Patterns of reduced use and abstinence in multi-site randomized controlled trials of pharmacotherapies for cocaine and methamphetamine use disorders.

Abstract

Many individuals with cocaine or methamphetamine use disorders who enter treatment do not achieve abstinence but reduce their use of the target drug. We aimed to compare change in pattern of drug use (i.e., achieving "abstinence", "reduced use" or no reduction in use) among participants in randomized controlled trials (RCTs) of treatment of cocaine and methamphetamine use disorder, irrespective of the type of treatment. The data were drawn from 10 multi-site pharmacotherapy RCTs of cocaine (n = 1,134) and methamphetamine (n = 555) use disorders. The outcome patterns and their sociodemographic and clinical correlates were compared in cocaine and methamphetamine RCTs, using multinomial logistic regression models. Analyses were adjusted for missing data, clustering within RCTs, socio-demographic and baseline clinical characteristics, and treatment arms. Those in cocaine RCTs were more likely to experience reduced use compared to participants in methamphetamine RCTs (20.6% vs. 13.2%, respectively), but less likely to experience "abstinence" (7.6% vs. 20.3%; Chi-squared = 14.20, df = 2, P < 0.001). Differences in "abstinence" persisted after adjustment for baseline covariates. Association of sociodemographic and clinical correlates with outcomes differed in cocaine and methamphetamine RCTs. A sizeable proportion of individuals in RCTs of pharmacological treatment for stimulant use disorder who do not attain "abstinence" nevertheless reduce their use. The outcome patterns of drug use are different for cocaine and methamphetamine use disorders and reliance on abstinence as the sole outcome may obscure these differences.