‘When I have these kits… I have confidence’— HIV self-testing provides agency for women in their sexual decision-making

Abstract

Oral fluid-based HIV self-testing (HIVST) has emerged as a promising approach to increasing HIV testing coverage, particularly among high-risk populations. Understanding the experiences of women using self-tests and offering them to their sexual partners (secondary distribution) is crucial for determining the potential of HIVST. Qualitative in-depth interviews were conducted among 32 women at high-risk of HIV infection, including women who engage in transactional sex, who participated in a cluster randomized trial of a secondary distribution strategy in western Kenya. Interviews explored how women used self-tests within relationships and how this affected their sexual decision-making. Three key themes emerged: women used HIVST to assess risk prior to engaging in sex with partners; HIVST provided women with increased agency to engage in or end relationships; and women appreciated these benefits and urged expanded access to self-tests. HIVST has the potential to support HIV prevention objectives in settings with high prevalence of HIV.

Introduction

Despite significant progress in access to HIV services in sub-Saharan Africa contributing to a 38% reduction in new HIV infections between 2010–19 (UNAIDS, 2020), 62% of new HIV infections over this time period were concentrated among key populations, including female sex workers (FSW), females engaged in transactional sex, and their sexual partners (UNAIDS, 2020). Additionally, women and girls continue to carry a disproportionate burden of HIV in sub-Saharan Africa, accounting for 59% of all new HIV infections in 2019 (UNAIDS, 2020). Therefore, identifying novel HIV prevention strategies for women at high-risk of exposure remains essential.

HIV prevalence among women who sell sex in Kenya is estimated to be 29.3% (Bhattacharjee et al., 2019; National AIDS Control Council & Ministry of Health, 2018) and recent data indicate a high number of active sex work ‘hotspots,’ which are areas characterized by transactional and commercial sex (National AIDS and STI Control Programme (NASCOP), 2019). Notably, the difference between those engaged in sex work and those engaged in transactional sex comes down to self-identifying as a sex worker versus an implicit exchange of money or goods (Wamoyi et al., 2019). The term ‘transactional sex’ recognizes that there is a spectrum of paid sex arrangements; it does not need to be differentiated from sex work when it comes to HIV risk (McMillan et al., 2018).

The burden of HIV and prevalence of transactional sex is especially concentrated in western Kenya, along Lake Victoria (Camlin et al., 2013; Kiwanuka et al., 2014; Kwena et al., 2014). Studies have reported HIV prevalence between 25–29% and annual incidence rates of 3.39–4.9% in western Kenya’s fishing communities (Camlin et al., 2013; Kiwanuka et al., 2014; Kwena et al., 2014). As in other regions within sub-Saharan Africa, men in western Kenya have lower engagement in HIV services than women, resulting in higher HIV morbidity and mortality for men despite the higher HIV infection risk for women (Beckham et al., 2016; Cornell et al., 2011; Druyts et al., 2013; UNAIDS, 2017).

To broach these challenges, HIV self-testing (HIVST) has emerged as a promising approach to increase HIV testing coverage, particularly among men, young people, and in high-risk populations. HIVST offers greater privacy and convenience than clinic-based testing and has high acceptability among hard-to-reach groups, including key populations (Choko et al., 2015; Johnson et al., 2014; World Health Organization, 2016). In accordance with World Health Organization guidelines supporting its scale-up (World Health Organization, 2016), many countries have introduced HIVST in the retail and public sector, yet reaching people living with HIV who are unaware of their status or at high-risk of HIV infection remains a challenge. One WHO-endorsed strategy to overcome this challenge is the provision of multiple self-tests to individuals who then voluntarily distribute the kits to others, known as secondary distribution of self-tests.

