Promoting partner and couples HIV testing using self-test kits in Kenya

Publication Details

Thirumurthy, H, Omanga, E, Obonyo, B, Masters, S and Agot, K, 2017. Using HIV self-testing to promote male partner and couples testing in Kenya, 3ie Impact Evaluation Report 60. New Delhi: International Initiative for Impact Evaluation (3ie)

Link to Source
Harsha Thirumurthy, Eunice Omanga, Beatrice Obonyo, Samuel Masters and Kawango Agot
Institutional affiliations
None specified
Grant-holding institution
None specified
Sub-Saharan Africa (includes East and West Africa)
Health Nutrition and Population
Gender analysis
Gender analysis
Equity Focus
None specified
Evaluation design
Randomised Control Trials (RCT)
3ie Series Report
3ie Funding Window
HIV self-testing thematic window


The study evaluates the impact of distributing HIV self-test kits to women at antenatal care (ANC) clinics on the uptake of male partner or couple’s HIV testing.


Increasing the uptake of HIV testing and counselling (HTC) in Sub-Saharan Africa is essential for improving the effectiveness of HIV treatment and preventing new infections. Although Kenya has nearly met its target of 80 per cent coverage of HTC among adults, uptake of HTC among men, uptake of repeat testing, and knowledge of HIV status among HIV-positive persons remains low. Four in ten new HIV infections occur within stable heterosexual partnerships and the majority of persons in serodiscordant relationships are unaware of their HIV status. Couples testing increases likelihood of mutual HIV status disclosure and is recommended by the Kenyan government.

Research questions

Does the provision of HIV self-test kits increase HIV testing uptake among partners of women receiving antenatal care and postpartum care services relative to clinic referral testing?
Does HIVST increase the discussion of HIV testing with partners?
Does HIVST increase couples testing?
Does HIVST increase couples testing among those whose partners tested for HIV?
Does HIVST increase awareness of partner’s HIV status?
Does HIVST increase awareness of partner’s HIV status among those whose partner tested for HIV?
Does HIVST contribute to new HIV case identification?


Intervention design

The intervention was developed in consultation with community stakeholders, national and provincial HIV-testing task forces and with support from the local Ministry of Health (MOH) officials and a community advisory board.

Women in the treatment arm received two OraQuick Rapid HIV-1/2 Tests and instructions for use for them and their partners. They received training on proper use of the test, including a demonstration. Participants were counselled about the importance of partner testing and provided suggestions for broaching the topic of HIV testing. They also received a voucher and counselling on where they could go for confirmatory testing in case of a positive result. Women in the control arm received counselling on the importance of partner testing and encouraged to refer their partners for testing (standard of care). They were given referral cards for clinic testing to give to their partners.  The cards summarised the information presented, mentioned three study sites, and encouraged testing at these sites and to present the card. The referral cards were used to verify HIV testing in the control group.

Theory of change

HIV self-testing addresses some of the barriers often cited by men such as lack of time, opportunity costs and transportation costs. Home delivery avoids opportunity and travel costs. Self-testing can increase the perception of privacy and confidentiality, and be more convenient (one can take it at a time of one’s choosing and it only takes 20 minutes). Removing barriers lowers the cost of getting tested, increasing the value of the outcome, making HIVST more attractive.

Evaluation design

The study used RCT design. There were 286 participants in the control group and 284 participants in the experimental group. The primary, pre-specified outcome was the participant’s report on HIV testing by their primary partner within three months of enrollment. Ex-post comparison of testing was compared between intervention and control using an unadjusted modified Poisson regression.

Main findings


Male partner testing within three months of enrollment was higher in the self-testing group (91%) than the control (52% p < 0.001). Over 95 per cent of participants in both groups reported discussing HIV testing with their partner (no difference). The self-testing group participants were more likely to test as a couple than the comparison group (75% versus 33%, p < 0.001), and more likely to test as a couple conditional on testing, (83% vs. 64%, p < 0.001). Disclosure of status was more common in the self-testing group (89% vs 51%, p < 0.001), but not conditional on testing (98% for both).

Cost efficiency analysis

The total cost would be US$10,545 making the cost per additional partner test US$27. However, the value is highly dependent on the cost of the kits, which cost US$10 per kit. Costs with mass distribution may be lower, improving the cost-effectiveness of the intervention.

Implications for implementers

Offering HIV self-test kits to women to share with their partners is highly acceptable. Scaling up this intervention would require training of clinic staff on the use of HIV self-testing so that they can provide adequate instruction to women on their use and ways to broach the topic with partners.

Implications for policy and practice

Providing free HIV self-test kits to women in clinic settings is extremely effective at increasing partner testing rates and disclosure of HIV status. Given the sizable prevention benefits of testing, demand creation interventions such as this one warrant strong consideration by programs and countries seeking to increase HIV testing. Implemented at scale over a longer period could provide additional cost savings.

Implications for further research

Future studies should examine confirmatory testing and linkage to care following self-testing to ensure that those who test positive seek care. More research is needed to understand how partner distribution of self-tests would interact with widespread availability of self-tests, on optimal pricing or subsidy for test kit; how much instruction or counseling needs to accompany HIV self-test distribution; as well as to identify effective mechanisms for reaching single men.

Policy influence

This study was cited the WHO HIV self-testing guidelines. One of the researchers has been engaged by the Kenyan MOH to assist with writing the Kenyan HIV self-testing guidelines. HIV self-testing will be rolled out more widely in Kenya in 2017. Due to the WHO HIV self-testing guidelines, most PEPFAR supported countries are including HIVST in their country operational plans, and countries are putting in HIV self-testing in Global Fund concept notes.

Additional publications


Scroll to Top