Self-testing for COVID-19 in Durban and Eastern Cape, South Africa: a qualitative inquiry targeting decision-takers

ABSTRACT

Innovative diagnostic solutions are essential to improve COVID-19 case detection and slow its spread in resource-constrained settings. To understand how South African communities may utilise rapid SARS-CoV-2 antigen self-testing and react to self-test results, we conducted a qualitative study, involving semi-structured interviews and focus group discussions, of healthcare workers, representatives of civil society groups, and potential self-testing implementors. A sex- and location-sensitive thematic analysis approach was used to assess how 52 decision-takers on self-testing roll-out in South Africa perceive the value and utility of this innovative diagnostic approach. Informants suggested South Africans might value a device that allows them to self-test in private, at their own convenience, while avoiding the risk of social stigma and having to wait for COVID-19 test results in a facility. They also emphasised the need for awareness and sensitisation campaigns and to ensure pre-/post-self-test counselling services are easily accessible. Collaboration with traditional leaders and community-based organisations would improve results communication and linkage to counselling and confirmatory testing. SARS-CoV-2 self-testing was perceived to be acceptable to a population already sensitised to the use of HIV self-testing, being a decentralised solution that would help reduce the incidence of COVID-19 and prevent any further deterioration of socio-economic indicators.

KEYWORDS:

• South Africa
• self-testing
• COVID-19
• screening
• qualitative

Introduction

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the virus that causes coronavirus disease 2019 (COVID-19) (Khan et al., 2020). South Africa is among the countries where the COVID-19 pandemic has severely disrupted the national health system. The first COVID-19 case in South Africa was confirmed on 5 March 2020 (National Institute for Communicable Diseases, 2022). South Africa implemented a tiered system of nonpharmaceutical interventions (‘lockdowns’) to contain the pandemic; however, while this may have assisted health system capacity, the populations’ adherence to these regulations is unclear, and economic hardships were exacerbated (Moonasar et al., 2022). Co-morbid conditions that contribute to mortality from COVID-19 are highly prevalent in South Africa, including HIV and tuberculosis, which are both independently associated with 1.34- and 1.48-fold higher odds of death, respectively, following hospital admission with COVID-19 (Shabir & Jeremy, 2021).

South Africa has a history of community-grounded interventions to counteract the HIV epidemic, which has comprised efforts to prevent transmission and provide antiretroviral treatment and mass testing. To combat stigma against people living with HIV and to improve their wellbeing and self-management of the infection, there have been efforts to integrate HIV care with that of other infectious and non-communicable diseases, in decentralised, community-focused models in which nurses and other community health agents play a crucial role (Barnabas et al., 2020; Simelela, Venter, Pillay, & Barron, 2015). South Africa has also promoted ‘one-stop-shop’ approaches to integrate HIV and TB management at a single facility (National Department of Health, 2012). In this context, decentralised models of SARS-CoV-2 testing and care, integrated with responses to HIV and TB, could be welcome by societal sectors with difficulties to demand and access care.

In resource-constrained settings where healthcare delivery is decentralised, self-testing for infectious diseases presents an additional opportunity to alleviate the burden on healthcare systems, enabling responsibility for case detection to be shared with citizens. In South Africa, self-testing is already used for HIV. One study in South Africa showed that 22.3% of informants preferred HIV self-testing over practitioner-initiated testing (van Dyk, 2013). Another study in South Africa revealed that HIV self-tests could complement existing practitioner-assisted testing services (van Rooyen et al., 2015). As in other countries, HIV self-testing is perceived both as a case detection tool that empowers and capacitates the most vulnerable key populations in society and as a tool that allows public health authorities to increase awareness of HIV status among the most hard-to-reach individuals at-risk of HIV acquisition.

Self-testing for SARS-CoV-2 (hereafter, self-testing) may also represent a convenient, acceptable, and empowering tool to increase testing uptake. By enabling people to take action to detect infections early and to prevent transmission to third persons in a timely manner, self-testing may promote the reactivation of society. However, people’s values around self-testing must be understood before its promotion and distribution in South Africa. Known barriers to HIV self-testing, such as financial considerations, fear of stigma, and geographic proximity to self-test distribution locations (Harichund, Karim, Kunene, Simelane, & Moshabela, 2019) could deter South Africans from utilising self-testing. Nevertheless, factors such as testing in private, at a time convenient to the user, as reported for HIV self-testing in Khayelitsha (Martínez Pérez et al., 2016), could foster South Africans’ willingness to utilise self-testing.

