Association between self-reported marijuana use and incident diabetes in women and men with and at risk for HIV

Introduction: Marijuana use is common among persons living with HIV, but whether it's use increases the risk of type 2 diabetes in this population has not been explored. Objective: To determine whether self-reported marijuana use is associated with incident type 2 diabetes in women and men living with and at risk for HIV. Methods: We analyzed data from the Women's Interagency HIV Study (WIHS) and Multicenter AIDS Cohort Study (MACS), between 2000-2017 (WIHS) and 1999-2017 MACS. The association between self-reported marijuana use and incident type 2 diabetes was analyzed using time-dependent Cox regression models among 3578 and 2682 participants in the WIHS and MACS respectively. Results: Over the follow-up period, 452 (WIHS) and 326 (MACS) incident type 2 diabetes cases occurred. In multivariable models, the hazard ratios, collectively indicate a reduced risk of type 2 diabetes, in marijuana users compared to none users, although all associations were not statistically significant. The results were similar for HIV-positive and HIV-negative participants in both cohorts. Conclusions: In this prospective analysis of nearly 20 years of data for women and men with and at risk for HIV in the WIHS and MACS, although we found a pattern of reduced risk of type 2 diabetes among self-reported marijuana users, the associations were not statistically significant. To better inform clinical decisions and legal policy regarding marijuana use in this population, further longitudinal investigations that biologically quantify marijuana use to assess risk for incident diabetes is warranted.