Assessing rates and contextual predictors of 5-year mortality among HIV-infected and HIV-uninfected individuals following HIV testing in Durban, South Africa

Abstract Background Little is known about contextual factors that predict long-term mortality following HIV testing in resource-limited settings. We evaluated the impact of contextual factors on 5-year mortality among HIV-infected and HIV-uninfected individuals in Durban, South Africa. Methods We used data from the Sizanani trial (NCT01188941) in which adults (≥18y) were enrolled prior to HIV testing at 4 outpatient sites. We ascertained vital status via the South African National Population Register. We used random survival forests to identify the most influential predictors of time to death and incorporated these into a Cox model that included age, gender, HIV status, CD4 count, healthcare usage, health facility type, mental health, and self-identified barriers to care (i.e., service delivery, financial, logistical, structural and perceived health). Results Among 4816 participants, 39% were HIV-infected. Median age was 31y and 49% were female. 380 of 2508 with survival information (15%) died during median follow-up of 5.8y. For both HIV-infected and HIV-uninfected participants, each additional barrier domain increased the HR of dying by 11% (HR 1.11, 95% CI 1.05–1.18). Every 10-point increase in mental health score decreased the HR by 7% (HR 0.93, 95% CI 0.89–0.97). The hazard ratio (HR) for death of HIV-infected versus HIV-uninfected varied by age: HR of 6.59 (95% CI: 4.79–9.06) at age 20 dropping to a HR of 1.13 (95% CI: 0.86–1.48) at age 60. Conclusions Independent of serostatus, more self-identified barrier domains and poorer mental health increased mortality risk. Additionally, the impact of HIV on mortality was most pronounced in younger persons. These factors may be used to identify high-risk individuals requiring intensive follow up, regardless of serostatus. Trial registration Clinical Trials.gov Identifier NCT01188941. Registered 26 August 2010