Identification of Childhood Psychiatric Disorder by Informant: Comparisons of Clinic and Community Samples
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
Objective: To compare the identification of psychiatric disorder as informed by parents versus teachers in children aged 6–11 years and parents versus adolescents in youth aged 12–16 years in clinic versus community samples. Method: Study data come from parallel surveys in Hamilton, Ontario, of children aged 6–16 years. The surveys included consecutive referrals (N = 1150) between 1989 and 1991 to the region's 2 agencies providing outpatient child mental health services. Also, a simple random sample (N = 1689) was used, drawn in 1989 from students attending public schools. Conduct disorder, hyperactivity, emotional disorder, and somatization disorder were assessed by informants using the original Ontario Child Health Study scales. Results: The percentage of children identified with a disorder was markedly higher in the clinic sample, irrespective of the type of disorder, the age and sex of the child, and who provided the assessment. Also, there was a statistically significant differential shift between parents and teachers in the percentage of children identified with disorder. The ratio of children aged 6–11 years identified with conduct disorder or hyperactivity by parents versus teachers was higher in the clinic sample than in the community sample. Among youth aged 12–16 years, a similar pattern emerged for parents as informants versus the adolescents themselves, but it was statistically nonsignificant. Conclusions: The data suggest that the relative contribution of informants to the identification of childhood psychiatric disorder varies by sample type: clinic and community. If risk factors for child disorder are influenced by contextually specific factors wedded to informants, then studies conducted in clinic versus community samples may lead to discrepant information about the determinants of psychopathology. The extent of this problem needs to be assessed by comparing the results of parallel studies conducted in clinic versus community samples.
