Methodological challenges in assessing children's mental health services utilization.

This paper describes the reliability and validity of the service assessment for children and adolescents (SACA) for use among Spanish-speaking respondents. The test-retest reliability of the instrument was assessed in a randomly selected clinical sample of 146 Puerto Rican children and adolescents aged 4-17. Both parents and children were administered the SACA twice by independent interviewers over an average 12-day follow-up period. The accuracy of parental and youth self-reports was assessed by comparing these reports to information obtained from medical records. The results showed that parents and children (aged 11-17) were able to report with fair to moderate reliability any last year use of mental health services, any outpatient mental health services, and school services. Residential and hospitalization services were reported by both informants with substantial test-retest reliability. Slight or no test-retest reliability was obtained for parent and child on the use of the specific type of mental health professionals, as well as parental reports of several treatment modalities. Substantial sensitivity of the SACA was obtained when comparing medical records to parental and child reports to lifetime use of any service and outpatient mental health service. Moderate sensitivity was obtained for last year use of mental health services for both parent and child informants.