CHILDREN FROM THE GAZEL COHORT .1. PREVAL ENCE AND PATTERNS OF SERVICE UTILIZATION FOR EMOTIONAL AND BEHAVIORAL SYMPTOMS

Art epidemiological survey of French children aged 4 to 16 was conduct ed in order to estimate the 12-months rates of service utilization for psychological reasons and to assess the factors associated with servi ce use in this community sample. A large sample of 2582 children and a dolescents was recruited from the families whose one parent was employ ed by the national electricity and gaz company (EDF-GDF). Of these emp loyees, 20,000 have volunteered for a longterm prospective cohort stud y of their health and, since 1989, they have participated to annual su rveys and additional ad hoc research programmes. Families with a child aged 4 To 16 in 1991 were selected. Only one child was selected in ea ch family, and the sample was stratified by socio-economic status and family size according to census data. A survey questionnaire comprisin g a valid measure of child psychopathology (Child Behavior Checklist: CBCL) and an additional questionnaire including questions related to s ervice use was used as a means of data collection. The response rate w as 62.2 % and factors associated with participation in the survey were analysed. The 12-months prevalence rate of contact for psychological motives were. 42.3 % for general practitioners and family doctors, 7.8 % for speech and language therapists, 9.5 % for educational specialist s, and 6.0% for mental health professionals. With the exception of gen eral practitioners, rates of service contact were significantly higher for boys. Logistic regression analysis was used to identify separatel y factors associated with recent contact for each category of professi onals. Results showed that, for all professionals, high scores on the CBCL measure was significantly associated with service use, the strong est association being found for mental health professionals. Family st ructure was also predictive of the latter, with higher rates of contac ts for those children living in families whose parents are divorced, s eparated or widowed. Some differences for contacts with doctors were f ound according to the region, otherwise, no effects of socio-economic status, educational level of the parent, or other socioeconomic indica tors were found to predict service utilization. The implications for s ervices are discussed.