Correlates of prescription drug nonmedical use and problem use by adolescents

Objectives: This study characterizes patterns and correlates of prescription drug nonmedical (PDNM) use and prescription drug problem (PDP) use among U.S. adolescents. Methods: Using the 2003 National Survey on Drug Use and Health (NSDUH), we estimated annual prevalence of PDNM and PDP use. We stratified prevalence estimates by therapeutic class and key patterns of use (prescription drugs alone [Rx-only] and with other drugs [poly-drug]). We used weighted multivariate logistic regression to estimate odds ratios (OR; 95% confidence interval [CI]) of correlates associated with PDNM and PDP use, adjusting for complex sampling design. Results: Nearly 2.3 million adolescents (9.3%) reported PDNM use; of these, 357,000 (15.3%) met PDP use criteria. Nearly 60% of PDNM users and 70% of PDP users also reported past-year use of illicit drugs, predominantly marijuana, hallucinogens, inhalants, and cocaine/crack. Among PDP users, almost 60% reported use of 2 or more prescription drug classes. Multivariable analyses revealed variation in the associations of correlates, especially sex and race/ethnicity, with PDNM and PDP use. Correlates also differed across therapeutic classes and by patterns of use. For example, females were more likely to report Rx-only PDNM use (OR = 1.27, 95% CI = 1.05, 1.54) and Rx-only PDP use (OR = 1.62, 95% CI = 1. 15, 2.3); sex differences disappeared with poly-drug use measures. Hispanic ethnicity was strongly associated with all PDP use measures (OR = 2.13, 95% CI = 1. 16, 3.9 1). Conclusions: Correlates differed for PDNM and PDP use, in aggregate and by therapeutic class, and adolescents' preferences for prescription drugs as single agents or with other drugs use. These findings provide knowledge that can be used to better target approaches for identifying and preventing problematic prescription drug use.