In South Africa, SARS-CoV-2 testing efforts have primarily relied on facility-based polymerase chain reaction (PCR) tests (Baxter, Karim, & Karm, 2021). However, other tests, such as rapid SARS-CoV-2 antigen-detection tests (RATs), have become increasingly available (Republic of South Africa Department of Health, 2021). Self-testing is not yet regulated in South Africa; however, the population can access RATs through laboratories, and a locally produced RAT was approved in December 2021 (SADSI & SAMRC, 2021). Accessing testing as well as other forms of care has been challenging for some populations, and the economic consequences of COVID-19 have been profound (Moonasar et al., 2022). It is within this context that optimal modalities of self-testing must be explored.

To understand the general public’s values and preferences towards self-testing in South Africa, we conducted a qualitative inquiry in a rural and an urban setting which considered the insights and experiences of three groups of potential decision-takers in future self-testing programs as proxies for the general public’s wants and needs in relation to self-testing. The inquiry attempted to understand how valuable, and under what circumstances, self-testing would be for preventing the spread of SARS-CoV-2. Specific objectives were to understand the perceptions of current facility-based SARS-CoV-2 diagnostic modalities; to explore factors that may foster populations’ acceptability of self-testing; to understand what actions users of self-testing would take after receiving a reactive result; and to analyse strategies for the implementation of self-testing promotion and distribution programs.

Methods

Design, site, and population

This is a qualitative study in design that was conducted alongside a population-based survey to assess the acceptability of SARS-CoV-2 self-testing among the healthcare workforce and the public (Shilton et al., 2022). This study used semi-structured interviews (SSIs) and focus group discussions (FGDs) as data collection techniques.

The study was conducted between September and November 2022 in urban Durban (KwaZulu-Natal) and the rural sub-district of King Sabata Dalindyebo (Eastern Cape), where the implementer of the study, Advance Access & Delivery (AA&D), has ongoing public health programs, and where both AA&D and its partner university, Durban University of Technology, have relationships with local Department of Health entities. These relationships were integral to accessing health staff, particularly during lockdowns when accessing healthcare facilities was challenging.

The study population comprised three groups of stakeholders who have decision-making capacities in community or out-of-facility screening and diagnosis actions: healthcare workers (HCWs, e.g. community health workers, nurses, primary care physicians, etc.), because they can make recommendations on the use of self-testing; representatives of civil society communities (RCSs, e.g. members to the executive boards of grassroots, associations, professional councils, etc.), because they can influence various social, economic, or professional communities’ decision-making around the usability of self-testing; and potential self-testing implementers (PIs, e.g. traditional and elected community leaders, directors or general managers of factories or large corporations, etc.), because of their capacity to pool resources to procure and distribute self-testing at-scale in the workplaces they manage or the geographies where they have jurisdiction to regulate.

Inclusion criteria were being aged >17 years, willing to provide consent, and capacity to communicate in English, isiZulu, or isiXhosa. Efforts were made to ensure maximum variation sampling in terms of gender, urban and rural workplaces, and professional and institutional profile. A purposive sampling approach was used. The study teams produced sex-disaggregated lists of at least 50 profiles per each of the three study groups. These lists were randomly rearranged. Then, starting with the first name on each list, the interviewers contacted potential informants by phone or email informants and provided information about the study’s aim. Those who expressed interest in the study were invited to participate.

Data collection

Prior to any SSI or FGD, informed consent forms were shared with the informants, so they had more time to make an informed decision about participating. The informants chose the venue of data collection and decided if the SSI or FGD had to be conducted in English, isiZulu or isiXhosa. Informants who attended in-person data collection received a transportation refund. Data collection was conducted via Zoom® teleconferencing software with those who expressed time difficulties.

Data collection was led by a team of trained interviewers. The same structured guide was used for SSI and FGD and included questions around knowledge of conventional COVID-19 testing; values around self-testing; the public’s preferences for self-testing delivery; safe and effective use of self-testing; actions upon reactive self-test results (Shilton et al., 2022).

All encounters were audio-recorded and transcribed verbatim. Where applicable, responses in isiZulu or isiXhosa were translated into English. All transcripts were cross-checked against the recordings.

