Using CBCL Scores At Age 5 to Predict Psychiatric Diagnoses 1.5 Years Later in the Netherlands

Dutch researchers administered a translation of the DICA-R interview to the parents of 403 children 1.5 years after the children had been assessed with the CBCL/4-18 at age 5. Three groups were selected from a school-based cohort of 1,317 children screened at age 5: Group E–CBCL Externalizing score >90th percentile and/or Attention Problems score >95th percentile (N = 175); Group I–CBCL Internalizing score>90th percentile and did not meet criteria for Group E (N = 59); and Group N–from the children who had CBCL Total Problems score <90th percentile, children were selected for matching to a child in Group E or Group I on age, sex, and school (N = 220).

CBCL scores and psychosocial risk factors were used to predict five groups of DSM diagnoses: ADHD, CD/ODD, mood disorders, anxiety disorders, and elimination disorders. As measured by odds ratios (OR), ADHD was strongly predicted by clinical range scores on Attention Problems (OR = 23), Aggressive Behavior (OR = 16), Externalizing (OR = 126), and Total Problems (OR = 34). CD/ODD diagnoses were strongly predicted by Aggressive Behavior (OR = 39), Externalizing (OR = 28), and Total Problems (OR = 15). Mood disorders were predicted by clinical range scores on Anxious/Depressed, Social Problems, and Thought Problems (OR = 6, 7, 9, respectively), Internalizing (OR = 4), and Total Problems (OR = 3). Anxiety diagnoses were predicted by clinical range scores on Withdrawn (OR = 8), Sex Problems (OR = 8), Internalizing (OR = 6), and Total Problems (OR=3). Elimination disorders had the lowest OR (all OR < 4). OR for psychosocial risk factors, such as low parent SES, single parent family, and negative life events, were significant but generally lower than those for CBCL scores (all OR < 5).

The authors concluded that CBCL scores at age 5 were good predictors of psychiatric diagnoses 1.5 years later in a community sample and that they yielded higher OR than various psychosocial risk factors.

Reference: Kroes, M., Kalff, A.C., Steyaert, J., Kessels, A.G.H., Feron, F.J.M., Hendriksen, J.G.M., vanZeben, T.M.C.B., Troost, J., Jolles, J., & Vles, J.S.H. (2002). A longitudinal community study: Do psychosocial risk factors and Child Behavior Checklist scores at 5 years of age predict psychiatric diagnoses at a later age? Journal of American Academy of Child and Adolescent Psychiatry, 41, 955-963.