High uptake of HIV self-testing by men who have sex with men and transgender women in Burma

HIV self-testing is feasible and acceptable for men who have sex with men (MSM) and transgender women (TGW), engaging more people than usual testing services, according to the results of a randomised trial in Burma presented to the 25th Conference on Retroviruses and Opportunistic Infections (CROI 2018) in Boston last week.

The confidentiality and privacy of self-testing, in relation to both HIV status and sexual behaviour, may have particular advantages for stigmatised groups. The conference also heard positive results from a programme distributing self-testing kits through peer networks of MSM in South Africa.

The HIV epidemic in Burma disproportionately affects key populations, including MSM and TGW. The HIV prevalence in these populations is estimated to be 12% but may be as high as 27% in Rangoon, the city where the study was done. Only one in five has ever been tested and access to other healthcare services, including HIV treatment, is also poor.

Earlier qualitative research with MSM and TGW in Burma suggested that self-testing would be acceptable because of its confidentiality and privacy. As neither sexual behaviour nor HIV status would need to be disclosed, participants felt it could help them avoid stigma. While self-testing was expected to be convenient, there was some concern that provision would not include counselling and linkage to care.

In the randomised trial, HIV-negative MSM and TGW were recruited by getting the first recruits to refer their social contacts and peers to the study, with those individuals, in turn, referring people they knew and so on (respondent-driven sampling). Participants were randomised to either:

• HIV self-testing, using an Oraquick test kit, or
• Counselling and testing at a community-based organisation that serves MSM and TGW.

Of note, all participants received pre-test counselling and were also asked to return for a second study visit to report test results and receive post-test counselling and referrals, if necessary. This addressed some of the concerns raised in the qualitative research, as well as meaning that this is not an ‘unsupervised’ model of self-testing.

Of the 577 participants, 85 identified as transgender women. Just over half the participants were described as gay or homosexual, whereas 38% were bisexual. However, it’s worth noting that the categories and terms for gender identity and sexual orientation that are used in Burma are complex and do not neatly correlate with Western categories.

Self-reported HIV risk behaviour was high: 30% had engaged in sex work in the last six months; 30% had used condoms the last time they had sex, and 33% had ever been tested for HIV.

The acceptability of both testing methods was high, with some preference for self-testing: 99% in the self-testing arm and 93% in the community organisation arm said the method was easy; 98% and 95% said the method was convenient; 99% in both arms said they trusted the results of their test; and 99% in both arms said that they would recommend their assigned testing method to a partner, friend, or family member.

Those in the self-testing arm were more likely to return for the second study visit (54% vs 46%). More new HIV diagnoses were reported in the self-testing arm (28 vs 16).

In both arms, the majority of participants said that their preference for future HIV testing would be to use a self-test at home. Visiting a government clinic or hospital was the least popular option. Had self-testing been compared to testing at a government clinic (rather than at a community-based organisation), it is likely that its advantages would have been even more apparent.

South Africa

Similar recruitment methods to the Burmese study were used in Gert Sibande and Ehlanzeni, two districts to the east of Johannesburg in South Africa. A total of 127 men who have sex with men enrolled, two-thirds of whom were aged 18-24. Half only had primary or secondary school education and only 31% had had paid work in the past six months.

Two-thirds described themselves as bisexual – 83% had a regular male partner and 51% a regular female partner.

The study was not randomised and all participants were offered self-test kits. They were given a demonstration of two different kits – the OraQuick oral fluid test used in the Burmese study and the AtomoRapid device which has a built-in lancet and tests fingerprick blood. They were asked to choose the type of test they would like to take home.

Although it’s commonly reported that oral fluid tests are more popular, in this setting participants most commonly chose the fingerprick blood test. The researchers attribute this to familiarity with these tests from public clinics and also due to confusion about the difference between virus and antibody detection (a number of men asked counsellors how HIV would be detected in saliva if the virus cannot be transmitted through kissing).

The researchers used a network distribution approach. Each participant received five test kits and was encouraged to use at least one test kit himself and to share the others with people he felt safe distributing kits and discussing HIV with. The 127 men distributed kits to 376 friends, 217 family members and 135 sexual partners.

Most of the men (91%) used their test kit. Most tested on their own, but a third of participants sometimes tested with other people present, including taking a test at the same times as a friend, family member or partner. This suggests that self-testing can help open up discussion about HIV testing and HIV treatment – a number of participants told the researchers about improved dialogue with partners.

Not many self-testing studies have data on linkage to care, but this one does. Of six participants who were known to have seroconverted, four reported having linked to care and starting treatment. This is comparable to linkage to care data following diagnosis at health facilities in South Africa.

Whereas 38% of the cohort tested at least once every six months before joining the research cohort, six months later 84% reported recent HIV testing. When asked how they would prefer to test in the future, 83% said with self-testing.

The researchers concluded that HIV self-testing is acceptable and feasible for South African MSM and can be disseminated through high-risk peer networks. It increases testing frequency and partner testing, potentially reducing late diagnosis and facilitating access to HIV treatment.