Previous qualitative and quantitative studies in Kenya and elsewhere have demonstrated the acceptability and feasibility of a single distribution of multiple HIV self-tests to women for their own use and for secondary distribution to their sexual partners (Choko et al., 2017; Harichund et al., 2019; Maman et al., 2017; Matovu et al., 2018; Qin et al., 2018). Studies have also found that women, including FSWs, have the discretion and agency to safely introduce self-tests to some of their sexual partners, and that women may make better-informed sexual decisions by offering self-tests to their partners (Maman et al., 2017; Thirumurthy et al., 2016). Through one-time provision of self-tests to women at high-risk of HIV exposure, including FSW, pregnant women, and post-partum women, studies conducted in Kenya and Malawi have also shown that secondary distribution results in higher male partner testing (Choko et al., 2019; Masters et al., 2016). Beyond the antenatal clinic setting, a few studies have also examined how to reach high-risk women and their male partners with self-tests, as well as men’s experiences receiving and using HIVST kits through sustained secondary distribution of self-tests (Napierala et al., 2019; Napierala et al., 2020). However, there is little published literature examining women’s experiences using and distributing HIV self-tests when they have sustained access to self-tests. This study addresses such gaps in the literature by 1) giving women sustained access to free self-tests that they can use themselves and also offer to their male partners; 2) focusing on a population of women at high-risk of HIV exposure since they reported multiple partners and engaged in transactional sex; 3) longitudinally following-up with women to see how self-tests affected their relationship dynamics and HIV risk.

HIVST has the potential to serve as a powerful HIV prevention tool. Understanding the experiences of women, how they distribute self-tests, reasons for selecting partners for secondary distribution, whether and how male partners use the tests, sexual and health seeking behaviour post-HIVST, and implications for relationships are all crucial dimensions to determine the HIV prevention implications of HIVST secondary distribution in high-risk settings. Within the context of the Jikinge study, a matched-pair cluster randomized controlled trial that evaluated the effect of providing women with sustained access to free HIV self-tests (Thirumurthy et al., 2021), we conducted a qualitative study to examine how women used self-tests in their relationships and how this affected their sexual decision-making.

Materials and Methods

Qualitative interviews were conducted with women at high-risk of HIV infection who participated in the intervention group of the Jikinge study. HIV-negative women aged ≥18 years who self-reported ≥2 male partners in the past month were eligible for the Jikinge study. Sampling methods have been published in the primary paper (Thirumurthy et al., 2021). Notably, randomization units were communities with a high prevalence of transactional sex, including beach communities along Lake Victoria and inland communities with hotspots for transactional sex such as bars and hotels. 66 clusters comprised of beach communities and hotspots in Siaya County, Kenya were randomized into intervention and control groups. Participants in the intervention group received 5 oral fluid-based HIV self-tests at enrolment and additional self-tests at 3-monthly intervals or more frequently if needed. Participants were encouraged to offer self-tests to sexual partners with whom they did not anticipate using a condom. Follow-up study visits were conducted at 6-monthly intervals. At each follow-up visit, women received rapid HIV antibody testing and questionnaires that collected information on participants’ use and distribution of self-tests and partner testing, couples testing, sexual behaviour, and intimate partner violence. Participants were followed for one to two years, with a median follow-up duration of 18 months. The study protocol was approved by the ethics review committee at Maseno University and Institutional Review Board at the University of Pennsylvania.

The study team set out to explore women’s use of the self-tests over time. Qualitative interviews were conducted with a sub-set of participants at 12 months and 18 months, with a few control group members participating in the first round of interviews and only intervention group members participating in both 12- and 18-month interviews. We randomly selected interview participants from beach clusters and hotspot clusters; within each cluster, participants were stratified by whether they received above or below the median number of self-tests (three) since enrolment. We contacted participants and invited them take part in an interview at the location of their choosing and participation was based on interest and availability. The first interview was conducted approximately 12 months after enrolment. The same participants were contacted at 18 months for a second interview. Participants provided written informed consent for each interview and received compensation of 550 Kenya Shillings (~US$5) for their time, transportation fare refunds, and a light refreshment. The main Jikinge trial was conducted from June 2017 through March 2020 and qualitative interviews were conducted between July 2019 and February 2020.

