• Linwei Wang, Jeong Eun Min, Emanuel Krebs, Elizabeth Evans, David Huang, Lei Liu, Yih-Ing Hser, and Bohdan Nosyk.
• BC Centre for Excellence in HIV/AIDS, 608-1081 Burrard Street, Vancouver, BC V6Z 1Y6, Canada. Electronic address: lwang@cfenet.ubc.ca.
• Int. J. Drug Policy. 2017 Nov 1; 49: 32-40.

BackgroundPolydrug use may challenge effective treatment for substance use disorders. We evaluate whether secondary substance use modifies the association between treatment and primary drug use among primary heroin, cocaine and methamphetamine (MA) users.MethodsData were obtained from prospective cohort studies on people who use illicit drugs (PWUD) in California, USA. Using repeated monthly data on self-reported secondary substance use (heroin, cocaine, MA, alcohol or marijuana; ≥1day in a month), primary drug use (≥1day in a month), and treatment participation, collected via timeline follow-back, we fitted generalized linear mixed multiple regression models controlling for potential confounders to examine the interactions between treatment and secondary substance use on the odds of primary heroin, cocaine and MA use, respectively.ResultsIncluded in our study were 587 primary heroin, 444 primary MA, and 501 primary cocaine users, with a median of 32.4, 13.3 and 18.9 years of follow-up, respectively. In the absence of secondary substance use, treatment was strongly associated with decreased odds of primary drug use (adjusted odds ratios (aORs): 0.25, 95% CI: 0.24, 0.27, 0.07 (0.06, 0.08), and 0.07 (0.07, 0.09)) for primary heroin, MA, and cocaine users, respectively. Secondary substance use of any kind moderated these associations (0.82 (0.78, 0.87), 0.25 (0.21, 0.30) and 0.53 (0.45, 0.61), respectively), and these findings were consistent for each type of secondary substance considered. Moreover, we observed different associations in terms of direction and magnitude between secondary substance use and primary drug use during off-treatment periods across substance types.ConclusionThis study demonstrates secondary substance use moderates the temporal associations between treatment and primary drug use among primary heroin, MA and cocaine users. Disparate patterns of polydrug use require careful measurement and analysis to inform targeted treatment for polydrug users.Copyright © 2017 Elsevier B.V. All rights reserved.

39 keywords...

hide…