Dataset: the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) with a sample size of 43,093.

Procedure: Exploratory and confirmatory factor analyses were to examine the latent structure of non-medical prescription drug use and prescription drug use disorders on the full sample. The non-medical prescription drugs are sedatives, opiates, tranquilizers, and stimulants. One-factor model was captured by EFA and confirmed by CFA, which indicates a liability shared among these drug use and use disorders.

Multiple Indicators Multiple Causes (MIMIC) analysis was used to examine whether the effect of sociodemographic and psychiatric covariates occurred through the latent factor, directly on each drug class or both. It has been found that younger age, being White, having more intense pain or psychiatric conditions increased the risk of non-medical prescription drug use through the latent factor. The same covariates, except for anxiety disorders also significantly increased the risk of prescription drug use disorders through the latent factor. In addition, young age directly increased the risk of sedative use above and beyond the latent factor.

Conclusion: Treatment, prevention and policy approaches associated with these identified risk factors working on these drug use problems as a group maybe more effective than those focused on individual classes of drugs.