Abstract Body

Early retention in care (RIC), sex, and sexual mode of HIV acquisition have each been associated with mortality risk among persons living with HIV (PLWH). We therefore assessed whether early RIC mediates or modifies the effect of sex and sexual mode of HIV acquisition on mortality among PLWH on antiretroviral therapy (ART) in the Americas.

ART-naïve, adult PLWH (≥18 years old) enrolling at Caribbean, Central and South America network for HIV epidemiology (CCASAnet) and Vanderbilt Comprehensive Care Clinic sites 2000-2015, starting ART, and with ≥1 visit after ART-start were included. Early RIC was defined as ≥2 HIV care visits/labs ≥90 days apart in the first year after ART-start. Sex and sexual mode of HIV acquisition were categorized as women, heterosexual men, and men who have sex with men (MSM). Individuals were followed from one year after ART-start to date of death, last clinic visit or study closure. Cox regression models assessed the association between early RIC, sex and sexual mode of HIV acquisition, and mortality beyond the first year of ART; interactions between RIC and sex and sexual mode of HIV acquisition were tested. Associations were estimated for each site separately and pooled.

Among 11,721 PLWH with ≥1 visit after starting ART, 647 subsequently died (rate=10.9/1000 person-years) and 1985 were lost to follow-up (rate=33.6/1000 persons-years). Early RIC substantially decreased mortality during subsequent years at all sites (Figure), adjusting for age, sex and sexual mode of HIV acquisition, ART-start year, and pre-ART nadir CD4, AIDS, and HIV-1 RNA. Sex and sexual mode of HIV acquisition were also associated with mortality beyond the first year of ART, with MSM having lower risk (pooled adjusted aHR=0.75; 95%CI: 0.61-0.92) and heterosexual men having similar risk (aHR=0.92; 95%CI: 0.74-1.13) to women. Point estimates for sex and sexual mode of HIV acquisition were similar whether or not retention was included. Moreover, there was no evidence of an interaction between sex and sexual mode of HIV acquisition and early RIC (p>0.05).

In the Americas, early RIC significantly decreased mortality risk after one year on ART. Additionally, MSM had a lower mortality risk than women. We found no evidence of RIC mediating or modifying the association between sex and sexual mode of HIV acquisition and mortality in our population.