Data analysis

Transcripts were uploaded into computer-assisted qualitative data analysis software Quirkos®. A thematic analysis was applied. First, all transcripts were coded using a pre-defined coding scheme. Whenever emerging themes were identified, new codes were created inductively. In parallel to coding, the analysts (one American, US-based White female, and one Nigerian, Durban-based Black female, both social scientists) prepared reflexive memos to improve the study’s trustworthiness.

Iteratively with coding, the dataset was analysed using a four-stage approach: first, transcript-by-transcript, followed by a theme-by-theme sex-sensitive comparison of coded narratives across all transcripts, then by a theme-by-theme geography-sensitive comparison of coded narratives across all transcripts, and finally by a study group-sensitive comparison of key findings.

This article was prepared considering general insights as well as deviant cases’ insights. The informants’ own words were used to prepare the description of the informants’ voices. Attention was paid to the memos, to ensure that the analysts’ biases did not affect the interpretation of the findings. COREQ was considered (Tong, Sainsbury, & Craig, 2007).

Ethics

The study protocol, data collection instruments and informed consent forms received approval from the Institutional Research Ethics Committee at the Durban University of Technology.

Results

In total, 52 individuals (mean age 42.5 years; 28 females) participated in 35 SSIs and 5 FGDs; 35 informants were from Durban and 17 from King Sabata Dalindyebo. Individual interviews were conducted with 11 healthcare workers (HCWs), 14 potential self-testing implementers (PIs), and 10 representatives of civil society communities (RCSs). Two FGDs comprised four rural and four urban RCSs; two FGDs comprised three rural and three urban HCWs; and one FGD comprised three PIs. Regarding education, 6 informants had completed secondary schooling, 1 informant had received vocational training, 7 informants had received higher education or a diploma, 18 informants held a bachelor’s degree, and 20 informants held a postgraduate degree. The RCSs and PIs showed a wide range of socio-professional profiles and institutional affiliations. To ensure the informants’ anonymity, only aggregated socio-demographics are displayed in Table 1.

Table 1. Aggregate demographic data for the study informants.

Variable		Healthcare workers (HCWs) (n = 17)		Potential self-testing implementers (PIs) (n = 16)		Representatives of civil society (RCSs) (n = 18)
		Female	Male	Female	Male	Female	Male
Encounter	Semi-structured interview	6	5	7	7	4	6
	Focus group discussion	4	2	2	1	5	3
Setting (HCW and RCS)	Rural	5	3			5	4
	Urban	5	4	9	8	4	5
Sector (PI only)	Public			6	2
	Private			3	6
Age group (years)	18–25		1
	26–35	6	1	2		3	2
	36–45	2	4	3	4	2	1
	46–55	2		1	4	4	3
	56 and above		1	3			3
Education	Primary
	Secondary					4	2
	Vocational training						1
	Higher (diploma)	3	1				3
	Higher (bachelor’s degree)	3	3	4	3	4	1
	Postgraduate (masters, PhD)	4	3	5	5	1	2
For HCW only	Nurse	5	2
	Occupational/physical therapist	2
	Pharmacist	1
	Physician	2	5
For PI and RCS: Socio-professional sector	Traditional leader						1
	Commercial/for-profit sector			3	5
	Non-profit and community services			1		4	2
	Agriculture and trade					1
	Transportation services						2
	Local government official			3	1	3	1
	Academia				1	1
	Religious services						1
	Legal services						2
	Health and public health administration			2	1	0	0

Knowledge of conventional COVID-19 testing

Informants’ descriptions of currently available testing services varied. HCWs, urban RCSs, and urban PIs tended to describe testing services in greater detail and were more likely than informants from rural areas to identify antigen, antibody, and PCR tests as being available. Urban informants identified centralised testing at hospitals, clinics, and community health centres. PIs also identified private pharmacies as testing sites. Urban informants further identified ‘pop up’ (temporary, community-facing testing sites) and drive-through laboratories. Only one rural RCS, the leader of a taxi association, gave drive-through and mobile outreach as options in King Sabata Dalindyebo.

In Durban, informants identified a public–private sector dichotomy. Particularly among RCSs, testing services were described as distributed along a gradient of access, convenience, and cost, with free testing available at government facilities but with long waiting times. In contrast, private testing was considered by RCSs more available but also less convenient due to its cost. Accessing testing was considered burdensome and ‘dangerous’, given fears that SARS-CoV-2 could be transmitted while waiting in queues.

