IntroductionHIV infection is still a global public health problem. More than 75% of HIV-infected people are in Africa, and up to 54% are unaware of their HIV status, limiting access to antiretroviral treatment.Context and purpose of the studyThis review aimed to determine whether HIV self-testing (HIVST) increases the uptake of testing, the yield of new HIV-positive diagnoses, and the linkage to antiretroviral treatment, and social harms among adults in Africa.MethodsPubMed, The Cochrane Central Register of Controlled Trials (CENTRAL), Pan African Clinical Trials Registry, The Cochrane Database of Systematic Reviews (CDSR), Databases of Abstracts of Reviews of Effectiveness (DARE), Social Sciences Citation Index, Web of Science and African Index Medicus databases were searched from 1998 to 2019 (updated in December 2019). Eligible trials employed randomized controlled trials (RCTs), before/after studies, and interrupted time series design comparing HIVST to standard HIV testing services or comparing different approaches to HIVST among adults living in Africa were systematically sought.ResultsAfter searching 2,617 citations eleven trials were identified including 59,119 participants from four (4) African countries. Meta-analysis of seven trials showed a significant increase in the uptake of HIVST compared to standard HIV testing services: Both fixed-effects (Rate Ratio (RR) = 2.64, 95% CI: 2.51 to 2.79), and random-effects (RR) = 3.10, 95% CI: 1.80 to 5.37, and a significant increase in the uptake of couples’ HIVST (RR = 2.50, 95% CI: 2.29 to 2.73 in fixed-effects models; and RR = 2.64, 95% CI: 2.01 to 3.49 in random-effects model). A decrease in linkage to care and ART was observed in HIVST compared to standard HIV testing services (RR = 0.88, 95% CI: 0.88 to 0.95 in fixed-effects models; and RR = 0.78, 95% CI: 0. 56 to 1.08 in random-effects models). Six RCTs measured social harms, with a total of ten reported cases related to HIVST. One RCT comparing two approaches to HIVST showed that offering home-based HIVST with optional home-initiation of antiretroviral treatment increased the reporting of a positive HIV test result (RR: 1.86; 95% CI: 1.16 to 2.98), and linkage to antiretroviral treatment (RR: 2.94; 95% CI: 2.10 to 4.12), compared with facility-based linkage to antiretroviral treatment.ConclusionsHIVST has the potential to increase the uptake of HIV testing compared to standard HIV testing services. Offering HIVST with optional home initiation of HIV care compared to HIVST with facility-based HIV care increases HIV positivity and linkage to antiretroviral treatment. Reported incidences of intimate partner violence related to HIVST were rare. Future research should focus on the potential of HIVST to reach first-time testers, the effect of using different approaches to HIVST, and strategies for linkage to HIV services.Systematic review registrationThis systematic review was prospectively registered on the Prospero International Prospective Register of Systematic Review (CRD42015023935).