Methamphetamine Use and Hiv Risk Behavior Among Men Who Inject Drugs: Causal Inference Using Coarsened Exact Matching Publisher Pubmed



Noroozi M¹ ; Higgs P² ; Noroozi A³ ; Armoon B⁴ ; Mousavi B⁵ ; Alikhani R⁵ ; Bazrafshan MR⁶ ; Astaneh AN⁷ ; Bayani A⁸ ; Moghaddam LF⁹ Show Affiliations
Source: Harm Reduction Journal Published:2020

Abstract

Background: Understanding the association between methamphetamine (MA) use and HIV risk behavior among people who inject drugs (PWID) will assist policy-makers and program managers to sharpen the focus of HIV prevention interventions. This study examines the relationship between MA use and HIV risk behavior among men who inject drugs (MWID) in Tehran, Iran, using coarsened exact matching (CEM). Methods: Data for these analyses were derived from a cross-sectional study conducted between June and July 2016. We assessed three outcomes of interest - all treated as binary variables, including distributive and receptive needle and syringe (NS) sharing and condomless sex during the month before interview. Our primary exposure of interest was whether study participants reported any MA use in the month prior to the interview. Firstly, we report the descriptive statistics for the pooled samples and matched sub-samples using CEM. The pooled and matched estimates of the associations and their 95% CI were estimated using a logistic regression model. Results: Overall, 500 MWID aged between 18 and 63 years (mean = 28.44, SD = 7.22) were recruited. Imbalances in the measured demographic characteristics and risk behaviors between MA users and non-users were attenuated using matching. In the matched samples, the regression models showed participants who reported MA use were 1.82 times more likely to report condomless sex (OR = 1.82 95% CI 1.51, 4.10; P = 0.031), and 1.35 times more likely to report distributive NS sharing in the past 30 days, as compared to MA non-users (OR = 1.35 95% CI 1.15-1.81). Finally, there was a statistically significant relationship between MA use and receptive NS sharing in the past month. People who use MA in the last month had higher odds of receptive NS sharing when compared to MA non-users (OR = 4.2 95% CI 2.7, 7.5; P = 0.013). Conclusions: Our results show a significant relationship between MA use and HIV risk behavior among MWID in Tehran, Iran. MA use was related with increased NS sharing, which is associated with higher risk for HIV exposure and transmission. © 2020 The Author(s).