari, Livia

AU - Bellina, Monica

AU - Bonivento, Carolina

AU - Tesei, Alessandra

AU - Piccin, Sara

AU - Conte, Stefania

AU - Perna, Giampaolo

AU - Frigerio, Alessandra

AU - Castiglioni, Isabella

AU - Fabbro, Franco

AU - Molteni, Massimo

AU - Nobile, Maria

PY - 2017

Y1 - 2017

N2 - Researchers’ interest have recently moved toward the identification of recurrent psychopathological profiles characterized by concurrent elevations on different behavioural and emotional traits. This new strategy turned to be useful in terms of diagnosis and outcome prediction. We used a person-centred statistical approach to examine whether different groups could be identified in a referred sample and in a general-population sample of children and adolescents, and we investigated their relation to DSM-IV diagnoses. A latent class analysis (LCA) was performed on the Child Behaviour Checklist (CBCL) syndrome scales of the referred sample (N = 1225), of the general-population sample (N = 3418), and of the total sample. Models estimating 1-class through 5-class solutions were compared and agreement in the classification of subjects was evaluated. Chi square analyses, a logistic regression, and a multinomial logistic regression analysis were used to investigate the relations between classes and diagnoses. In the two samples and in the total sample, the best-fitting models were 4-class solutions. The identified classes were Internalizing Problems (15.68%), Severe Dysregulated (7.82%), Attention/Hyperactivity (10.19%), and Low Problems (66.32%). Subsequent analyses indicated a significant relationship between diagnoses and classes as well as a main association between the severe dysregulated class and comorbidity. Our data suggested the presence of four different psychopathological profiles related to different outcomes in terms of psychopathological diagnoses. In particular, our results underline the presence of a profile characterized by severe emotional and behavioural dysregulation that is mostly associated with the presence of multiple diagnosis.

AB - Researchers’ interest have recently moved toward the identification of recurrent psychopathological profiles characterized by concurrent elevations on different behavioural and emotional traits. This new strategy turned to be useful in terms of diagnosis and outcome prediction. We used a person-centred statistical approach to examine whether different groups could be identified in a referred sample and in a general-population sample of children and adolescents, and we investigated their relation to DSM-IV diagnoses. A latent class analysis (LCA) was performed on the Child Behaviour Checklist (CBCL) syndrome scales of the referred sample (N = 1225), of the general-population sample (N = 3418), and of the total sample. Models estimating 1-class through 5-class solutions were compared and agreement in the classification of subjects was evaluated. Chi square analyses, a logistic regression, and a multinomial logistic regression analysis were used to investigate the relations between classes and diagnoses. In the two samples and in the total sample, the best-fitting models were 4-class solutions. The identified classes were Internalizing Problems (15.68%), Severe Dysregulated (7.82%), Attention/Hyperactivity (10.19%), and Low Problems (66.32%). Subsequent analyses indicated a significant relationship between diagnoses and classes as well as a main association between the severe dysregulated class and comorbidity. Our data suggested the presence of four different psychopathological profiles related to different outcomes in terms of psychopathological diagnoses. In particular, our results underline the presence of a profile characterized by severe emotional and behavioural dysregulation that is mostly associated with the presence of multiple diagnosis.

KW - Adolescence

KW - Child Behaviour Checklist

KW - Childhood

KW - Dysregulation

KW - Latent class analysis (LCA)

KW - Psychopathology

UR - http://www.scopus.com/inward/record.url?scp=84995395678&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84995395678&partnerID=8YFLogxK

U2 - 10.1007/s00787-016-0918-2

DO - 10.1007/s00787-016-0918-2

M3 - Article

AN - SCOPUS:84995395678

VL - 26

SP - 549

EP - 557

JO - European Child and Adolescent Psychiatry

JF - European Child and Adolescent Psychiatry

SN - 1018-8827

IS - 5

ER -