HIV self-testing is not a ‘magic bullet’

18 March 2019

Self-testing reached men and youth with knowledge of their HIV status, but further intervention is needed to link people to antiretroviral treatment.  

Man, woman and baby at health clinic in Malawi

HIV self-testing in Malawi increased knowledge of HIV status among partners of people living with HIV, but it underperformed in linking people to care. Just 23% of people who self-tested HIV-positive initiated treatment within six months, compared to 75% of people who tested at a medical facility.

The results of this randomised control trial assessed the impact of index HIV self-testing (HIVST) among partners of antiretroviral treatment (ART) clients in Malawi. The findings were presented earlier this month at the annual Conference for Retroviruses and Opportunistic Infections (CROI 2019) held in Seattle, USA (4-7 March).

73% of people who were given an HIVST kit tested for HIV compared to just 27% who received a standard care model.

Partners of people living with HIV, particularly men, are among the most at-risk groups for HIV, but getting them to test for HIV is a challenge. Studies from Malawi and Kenya have shown HIVST can increase access to HIV testing among partners of sex workers and antenatal patients, however among partners of people living with HIV effectiveness remains less clear. Among this group of patients there were concerns that HIVST would lead to automatic disclosure, if they have not already told their partner they are living with HIV.

The study randomised into two arms 484 patients (113 male and 371 female) attending an antiretroviral treatment clinic across three hospitals in Malawi. The final analysis included 365 people who completed baseline follow-up.

The standard of care (SOC) arm consisted of a passive partner referral slip which the patient took home to their partner. In the HIVST arm (intervention), patients were given the passive referral slip in addition to an OraQuick self-test, which they were shown how to use in case their partner needed assistance. The primary outcome was knowledge of status, while secondary outcomes also looked at positivity rates, ART-initiation at 6 months and adverse events. A cost-analysis was also conducted.

The demographics across both arms were similar – most ART patients were female, in stable relationships and had disclosed their HIV status to their partner. Distribution of HIVST and the SOC was high across the board, except in youth (15-24) where HIVST was more likely to be distributed over SOC.

HIVST was very high in this study, but the benefit was most effective among male partners. HIV-positivity results were similar across all arms, but again, it was men who were most likely to test positive and benefit from the intervention.

Unfortunately, ART-initiation at six months was poor in this study among people who received the intervention. This was mainly driven by male partners of whom just 22% were linked to care compared to 75% in the SOC arm. The main reasons for non-initiation were: no time/travelling; afraid of unwanted status disclosure; not ready/not accepted HIV status; and one person went to the hospital but the ART clinic was closed.

HIVST was highly acceptable and there were no adverse events. Usability results indicate that HIVST are not necessarily intuitive and people do need assistance in carrying out the tests – 65% of the partners received help from their partners to carry out the test while 8% could not interpret their results.

In her discussion, lead researcher Kathryn Dovel of UCLA remarked that self-testing alone cannot be a ‘magic bullet’ for getting people diagnosed and onto treatment, and that additional treatment linkage interventions are needed, and particularly ones that target men. These could include home ART-initiation with motivational counselling, as well as providing more male-friendly spaces for health services.

Photo credit:
Corrie Wingate

Written by Caitlin Mahon

Content Specialist - HIV & Sexual Health

Community guidelines for comments can be found in our website T&Cs