Are violence, harmful alcohol/substance use and poor mental health associated with increased genital inflammation?: A longitudinal cohort study with HIV-negative female sex workers in Nairobi, Kenya.

Abstract

Violence, alcohol use, substance use and poor mental health have been linked with increased HIV acquisition risk, and genital inflammation enhances HIV susceptibility. We examined whether past 6 month experience of these exposures was associated with increased genital inflammation, thereby providing a biological link between these exposures and HIV acquisition risk. The Maisha Fiti study was a longitudinal mixed-methods study of female sex workers in Nairobi, Kenya. Behavioural-biological surveys were conducted at baseline (June-December 2019) and endline (June 2020-March 2021). Analyses were restricted to HIV-negative women (n = 746). Women with raised levels of at least 5 of 9 genital inflammatory cytokines were defined as having genital inflammation. Multivariable logistic regression models were used to estimate (i) baseline associations between genital inflammation and violence, harmful alcohol/substance use, and poor mental health, and (ii) longitudinal associations between these exposures at different survey rounds, and genital inflammation at follow-up. Inflammation data was available for 711 of 746 (95.3%) women at baseline; 351 (50.1%) had genital inflammation, as did 247 (46.7%) at follow-up. At baseline, 67.8% of women had experienced physical and/or sexual violence in the past 6 months, 33.9% had harmful alcohol use, 26.4% had harmful substance use, 25.5% had moderate/severe depression/anxiety, and 13.9% had post-traumatic stress disorder. In adjusted analyses, there was no evidence that these exposures were associated cross-sectionally or longitudinally with genital inflammation. We report no associations between past 6 month experience of violence, harmful alcohol/substance use, or poor mental health, and immune parameters previously associated with HIV risk. This suggests that the well-described epidemiological associations between these exposures and HIV acquisition do not appear to be mediated by genital immune changes, or that any such changes are relatively short-lived. High prevalences of these exposures suggest an urgent need for sex-worker specific violence, alcohol/substance use and mental health interventions.