International Comparisons of the Youth Self-Report Dysregulation Profile: Latent Class Analyses in 34 Societies

J Am Acad Child Adolesc Psychiatry. 2016 Dec;55(12):1046-1053. doi: 10.1016/j.jaac.2016.08.012. Epub 2016 Oct 5.

Abstract

Objective: We used latent class analysis (LCA) to examine the prevalence and characteristics of the Dysregulation Profile (DP) based on data from the Youth Self-Report (YSR). The DP comprises elevated scores on the Anxious/Depressed, Attention Problems, and Aggressive Behavior syndromes and thus reflects significant problems in self-regulation of affect, attention, and behavior.

Method: We examined YSR data for 38,070 adolescents (48.1% male) in 34 societies. Participants ranged in age from 11 to 16 years. Researchers in 31 societies used translations of the YSR (not in Jamaica, Australia, or the United States).

Results: The various statistical indices for good LCA model fit (entropy, bootstrapped parametric likelihood ratio test, adjusted Bayesian Information Criterion, and probability of correct assignment) were not always consistent but generally supported a DP class in every society. However, prevalence of the DP ranged from 1% to 26% and the T score syndrome profile for the DP class in many societies featured elevations on all scales. In every society, the DP class had significantly higher scores than the pooled non-DP classes on all 3 DP syndromes, with large d values.

Conclusion: Because model fit, the number of classes, and the prevalence of the DP class varied across societies, and because the DP "3-peak" profile was relatively uncommon, results for the DP based on adolescents' ratings in 34 societies must be considered as mixed.

Keywords: Dysregulation Profile; YSR; international comparisons; latent class analysis.

Publication types

  • Comparative Study

MeSH terms

  • Adolescent
  • Child
  • Child Behavior Disorders* / classification
  • Child Behavior Disorders* / epidemiology
  • Child Behavior Disorders* / physiopathology
  • Female
  • Global Health / statistics & numerical data*
  • Humans
  • Male
  • Models, Statistical*
  • Self Report