Acceptability and Usability Evaluation of Finger-Stick Whole Blood HIV Self-Test as An HIV Screening Tool Adapted to The General Public in The Central African Republic.

Background: Opportunities for HIV testing could be enhanced by offering HIV self-testing (HIVST) in populations that fear stigma and discrimination when accessing conventional HIV counselling and testing. Field experience with HIVST was poorly reported in French-speaking African countries.

Objective: To investigate the usability of HIVST in Bangui, Central African Republic.

Methods: The prototype self-test Exacto^® Test HIV (Biosynex, Strasbourg, France) was used to assess the usability of HIVST in 300 adults living in Bangui, according to WHO technical recommendations. Simplified and easy-to-read leaflet was translated in French and Sango.

Results: Preliminary survey in 3,484 adult volunteers including students, men who have sex with men and female sex workers living in Bangui showed that previous HIV testing in conventional centres for HIV counselling and testing was relatively infrequent and that acceptability of HIVST was elevated, although high heterogeneity could be observed between groups. The notice in French and Sango of Exacto^® Test HIV were chosen in 242/300 (80.6%) and 58/300 (19.4%), respectively. It was correctly understood in 273/300 (91.0%). The majority (275/300; 91.6%) correctly performed the HIV self-test; however, 71/300 (23.0%) asked for oral assistance. Most of the participants (273/300; 91.0%) found that performing of the self-test was very easy or easy, and less than Most of participants (273/300; 91.0%) found that performing of the self-test was very easy or easy and less than 1.0% (2/300) found it difficult. Overall the result were correctly interpreted in 96.9% (3,782/3,900), the reading/interpretion errors concerned the positive (96/1,800;5.3%), invalid (17/600;2.8%) and negative (5/1,500; 0.3%) self-test. The Cohen's coefficient κwas 0.94. The main obstacle for HIVST was the educational level, with interpretation difficulties in poorly educated people.

Conclusions: Our observations on profane adults living in Central African Republic, demonstrate: (i) the need to adapt the notice of instruction to African public, including educational pictograms as well as notice in vernacular language(s); (ii) the frequent difficulties in understanding the notice with frequent misinterpretation of test results; (iii) and the generally good usability of the HIV self-test despite these latter pitfalls. More research on exploring the best strategy (i.e. supervised versus unsupervised strategies) for different high- and low- risk populations in resource-constrained settings remains needed.