BACKGROUND: Adolescent nonmedical prescription opioid use is associated with overdose and other adverse outcomes, but its risk factors are poorly understood.METHODS: Data were drawn from a prospective cohort study of Los Angeles, California, high school students. At baseline (mean age 5 14.6 years), students completed self-report screening measures of problem alcohol, cannabis, and drug use and 6 mental health problems (major depression, generalized anxiety, panic disorder, social phobia, obsessive-compulsive disorder, and hypomania or mania). Past 6-month nonmedical prescription opioid use (yes or no) was assessed across 7 semiannual follow-ups.RESULTS: Among baseline never users of nonmedical prescription opioids (N 5 3204), average past 6-month prevalence of new nonmedical prescription opioid use across the 42-month follow-up was 4.4% (range 3.5%-6.1%). In a multivariable model co-adjusting for 9 baseline behavioral problems and other factors, major depression, hypomania or mania, cannabis, alcohol, and other drug use problems were associated with increased odds of nonmedical prescription opioid use over follow-ups. Cumulative indices of behavioral health comorbidity showed successively greater odds of subsequent nonmedical prescription opioid use for students with 1