This study focuses on the management of psychosocial problems by pediatricians and pediatric nurse practitioners in an HMO, the Columbia Medical Plan. The proposed research has three aims: 1) To examine the impact of sharing the results of a psychological screening questionnaire (The Child Behavior Checklist, CBCL) with pediatric providers on their management of psychosocial problems in children, 2) To examine the impact of this feedback on the psychosocial outcome of the child, 3) To design interventions, based on the findings, that will improve the management of children with psychosocial problems. The sample will consist of all children aged 6-10 visiting the Department of Pediatrics during one year. At the index visit, the parent will complete the CBCL and provide information on sociodemographic characteristics and attitudes toward psychiatry. At the same visit, without knowing the results of the CBCL, pediatric providers will assess the child using a multi-axial scheme on the presence of physical, mental disorder, psychosocial, and social problems. The children will then be randomized into one of two conditions: the provider is told the results of the CBCL or the provider is not told the results. The children will be followed over a six month interval after the index visit. All children with positive scores on the CBCL or a provider diagnosis of mental disorder or psychosocial problems will be asked to repeat the CBCL. A one-in-three sample of the children who are """"""""normal"""""""" will also be asked to repeat the CBCL. Data will be obtained from the providers and the computerized information system to ascertain the type of treatment received by the patient during the six month follow-up period. Following the follow-up study, providers will be debriefed to understand the usefulness of the CBCL. The data analysis will test these two hypotheses, which are stated in the null form: Hypothesis 1: Feedback of the results of the CBCL will not affect the management of the child by the pediatric provider, specifically whether or not the child is seen in psychiatry. Hypothesis 2: Feedback of the results of the CBCL will not affect the child's outcome at follow-up, as measured by the score on the CBCL administered six months after the initial visit. Mixed logic and multiple regression techniques will be used. The analysis will control for child's, parents', and provider's characteristics. The results will be used to design new treatment programs conjointly with CMP staff.