The interviewers asked participants about how they decided which partners to offer the self-tests to, how they discussed the self-tests with partners, self-testing experiences with their different partners, and how/whether the testing experience with partners affected their sexual behaviour with those partners. Using the two time points, we set out to understand their experiences with secondary distribution of self-tests over time, whether they offered self-tests to new partners or conducted repeat testing with existing partners, and how discussions and negotiations after self-testing evolved during the study. Finally, we used these interviews to collect information on women’s reactions to the process by which the self-tests were distributed—what worked or not, whether test kits were available when needed, and what suggestions they had for scaled up implementation, including places where they would prefer to obtain self-tests.

Interviews were conducted by Research Assistants (RA) in Kiswahili, Dholuo, or English depending on participant preference, and translated and transcribed by local study staff at Impact Research and Development Organization (IRDO) in Kisumu, Kenya. The RA that conducted the interview reviewed the final transcript for accuracy. All data were coded and analysed using Dedoose software (Dedoose, 2019). The team developed a baseline code book for the first round of interviews and a separate code book for the second, follow-up interviews. The codebooks included topical or deductive codes, allowing the team to index the interviews by the topics that were covered. The data were then coded using these deductive codes by two coders for the baseline interviews and by a single coder for the follow-up interviews. Quality control checks were done on the initial eight transcripts to confirm intercoder reliability. We generated code reports for each of the deductive codes and produced a comprehensive summary report of the data using these code reports. We created matrices to summarize the data by participant and compare responses and patterns across the participants and over the time periods. We used these matrices to identify emerging patterns and conduct thematic analyses, which is the focus of the results below.

Results

Baseline interviews were completed by 45 participants, 35 of whom were part of the intervention, and 32 intervention participants completed follow-up interviews. Ten of the 45 were part of the control group and therefore were not chosen for follow-up interviews as part of this study; three of the 35 intervention individuals could not be located to participate in a follow-up interview. Of the 32 follow-up interview participants, 15 were sampled from sex work hot spots and 17 were from beach communities. The average age of the participants who completed follow-up interviews was 26.6 years (IQR 21.5, 30.5) (Table 1). 12 women completed some primary school or less, eight completed primary school, and 12 completed more than primary school. Four women reported that sex work was their primary income source and 19 reported that sex work was a secondary income source (Table 1). In the broader Jikinge study, 15% of women reported sex work as their primary source of income and an additional 51% reported it as a secondary source of income; therefore 67% of the overall study population was engaged in transactional sex. There were no major differences in the demographic characteristics between the two populations (Thirumurthy et al., 2021). Throughout the qualitative study, we do not emphasize differences between the primary income and secondary income sex work population, because sex work was a matter of self-identification.

Table 1.

Characteristics of Jikinge Study qualitative interview participants

Variable	N (%) (N = 32)
Age, mean (SD)	26.6 (6.3)
Age, median (IQR)	25 (21.5, 30.5)
Marital status
Married	19 (59.4)
Single	7 (21.9)
In a relationship, but not living together	4 (12.5)
Divorced or Widowed	2 (6.3)
Highest education completed
Some primary or less	12 (37.5)
Primary	8 (25.0)
Some secondary	6 (18.8)
Secondary or higher	6 (18.8)
Primary income source
Sales/service	12 (37.5)
Skilled manual labor	2 (6.3)
Unskilled manual labor	6 (18.8)
Fishing/trade	2 (6.3)
Sex work	4 (12.5)
Informal/seasonal	1 (3.1)
Student	2 (6.3)
Unemployed	3 (9.4)
Sex work is a secondary source of incomea	19 (67.9)

Note: Percentages may not add to 100% due to rounding.

