Acceptability and usability of home-based HIV and blood glucose self-testing and home-based blood pressure measurement in Kenya, South Africa, and Zambia

Read the publication

Abstract

Home-based self-testing may improve individual health outcomes and public health programs by lowering barriers associated with clinic-based testing in low- and middle-income countries (LMICs). We assessed the acceptability and usability of conducting home-based self-testing for HIV and blood glucose and home-based researcher-conducted testing for blood pressure in sub-Saharan Africa. We enrolled participants (≥15 years old) from households in peri-urban and rural communities of Kenya, South Africa, and Zambia. Participants opted in to self-directed testing for HIV and blood glucose and had their blood pressure measured by a research team member. Our primary measures included HIV status and testing history, HIV and blood glucose test results, blood pressure, self-reported usability and acceptability of self-testing, and participant preferences for future self-testing. Among the 526 participants from 100 households enrolled per country, the average age was 41 years and 63% were female. Overall, 16% of participants reported living with HIV. Over half of participants (52%) had last tested for HIV > 12 months ago or had never tested for HIV, and 8% of participants were unsure of their HIV status. Among participants who self-tested, 2% (n = 6/330) tested positive for HIV and 4% (n = 18/469) had high blood glucose, while 26% (n = 131/502) had high blood pressure measured by the study team. Only 14% of study participants reported previous self-testing. Most (>90%) participants who self-tested rated all procedures for HIV and blood glucose tests as either “very easy” or “fairly easy” to use. Most participants (87%, n = 458/526) preferred home-based testing. Home-based self-testing for HIV and blood glucose and home-based blood pressure measurement were acceptable, usable and preferred in peri-urban and rural areas of Kenya and South Africa; findings from Zambia suggest promise for home-based testing but were limited by staff-administered HIV tests. Self-testing has the potential to expand and accelerate access to healthcare delivery in LMICs.

Study design, sites and partners

The DASH (Diagnostic Access to Self-Care & Health Services) Study was a cross-sectional study evaluating the acceptability and usability of home-based testing via household survey and provision of self-testing tools across varied implementation contexts in Kenya, South Africa, and Zambia. Peri-urban communities with populations of 200–2,000 people were chosen within each study site, which included Migori County, Kenya; KwaZulu-Natal Province, South Africa; and Luanshya District, Copperbelt Province, Zambia.

Migori County, Kenya is a peri-urban county of 1.1 million people and faces high burdens of HIV and malaria. Umsunduzi sub-district in KwaZulu-Natal province, South Africa, has approximately 618,000 people living in urban, peri-urban, and rural areas and has a high burden of HIV, sexually transmitted infections (STIs), early pregnancy, and diabetes. Luanshya District, Zambia has 118,000 people and includes peri-urban and rural areas with a high burden of HIV, malaria, and early pregnancy.

The DASH Study used a mobile app, HealthPulse TestNowTM (Audere; Seattle, USA) to guide study participants as they self-tested for HIV and blood glucose. The app was tailored to each study site, accounting for the types and brands of rapid tests used and the local language.

Results

The research teams conducted 300 household visits, with 100 households at each study site, between 31 May and 4 August 2023, enrolling 526 adolescent and adult participants (15 years and older) in Kenya (n = 197), South Africa (n = 148), and Zambia (n = 181) (Table 1). An additional 70 participants <15 years old were enrolled in a separate arm of the DASH study but are not included in this analysis as only participants 15 years and older were offered self-testing and surveyed about their testing preferences. The household composition, including enrolled household members and those who did not participate in the study, was variable across study sites. In Kenya, households had 5 members on average, while South Africa and Zambia had an average of 2 and 3 household members, respectively. The Kenya site households had an even number of male and female members (48% female) while South Africa and Zambia had majority female participants at 62% and 65%, respectively. Occupation of participants varied across study sites. In Kenya and Zambia, participants primarily reported farming/agriculture, manual labor, and small-market sales or trade occupations. South Africa and Zambia reported high rates of unemployment, 72% and 29%, respectively.

Participant flow through each stage of the study, including enrollment, testing, answering usability questions, results and referral to care, is presented separately for HIV, blood glucose, and blood pressure testing in Fig 2.

Usability of self-administered rapid diagnostic tests

Prior experience using a rapid test at home was low — 14% across all sites (Table 2). Despite limited experience with rapid tests, more than 90% of participants across all study sites reported that the app-based instructions for HIV and blood glucose testing were “very easy” or “fairly easy” to understand. Over 90% of participants from Kenya and South Africa who self-tested, found the HIV oral swab, HIV finger prick and blood glucose finger prick tests were “very easy” or “fairly easy” to conduct themselves (Fig 3a, S1 Table, S2 Table). Participants from Zambia were precluded from self-testing for HIV and so did not answer usability questions for collecting and applying the specimen to the HIV test (S1 Table), however reported high ease of understanding HIV testing instructions, similar to the findings in Kenya and South Africa. While 90% of Zambia participants rated the self-testing instructions for blood glucose as “very easy” and “fairly easy,” most participants from Zambia opted not to conduct the self-test for blood glucose themselves (S2 Table). All participants who self-tested for HIV or who were tested for HIV by the research team, using either finger prick or oral swab tests, accurately interpreted their test results, and over 95% of participants reported that they were “very confident” or “fairly confident” with their interpretation (Fig 3b, S1 Table).

Acceptability of self-administered rapid diagnostic tests

When asked about their preferences for diagnostic testing and disease monitoring, most participants preferred home-based testing (87%) over clinic-based testing (9%) (Table 2). When accounting for possible clustering effects within households using a GEE model, estimates for participants’ preference for home-based testing were similar, 85% (95% CI 79–90%) in Kenya, 95% (95% CI: 91–98%) in South Africa, and 81% (95% CI: 74–87%) in Zambia, with an overall preference estimate of 87% (95% CI 84–90%) for home-based future self-testing.

Of participants who used an HIV oral swab test and a blood glucose finger prick test, 71% said they would be comfortable using either type of test (oral swab or finger prick) on their own in the future, whereas 23% reported they would only be comfortable using an oral swab test on their own in the future (Fig 4). For participants who used finger prick tests for both HIV and blood glucose tests, 37% reported they would be comfortable using either type of test (oral swab or finger prick) on their own in the future, with 53% of participants reporting they would only be comfortable with using a finger prick test in the future. Fewer than 5% of participants in the study population reported they would not be comfortable or were unsure if they would be comfortable conducting a finger prick or oral swab test on their own in the future.