Background: Frequent testing of gay and bisexual men (GBM) at higher-risk of HIV is central to current prevention strategies. We conducted the first randomised trial to determine if access to HIV self-testing would increase testing frequency in two groups of higher-risk GBM; those who had tested within the past two years and those who had not.
Methods: In this wait-list control randomised trial, HIV-negative higher-risk GBM reporting condomless anal intercourse or >5 male sexual partners in the past 3 months were recruited at three clinical and two community-based sites in Australia. Enrolled participants were randomly assigned (1:1) via computer-generated randomisation codes to have free access to HIV self-testing (intervention) or not (standard-care). Participants completed 3-monthly online questionnaires. The primary outcome was the number of HIV tests over 12 months, analysed by intention-to-treat. The study was designed to evaluate the primary outcome overall and in two strata: frequent (last HIV test ≤2 years ago) and infrequent (>2 years ago or never tested) testers.
Results: Between Dec-2013, and Nov-2014, 180 men were randomised to self-testing and 179 to standard-care. The intention-to-treat analysis included men who completed any follow-up questionnaire: 179 (98%) in self-testing; and 164 (92%) in standard-care. The mean number of HIV tests over 12 months in the self-testing and standard-care arms was 3.9 and 1.6 per-person overall (rate ratio (RR):2.39, 95% CI: 2.08-2.76,p< 0.001), 4.0 and 1.8 among frequent testers (RR:2.23, 1.93-2.59,p< 0.001), and 3.2 and 0.6 among infrequent testers (RR:5.54, 3.15-9.74,p< 0.001), respectively. There was no statistical difference between the two arms in the mean number of facility-based HIV tests (1.4 vs 1.6, RR:0.89, 0.75-1.06) and any STI test (1.6 vs 1.7, RR:0.93, 0.79-1.10).

Type of testOverallFrequent testersInfrequent testers
 Self-testing (n=177)Standard-care (n=164)Rate-ratio (95%CI)Self-testing (n=147)Standard-care (n=140)Rate-ratio (95%CI)Self-testing (n=30)Standard-care (n=24)Rate-ratio (95%CI)
Self/ facility-based HIV3.9 (0.2)1.6 (0.1)2.39 (2.08-2.76)4.0 (0.2)1.8 (0.1)2.23 (1.93-2.59)3.2 (0.5)0.6 (0.2)5.54 (3.15-9.74)
Facility-based HIV1.4 (0.1)1.6 (0.1)0.89 (0.75-1.06)1.6 (0.1)1.8 (0.2)0.88 (0.73-1.05)0.8 (0.3)0.6 (0.2)1.41 (0.73-2.73)
Any STI1.6 (0.1)1.7 (0.2)0.93 (0.79-1.10)1.8 (0.1)1.9 (0.2)0.92 (0.77-1.09)0.9 (0.3)0.7 (0.2)1.35 (0.73-2.49)
Chlamydia/ Gonorrhoea1.5 (0.1)1.6 (0.1)0.94 (0.80-1.12)1.6 (0.1)1.8 (0.2)0.93 (0.78-1.11)0.9 (0.3)0.6 (0.2)1.33 (0.71-2.50)
Syphilis1.4 (0.1)1.5 (0.1)0.90 (0.76-1.08)1.5 (0.1)1.7 (0.1)0.89 (0.74-1.07)0.8 (0.3)0.6 (0.2)1.41 (0.72-2.76)
[Mean number of HIV and sexually transmitted infection (STI) tests over 12 months in intention-to-treat population. Mean(SD) unless otherwise specified]

Conclusions: HIV self-testing among higher-risk GBM increased HIV testing frequency by more than two-fold overall, and more than five-fold among infrequent testers, without reducing facility-based HIV/STI testing frequency. Self-testing should be provided more widely to achieve public health goals of increasing HIV testing frequency.