In Durban … the only options available are to go to a test center, either to government test centers, which are free and slow, or go to private ones, which are fast and expensive. (RCS, 55, male)

In King Sabata Dalindyebo, hospitals, private clinics, and consultations of general practitioners were identified as the primary testing sites available. One exception was mentioned by a male RCS who recounted how a ‘testing team’ had come to his workplace. Generally, rural informants described a high degree of trust in public-sector testing.

The barriers identified to conventional facility-based testing were diverse. The expense of COVID-19 testing was mentioned by all groups, either incurred from lost wages while waiting in queues for government testing or the cost of private testing. The cost of transportation was a barrier in King Sabata Dalindyebo, where testing facilities are geographically dispersed.

Broadly, stigma and fear were the most cited social barriers to testing. Stigma was associated with the fear of being seen testing (particularly in King Sabata Dalindyebo), of a positive result, and of disclosing positive results to contacts. Fears of being separated from family, losing income, and isolation were also cited. One element, called an ‘ostrich mentality’ by a female PI, affecting willingness to seek testing was people’s tendency to prefer not to know whether they have acquired SARS-CoV-2 to avoid the hassle of requisite actions. As this female PI put it: ‘If you don’t know [that you have SARS-CoV-2], you don’t have to deal with it.’

Barriers to access testing were drawn along socio-economic and education lines. In the rural area, informants described how COVID-19 was once considered a disease of ‘people who have money’, which has contributed to ‘finger-pointing’ toward the SARS-CoV-2 carriers and the persons who develop symptomatic COVID-19 disease. Also described as ‘apathy’ among poorer groups toward the disease, this is more likely a manifestation of general lack of knowledge about COVID-19, the test, and appropriate actions to be undertaken if an individual believes they are sick. To many, all SARS-CoV-2 prevention and testing protocols and regulations felt very difficult to comprehend. As a result, the public’s ability to adhere to them was described as uneven.

Before, there were all these restrictions on when you could go, and where you could go, and now there’s sort of drive-through concept, and people were nervous about being infected and infecting. So, it became quite complicated. I think people who have access to private healthcare were more able to access testing than people who were more reliant on public healthcare system. (PI, 52, male)

There was consensus that the health system was under-resourced prior to the pandemic. A long-lived lack of resources, including insufficient personnel and transportation for providing community-based care, was commonly cited as the reason for the system’s inability to reach everyone who requires testing in current times.

Values around self-testing

While only a few informants were familiar with self-testing, most felt that self-testing would be valuable as a case detection approach. Nevertheless, there were a few informants who were against self-testing. One male, Durban-based RCS indicated that the public health response against the pandemic would not benefit from self-testing because current surveillance systems in South Africa were poor. As one female HCWs explained, placing the responsibility solely in the individual using self-testing would not be helpful to ensure this complementary approach to facility-based use of RATs can lead to improve health authorities’ control over the SARS-CoV-2 cases that will go unnoticed:

For me one of the biggest things is: If you self-test at home, there’s nobody that is recording that you are positive. And that whole thing again about ‘Oh, I don’t want to deal with it’. So, you know you’re positive. Nobody else does. Who’s going to actually monitor your contacts and ensure that you actually then do isolate and do quarantine? If you leave the responsibility solely on the patient … (RCS, 46, female)

Despite these concerns, generally, informants felt that self-testing would provide numerous advantages, including reduced burden on healthcare workers if individuals could self-test without their assistance. HCWs felt that self-testing would help conserve diagnostic resources. One RCS, a process engineer, felt that self-testing would reduce bureaucratic ‘red tape’ involved in reporting. Should self-testing be provided for a charge, HCWs and PIs also felt that its scale-up may help to reduce the financial burden on the health system caused by the high demand of testing.

I think the advantage would be that people didn’t have to consult a healthcare worker, they didn’t have to travel to a clinic or a hospital to get the test, uhm … there they would save themselves. I suppose if it was paid for by the patient it would save the healthcare services a lot of money. (PI, 51, female)

Self-testing was understood to be a tool that promotes privacy, the lack of which was considered, particularly by rural informants, a deterrent to facility-based testing. Self-testing was considered by RCSs to promote health decision-making autonomy, much like ‘knowing your status’ with HIV/AIDS. Self-testing was viewed as particularly critical for hard-to-reach groups who are unwilling or unable to seek health services. For these groups, such as ‘homeless’ individuals or ‘people who use drugs’, self-testing was seen as in some way ‘liberating’. It meant that these hard-to-reach groups would not have to navigate institutional arrangements, which are often stigmatising and humiliating. Self-testing, then, needs to be grounded within contexts of systemic inequality and self-responsibilisation governance strategies.