^aAmong those who did not list sex work as a primary income source (N = 28)

The participants had primarily positive experiences initiating secondary distribution of HIV self-tests; none of the interviewees expressed significant adverse experiences as a result of HIV self-testing. As explored in the primary paper, there was no significant difference in intimate partner violence between participants in the control and intervention groups (Thirumurthy et al., 2021).

Broadly, women recognized that HIVST was a useful component in the fabric of HIV prevention, one that could be used as a tool to help determine when other prevention mechanisms, such as condoms or limiting the number of concurrent partners, were necessary. For example, a male partner’s decision to accept or decline a self-test offer, as well as the result of his test, helped women discern his HIV risk and decide whether unprotected sex was safe or whether condom use was necessary. It also helped participants decide whether sex was too risky and should be avoided altogether. Women’s engagement in unprotected sex was typically initiated by their male partners, but interviews showed that the women used self-tests to make their own determinations regarding which partner(s) represented a low enough risk for condomless sex. We identified three major themes from the interviews: 1. the effect of HIVST on condom use decision making and partner selection; 2. the effect of HIVST on relationships and power dynamics; and 3. ways in which access to HIVST could be expanded.

The effect of HIVST on condom use decision making and partner selection- Using HIVST to assess risk

HIVST was used by women as one way to assess risk and make decisions about condom use. While condom use was low at the study’s baseline, with 35.9% of all study participants reporting condom use at last sexual encounter at baseline (Thirumurthy et al., 2021), the HIVST experience offered women and their partners additional information to negotiate the terms of condom use in their relationship. Many women perceived themselves to be avoiding HIV infection by initiating testing before engaging in sex.

Consistent with the baseline survey data, women reported low use of condoms in the qualitative interviews; more than two-thirds of the women who completed a follow-up interview reported inconsistent condom use in their respective baseline qualitative interviews. Condom use was again low in the follow-up interviews with 21 of the 32 women explicitly mentioning inconsistent condom use and only two participants describing consistent condom use. Some participants continued to use condoms, inconsistently, for contraceptive purposes. However, through women’s narratives we see that HIVST gave women an opportunity to better assess their risk prior to insisting upon or foregoing condom use. For women who had partners who tested negative, women continued to engage with them in condomless sex:

Once I know my status there is no need for me to use condoms and I have also known his status.

(Age 21, Primary Source of Income: Informal/Seasonal, Beach)

For women who had partners who either refused to test or tested positive, women insisted on condom use or ended the relationship(s):

The one I had to put aside [we] had to use condoms because he was positive [identified by HIVST]. That is why I insisted that anytime we were to meet he had to use a condom.

(Age 31, Primary Source of Income: Sex Work, Hotspot)

The women discussed that knowing their partner’s HIV-negative status decreased the risk of HIV and made them feel that condoms after testing was unnecessary, especially if they trusted each other not to have multiple partners or to use condoms with non-primary partners. Rationale for not using both HIVST and condoms included: no need to use additional protection when you both know your negative status; no need for condoms once trust is established; and, push-back from partners because the men did not see a need for testing and condoms. Because condom use was dependent on perceived risk, women were more inclined to use condoms with extra-marital partners, compared to their husbands or primary partners.

Yes, if I tell him [to use HIVST] he will not refuse… We use [condoms] with the partner from outside [the primary relationship] but this one we can’t. With him we just have sex without condoms.

(Age 28, Primary Source of Income: Sales/Service, Beach)

Some women specifically noted that they previously used condoms consistently but replaced condom use with regular self-testing: ‘It changed because before we used condoms… as we continued testing, we stopped using it’ (Age 19, Primary Source of Income: Unskilled Manual, Beach). Women who continued using condoms alongside self-testing used them for contraception, to protect themselves from other sexually transmitted infections (STI), with extra-marital or non-primary partners, or to protect themselves from infection when their partner was known to be living with HIV.

There are many diseases, we can’t test STIs using [HIV self-tests]. So… there are other situations that forces you to use the condoms.

