The National Comorbidity Survey–Adolescent Supplement (NCS-A) was conducted between 2001 and 2004. Adolescents were selected from households participating in the National Comorbidity Survey Replication (NCS-R) at a response rate of 86.8% (879 adolescents). An additional sample was derived from schools at a response rate of 82.6% (9,244 adolescents). The total NCS-A sample was 10,123 English-speaking adolescents, ages 13-18 years old. Data were weighted for variability in probability of selection and differential non-response based on population sociodemographic distributions. Merged weighted household and school samples were adjusted for design effects, and closely approximate the sociodemographic distribution of the 2000 US census population.42-44 Additional details about the survey design and sampling procedures are provided elsewhere.43-45 For the current study, we excluded any participant who was missing a response for any of the PTEs or survey stratification variables, creating a final analytic sample of 9,956 adolescents (98.4%). Written informed consent from parents and assent from adolescents were obtained from all participants. Study procedures were approved by Human Subjects Committees at Harvard Medical School and the University of Michigan. The current study was approved by the Human Subjects Committee at Columbia University.
