Predicting prognosis for the conduct-problem boy: can family history help?

Journal Article

OBJECTIVE: Many children with conduct disorder develop life-course persistent antisocial behavior; however, other children exhibit childhood-limited or adolescence-limited conduct disorder symptoms and escape poor adult outcomes. Prospective prediction of long-term prognosis in pediatric and adolescent clinical settings is difficult. Improved prognosis prediction would support wise allocation of limited treatment resources. The purpose of this article is to evaluate whether family history of psychiatric disorder can statically predict long-term prognosis among conduct-problem children. METHOD: Participants were male members of the Dunedin Study, a birth cohort of 1,037 children (52% male). Conduct-problem subtypes were defined using prospective assessments between ages 7 and 26 years. Family history interviews assessed mental disorders for three generations: the participants' grandparents, parents, and siblings. RESULTS: Family history of externalizing disorders distinguished life-course persistent antisocial males from other conduct-problem children and added significant incremental validity beyond family and child risk factors. A simple three-item family history screen of maternal-reported alcohol abuse was associated with life-course persistent prognosis in our research setting and should be evaluated in clinical practice. CONCLUSIONS: : Family history of externalizing disorders distinguished between life-course persistent versus childhood-limited and adolescent-onset conduct problems. Brief family history questions may assist clinicians in pediatric settings to refine the diagnosis of conduct disorder and identify children who most need treatment.

Full Text

Duke Authors

Cited Authors

  • Odgers, CL; Milne, BJ; Caspi, A; Crump, R; Poulton, R; Moffitt, TE

Published Date

  • October 2007

Published In

Volume / Issue

  • 46 / 10

Start / End Page

  • 1240 - 1249

PubMed ID

  • 17885565

International Standard Serial Number (ISSN)

  • 0890-8567

Digital Object Identifier (DOI)

  • 10.1097/chi.0b013e31813c6c8d

Language

  • eng

Conference Location

  • United States