This proposal seeks funding to continue to test two competing hypotheses about the etiology of problem drug use: H1 : Early onset drug users (before age 16) are at increased risk of later problem drug use, relative to later onset drug users; H2: Early-onset drug users are at a significantly increased risk of problem drug use, compared with later-onset users, only if they have comorbid psychiatric problems in childhood. The proposal responds to NIDA's program announcement DA-91-33, """"""""Vulnerability to drug abuse"""""""". The proposal capitalizes on the Great Smoky Mountains Study (GSMS), a longitudinal research program on childhood predictors of alcohol, drug, and mental (ADM) disorders and comorbidity. In 1992-93 a representative population sample (N = 1,422) of 9, 11, and 13-year-olds, stratified on parent-reported behavioral and drug problems, was selected from a mixed rural-urban area of the southeastern United States. American Indians (mainly Cherokee) were over sampled so that all age-appropriate youth (N=349) were recruited. Response rate was 80% for both groups. Participants ( 50% female) and parents/caretakers have been interviewed annually for 4 years, using a highly-structured interview for ADM disorders. Data include use and abuse of tobacco, alcohol, marijuana and hard drugs, psychopathology and DSM-IV diagnoses, level of functioning at school, at home, and with peers, need for and use of services for ADM disorders, risk and protective factors including physical health and development, family functioning and family history of ADM problems and treatment, and community resources. Analyses to date show early-onset drug users (N=357) to have high rates of emotional (girls) and behavioral disorders, and a family history of drug problems. Cherokee youth have started drug use earlier and escalated faster than others. In the next 5 years we propose to reinterview each cohort at age 15 and 18, and the oldest cohort at age 21, keeping contact between interviews using computer-assisted telephone interviews. Tests of the hypotheses will use (weighted) random regression modeling to capitalize on the strengths of the longitudinal data. This is the first study capable of testing these competing hypotheses prospectively using state of-the-art assessment of ADM disorders, in a representative population sample of both sexes, with two ethnic groups to test generalizability. Support for the first hypothesis makes the case for primary (universal) prevention programs: support for the second mandates interventions targeted at high-risk groups.
Showing the most recent 10 out of 74 publications