(Age 21, Primary Source of Income: Unskilled Manual, Hotspot)

The effect of HIVST on relationships and power dynamics- HIVST provided women with increased agency

Women demonstrated increased confidence in the use of HIVST with partners over time. This confidence resulted in increased agency to negotiate condom use with partners deemed to be high-risk and to refuse sex or end relationships with those partners who refused condom use.

Over time women described greater confidence in introducing HIVST to their partners. As part of the education they received during enrolment, women role played with the study staff on how to introduce the kits and how to respond to partner questions. These role plays helped them develop their own strategies to communicate about the benefits of testing, the testing procedures, and the implications of the HIV test results. Because of the longitudinal nature of this investigation, women described being uncomfortable introducing the kits initially but gaining self-efficacy over time: ‘Before I had difficulty to convince someone, or any partner to test with these kits… I did not have courage and experience. But after that, after having more experience concerning those things, I am free and I can see it does not take much time’ (Age 22, Primary Source of Income: Sales/Service, Beach). The participants also reported that learning how to use the self-tests and then offering them to their partners allowed them to gain self-confidence and bravery in their overall lives: ‘When I have these kits, I know… I have confidence’ (Age 29, Primary Source of Income: Sales/Service, Beach).

As women’s comfort in introducing the self-tests over time increased, so did their agency in using HIVST to screen out partners who were high-risk. Though male partners who tested HIV-negative initiated condomless sex in most cases, women had the agency to act in situations that put them at risk. Once women determined they were at risk, they took actions such as ending relationships and turning down sex where their partner(s) refused to test or tested HIV-positive and did not want to use condoms. Couples who self-tested and were comfortable forgoing condom use, despite other risks, at least incorporated an HIV prevention tool into their routine, considering they may not have consistently used condoms anyways. This ability to assess risk gave women increased control in HIV prevention, greater confidence, and a greater feeling of autonomy.

I can say the kit helped me because I would have done something without knowing what was lying ahead. So the kit prevented a very big thing for me… You know someone who knows his status, he can cheat you just to infect you with the virus, then leave you. So when I saw the result, I told him there is nothing I can do with you… So I can say this kit is what helped me to avoid what could have happened [unprotected sex with HIV infected partner].

(Age 19, Primary Source of Income: Unemployed, Beach)

Women also displayed agency through their acceptance of gifts and money from partners following HIVST. Women noted that partners offered more gifts and gifts of higher value following HIVST, both as a reward for women’s HIV-negative status and as gratitude for offering men the opportunity to test. Women leveraged this agency to ask for specific needs such as their child’s school tuition, food, phones, and greater compensation.

I got many gifts from my partner who is not my primary partner when he knew my status because he had doubted me. So now at least he trusts me a little bit.

(Age 29, Primary Source of Income: Sales/Service, Hotspot)

Ultimately, HIVST is a powerful tool that shifted power dynamics and allowed women to have greater self-efficacy and avoid engaging in behaviours that they considered high-risk.

Ways in which access to HIVST could be expanded- Based on women’s suggestions

Women viewed HIVST as an equivalently useful and complementary prevention tool to condoms, and they believed self-tests should be widely distributed in the same way condoms are distributed. When asked about how they would like to access HIVST kits after study completion, participants described various options including distribution at pharmacies, distribution by community health workers and village elders, secondary distribution programs through peer educators, and self-test provision at government facilities with accessibility at all days/hours. Women also suggested making self-tests available at market centres, colleges, or bars because they believed that people would only use self-tests if it was convenient to obtain them. One woman recommended church distribution days, and another recommended providing self-tests at antenatal care centres or children’s clinics so that women could access self-tests when receiving other, more socially acceptable, health services.

They should organize [HIVST] like condoms, I always see condoms in school. They are everywhere, if you go to the bathroom, they are there. So if there is a way that they can make even youths to get [HIVST] it can be good.

