The overall goal is to investigate the relation between ethnicity, mental health, and medical care among adolescents. The major objective is to estimate incidence and prevalence of depression and anxiety among Anglo, African, and Hispanic American adolescents being seen in primary medical care clinics and to assess the impact of psychological dysfunction on the type and volume of services received. The descriptive aims are to: (1) Assess phenomenology of depression and anxiety across ethnic groups; (2) assess comorbidity of depression, anxiety, and somatic illness across ethnic groups; (3) estimate incidence and prevalence of depression and anxiety across ethnic groups; (4) determine whether adolescents with depression and anxiety are more frequent users of general medical services, and for what types of services; (5) determine the rate at which primary care providers recognize, label and treat mental health problems among adolescents; and (6) determine the rate at which primary care physicians refer adolescents with mental health problems to specialty mental health services. The methodologic aims are to: Assess concordance between physician diagnosis of depression and anxiety and diagnosis made independently using a structured diagnostic research interview, (2) assess the operating characteristics of screening scales for anxiety and depression across ethnic groups, and (3) assess concordance between parent and child reports of psychopathology using a structured psychiatric interview, and correlates of disagreement. In addition to examining the role of ethnic status, we also will assess the impact of ethnic identify, response styles, and selected psychosocial risk factors drawn from two main domains - stressors and resources. The analytic objective is to examine the efficacy of alternative models of stress and mental health, incorporating multiple dimensions of ethnicity into these models. A panel of youths 10-17 years of age will be sampled from Harris County Hospital District Health Centers over a 12 month period, and followed for 12 months. The T/1 sample will be 3,150, about 1,050 in each adolescent ethnic group. The parent sample will be about 1,380, also equally distributed across ethic groups, sampled on the basic of the diagnostic status of the adolescents. Data will be collected using self-report questionnaires, structured psychiatric interviews, physician report forms, and medical records. To examine the influence of risk factors, multivariate procedures such as logistic regression analysis, random regression, linear-models, GSK models, hierarchical discriminant function analysis, and linear structural equation models will be employed. Differential manifestation of symptoms will be examined using multivariate classification procedures (cluster analysis, factor analysis, latent trait analysis) to analyze symptom patterns within and across diagnostic groups as a function of ethnic status and selected psychosocial risk factors. Screening scales will be assessed for reliability, dimensionality, and efficacy in detecting cases of diagnosed disorder using, for example, Receiver Operating Characteristic (ROC) analyses, to estimate optimal screening scores in different subgroups.
Showing the most recent 10 out of 20 publications
