ABSTRACTBackground: Marijuana use is common among HIV+ individuals, but few studies have examined long-term trends in prevalence and correlates of use. Methods: We evaluated trends (1984–2013) in the annual prevalence of current (past 6-month use) and daily (among current users) marijuana use and determined correlates of use among 2742 HIV-seropositive (HIV+) and 3172 HIV-seronegative (HIV−) men who have sex with men in the Multicenter AIDS Cohort Study (MACS). Poisson regression models were used to estimate prevalence ratios of marijuana use separately for the men who were enrolled before 2001 (early-cohort) and after 2001 (late-cohort). Results: Over the 29 years of the study, the prevalence of current marijuana use declined significantly, whereas daily use among users increased among all men in the early and late-cohorts. A HIV+ status was associated with higher prevalence of marijuana use among the men in the early-cohort (adjusted prevalence ratio [aPR] = 1.53, 95% confidence interval [CI]:1.42, 1.64, p = <0.0001), but not in the men in the late-cohort (aPR = 0.90, 95% CI: 0.79, 1.03, p = 0.1424). Alcohol use and cigarette smoking were being positively associated with marijuana use. Conclusions: Although the annual prevalence of current marijuana use decreased significantly over time in the MACS, daily use among users increased significantly. Further, among the HIV+ men, our study did not show clinically significant adverse effects of marijuana use on highly active antiretroviral therapy use, CD4+ count, or HIV viral load.