(Age 21, Primary Source of Income: Skilled/Manual, Hotspot)

Women also voiced that their male partners would value convenience, suggesting distributing self-tests at colleges, bars, and betting shops where men frequent and may be comfortable accessing self-tests.

You know like now most men don’t go for tests… I am the one who can go to pick and bring them home. After I have brought them home, then we can teach ourselves here in the house… That is the only way I can help the men because most men, just going to get those kits is difficult for them, or to go for a test is difficult for them. If you put them at home, then it is easy.

(Age 38, Primary Source of Income: Sales/Service, Hotspot)

Discussion

These qualitative findings demonstrate that women who had sustained access to HIV self-tests can safely and effectively use self-tests to learn their partners’ HIV status, make decisions about sexual behaviour, and gain power and autonomy in relationships. Secondary distribution of self-tests can thus be an empowering HIV prevention tool for women. Women used self-tests as a tool for gauging partners’ risk levels, typically feeling safe to have condomless sex if their partner tested negative, and they had the power to insist on using condoms or refuse sex if a partner was too risky. Over time women expressed increased confidence and agency in the use of HIVST with their partners. Women’s experiences using self-tests were so positive that they urged routine, widespread availability of self-tests. Considering this is the first study to examine women’s experiences when they have sustained access to HIV self-tests, the results provide valuable guidance and implications of expanded access to self-testing.

Across the timepoints women valued self-testing as an empowering tool leading to augmented agency and autonomy over one’s health. Women embraced self-testing and reported that it made them feel more brave, confident, and empowered, perhaps because they were seeking increased agency in a world where men traditionally hold the decision-making power. While gendered power dynamics have evolved in the 21^st century, traditional gender archetypes and Kenyan women’s quest for power still remain (Aberman et al., 2018; Harrington et al., 2016). Gendered power, norms, and inequities have been routinely documented in the literature and are shown to hamper a woman’s HIV prevention behaviours (Celum et al., 2015; Corneli et al., 2014). Female-initiated HIV prevention methods, such as PrEP, have been globally documented to increase women’s autonomy and power over HIV prevention (Flash et al., 2017), particularly in non-monogamous relationships (Eakle et al., 2019). Similarly, female-initiated secondary distribution of HIVST has the ability to shift traditional power dynamics by bringing testing out of the clinic setting, where FSWs and women engaged in transactional sex are traditionally stigmatized (Lancaster et al., 2016), and into the hands of those with heightened risk. This then enhances one’s agency and power over their own risk—all while decreasing the reliance on condoms, a tool often met with resistance from male partners. The benefits experienced led the Jikinge participants to urge easy access to self-tests, much like current condom distribution. Not only did the participants report feeling more confident and empowered due to the introduction of HIVST, but they also acted on the power shift by being willing to end relationships with men who refused to use the self-tests. Scaling up female-initiated HIV prevention methods across Kenya and sub-Saharan Africa is integral to ending the HIV epidemic—a public health burden carried disproportionately by women (amFAR, n.d.; Avert, 2019).

The finding that condom use was dependent on the risk level of the partner (i.e. more consistent use with non-primary partners) is congruous with previous studies on determinants of condom use among females engaged in sex work (Broel et al., 2017; Gallo et al., 2011; Macaluso et al., 2000). Even so, the frequency of condomless sex, with and without HIVST, presents other risks to women including unintended pregnancy, STIs, or HIV transmission should one be in the HIV window period (HIV.gov, n.d.). Regardless of these concerns, condom use behaviours were low and inconsistent across timepoints and therefore point-of-sex self-testing is an important HIV prevention mechanism to add to women’s repertoire. Many of the women would not have been using condoms regardless of HIVST, in which case HIVST allowed them to assess their risk and insist on condom use or end relationships if their partners tested positive or refused to test.

