Using data collected in the National Institute of Mental Health (Rockville, Md) Epidemiologic Catchment Area Program, we examined the proposed hypothesis that there has been a shift in major depression to younger ages at onset, or increased prevalence in younger age periods, for recent birth cohorts. Life-table survival methods were used to examine the hazard rates for major depression as well as for other specific mental disorders. The findings are consistent with a gradual shift to increased rates for major depression between the ages of 15 and 19 years for Epidemiologic Catchment Area respondents born more recently. The findings also suggest a similar shift for drug abuse/dependence; similar but less pronounced changes were found for alcohol abuse/dependence and obsessive-compulsive disorder. However, in this study, bipolar disorder, panic disorder, and phobias did not exhibit a consistent increase in onset at younger ages. Further research is required to determine if the shifts in major depression, drug abuse/dependence, and possibly alcohol abuse/dependence are linked. It is important to note that these shifts to adolescent onset are occurring when nearly half the 31 million Americans without health insurance are aged 24 years or younger.