Background: Universal and regular HIV testing for early HIV-diagnosis is critical to reduce HIV incidence. Home-based HIV counselling and testing campaigns (HBHCT) have been shown acceptable and successful in high HIV prevalence setting, but little is known about the capacity of such strategy to re-test people regularly. We aimed to describe time to, and factors associated with, repeat HIV testing through HBHCT in rural South Africa.
Methods: From March 2012 to June 2016, the ANRS 12249 TasP cluster-randomized trial aimed at evaluating the impact of a Universal Test and Treat (UTT) approach on population-based HIV incidence in a rural area of KwaZulu-Natal with about 30% HIV prevalence. In both arms, rapid HIV testing was offered at home every six month to all resident members aged ≥16 years old (about 1,000 individuals/cluster). Individuals who tested HIV-negative at first contact with TasP fieldworkers were included in the analysis. Times to HIV-retests since first HIV-negative test were described using Kaplan-Meier curves, with right censoring at the first positive rapid test, death, out-migration and end of observation. Factors associated with first HIV-retest were identified using Cox regression with time-dependant variables, and taking account of cluster effect.
Results: 16,372 individuals with a first HIV-negative test were included in the analysis. From the first HIV-negative test, the probability of one repeat HIV test after one year was 61.2%, the probability of two repeat HIV tests after two years and three repeat HIV tests after three years were 52.3% and 51.4%, respectively (Figure 1). Factors associated with first repeat HIV testing were female gender (HR=1.23; 95%CI=1.18-1.29), being 30-59 years old (HR=1.12; 95%CI=1.05-1.18) or >60 years old (HR=1.18, 95%CI=1.07-1.29) versus < 30 years old, being engaged but not married versus never engaged (HR=1.13; 95%CI=1.05-1.23), having low educational level (HR=1.20; 95%CI=1.12-1.28).
Conclusions: Home-based HIV testing campaigns could contribute to achieve the target of universal HIV testing and ultimately UTT but may not be sufficient considering the assumptions of the Granich''s mathematical modelling (Lancet 2009). A combination of clinic-based and community-based strategies will be needed to maximise HIV testing coverage and specially reach men and youngest individuals.

Time to HIV restest from first HIV-negative test in the ANRS 12249 TasP trial
[Time to HIV restest from first HIV-negative test in the ANRS 12249 TasP trial]

Download the e-Poster