Given that transactional sex was prevalent in our study population and condomless sex is typically better compensated (George et al., 2019; Quaife et al., 2018; Robinson & Yeh, 2011), some women described being able to use HIVST as a way to receive higher compensation without taking on greater HIV risk. In a setting in which women may previously have had condomless sex with partners whose status they did not know, having access to self-tests can be a useful tool for women. Seeing as participants noted increased gifts and compensation associated with HIVST, which was also associated with condomless sex, women may have been able to leverage their power by agreeing to have condomless sex after using HIVST to assess their risk. Therefore, women who engage in sex work may benefit more over time by using HIVST because they can both protect themselves and increase their earnings. It is important to note, however, the unintended consequences of replacing condom use with self-tests, including higher risk of STIs, pregnancy, and acute HIV infection.

Beyond benefiting individuals, self-testing provides potential benefits to the system at large. This task shifting strategy—delegating tasks from a more to less specialized workforce to enhance efficiency (Zachariah et al., 2009)—may reduce the burden on healthcare providers, thus representing an opportunity for positive future impact of self-testing on the broader healthcare system. Several participants noted that they formerly underwent regular clinic-based HIV testing, even with their partners, but have now transitioned to HIVST when available. HIVST enables people to test as frequently as they wish without the individual burden of going to a hospital/clinic, while relieving the strain on overextended health systems. Given the women’s interest in HIVST and the success of this approach in promoting male partner testing, we recommend additional implementation science research to identify the most effective ways to scale up HIVST distribution to members of priority populations as a task-shifting opportunity.

While previous studies have investigated male-initiated violence due to secondary distribution and have determined that there is no evidence that HIVST increases women’s risk for violence or other negative outcomes from partners (Maman et al., 2017; Thirumurthy et al., 2016), there remain other concerns with partner-initiated HIV testing strategies, including HIV self-test secondary distribution. One such concern is the potential for a breach of confidentiality should a partner disclose the other’s HIV status. A breach may also occur deductively, for example when others make assumptions about why a woman ends the relationship. As a result, the person testing positive may face stigma from the testing initiator, or others, and be at greater risk for social and psychological harm. However, the concern around stigma or other unintended consequences of disclosing one’s status are not unique to HIVST and may be inherent to HIV testing, but widespread testing has been shown to reduce such stigma (Genberg et al., 2009; Kalichman & Simbayi, 2003). Another concern that arises with partner initiated HIVST includes the introduction of HIVST without adequate facilitation of linkage to care. Linkage to care challenges may lead to those obtaining reactive results to have low uptake of HIV care. Therefore, there is a need to reinforce the importance of confirming reactive self-test results with a provider, where the individuals can confirm their status, link to care, and receive any counselling and support that they may need. More research is needed to better understand whether people who receive self-tests secondarily link to confirmatory testing and other HIV services.

This qualitative study is not without limitations. First, we may not be able to generalize these findings to all women at high-risk of HIV infection since those who consented to participate in the study may have been more capable of initiating testing with their sexual partners. However, our study recruited women who distributed varying numbers of self-tests to partners, thus ensuring that we captured a range of women’s experiences. Second, the experiences in this particular paper are reported from the perspective of the women and not their male partners, thus we cannot speak to men’s experiences with HIVST and whether they echoed the reported sentiments of women in this analysis. Despite these limitations, we believe that the results reported have important implications for the scale up of HIVST among priority populations.

This qualitative study explored how women at high-risk of HIV infection distributed and used self-tests when they had sustained access over an extended period. Our findings suggest that enhanced availability of self-tests may translate to HIV prevention benefits over time. Women noted many advantages to having access to self-tests, including increased power and autonomy that allowed them to determine when additional HIV prevention behaviours (e.g., condom use or abstinence) were necessary or whether to disengage from high-risk sexual relationships. These effects of the intervention contributed to a perception that access to self-tests facilitated HIV prevention. This paper shows the multitude of ways in which women at high-risk of HIV infection are likely to value sustained access to HIV self-tests.

Data Availability Statement:

De-identified qualitative data may be made available upon request.