Global measures of impairment for epidemiologic and clinical use with children and adolescents

Objective: Two global measures of impairment developed for the NIMH Methods for the Epidemiology of Child and Adolescent Mental Disorders (MECA) Study and their potential for epidemiologic and clinical use are assessed. The measures are a non-clinician version of the Children's Global Assessment Scale (CGAS), designed to be scored by parents or by lay interviewers, and the Columbia Impairment Scale (CIS), a 13-item scale that can be administered directly by lay or clinical interviewers to parents or youth. Method: The validity of these impairment measures is evaluated by examining their correlations with clinician CGAS scores and with other factors expected to be associated with impairment. Discriminant function analyses are also conducted to assess thresholds on each measure. Results: All versions of the CIS and the CGAS show good variability in the community and correlate significantly with other indicators of psychological dysfunction and with the clinician's CGAS score. There is evidence of good construct, discriminant, and concurrent validity. Inter-site differences are assessed and appear to be culturally determined. Suggested thresholds for each measure are provided. Conclusions: The findings support the construct and concurrent validity of these measures of impairment. The ratings provided by lay interviewers based on parent interviews, relate better to ratings provided by clinicians, than do ratings based on youth interviews or provided by parents themselves. The CIS has an advantage over the non-clinician CGAS in that its rating does not rely on clinical judgment. Both measures are useful for clinical as well as epidemiologic purposes. The findings support the integration of both symptomatology and functional impairment in child psychiatric diagnostic assessments. Limitations to these findings are discussed.