Especially with the population that I work with [people who use drugs and homeless people], I think that [self-testing] will allow them the autonomy to do it wherever they’re based and not being forced to go into a facility where they [can be] discriminated. (PI, 31, female)

Informants felt that self-testing had the potential to be more acceptable as a SARS-CoV-2 diagnosis, if it were low-cost, widely available, and had shorter turn-around times for results, compared with provider-performed testing. The psycho-social benefits were also considered an advantage, as self-testing would allow individuals to mentally prepare for testing. Other benefits were the reduced risk of SARS-CoV-2 transmission in crowded facilities and an increased likelihood that individuals could self-isolate quickly.

Potential disadvantages were related to concerns about the potential lack of accountability afforded to people who would use self-testing and whether people would disclose their results, seek appropriate care, or self-isolate at the risk of lost income or social ostracisation. There were worries that widespread use of self-testing would compromise reporting and surveillance, should people be unable to report their results.

I’m not sure if they’ll give their details to surveillance mechanisms wherever they are … So if people do their own tests, we may not know who has COVID and what the numbers are. (PI, 51, female)

RCSs feared that the same barriers to facility-based testing would apply to self-testing. Financial considerations could cause difficulty in obtaining self-testing. One rural-based physician felt that poor knowledge about COVID-19 in general and the use of self-testing could lead to the incorrect use of self-tests. A few RCSs felt that false negative results could drive unhelpful behaviours, leading to further SARS-CoV-2 transmission if symptomatic individuals are not properly informed on the need to demand a RATs in their nearest clinic even if they receive a self-test negative result.

The informants considered that ‘homeless’ individuals, those living in informal settlements, people with co-morbidities, and the elderly would be the most likely to benefit from self-testing. Despite recognising that vulnerabilities would warrant intentional efforts to improve access to self-testing, informants also recognised that populations other than vulnerabilised groups might be more likely to use self-testing. For instance, it was suggested that men would benefit from self-testing, as they were understood to exercise health-seeking behaviours less frequently than women. PIs and RCSs also felt that young people may be likely to use self-testing because education rates tend to be higher in this demographic. RCSs’ opinions also reflected expectations for health behaviours, saying that ‘responsible’ people would be more likely to use self-testing, even associating a propensity toward certain behaviours as a reflection of ones’ character:

The disadvantage is that the person that is self-testing, we don’t know their character … how would they react? Maybe the test will say positive, and the person decides to walk around and that would increase the transmission of the virus to be something we don’t know. There are situations like that, but if people can be trained, maybe that would be helpful. (RCS, 40, male)

It was felt that individuals who would be less likely to use self-testing are those who are ‘complacent’ about COVID-19, those who ‘passively accept’ the situation, and those who deny the COVID-19 pandemic. While it may appear that this contradicts views of self-testing as ‘liberatory’, this is not the case. For those who are described as ‘unconcerned’ or ‘complacent’ about the COVID-19 pandemic and its impact, self-testing has little relevance since they are thought to be unlikely to test at any rate. Nevertheless, a few informants opined that access to self-testing may be a mechanism for increasing the reluctant persons’ uptake of SARS-CoV-2 testing if symptoms are evident.

Preferences for COVID-19 self-testing

Saliva samples were the preferred sample for self-testing, although HCWs felt that throat or nose swabs would produce the most accurate results. However, the swabbing process was thought unpleasant, and it was suggested that self-testers might collect swab samples incorrectly.

The trustworthiness of the distributor was an important consideration to facilitate uptake. Compared with Durban institutions, rural healthcare institutions were considered more trustworthy for distributing self-testing. One PI, a male public service administrator, felt that NGOs are not trustworthy. An RCS, the female head of an academic department, stated that people in Durban would not trust door-to-door initiatives delivering self-testing. In contrast, many rural RCSs felt that religious leaders and chiefs would have sufficient rapport with the population to encourage follow-up activities by self-test users. Many informants referred to local community leaders as the appropriate individuals to provide education about and even administer self-testing.

I think people do trust their pastors, their healthcare workers, nurses and general practitioners, pharmacists, principles … They generally don’t trust politicians. So, I wouldn’t include them there. But, generally, the community leaders, non-politically aligned, I think would be people that would be trustworthy. (RCS, 60, male)

Regarding locations to obtain a self-test, informants generally agreed that self-tests should be widely available, including at clinics, pharmacies, and hospitals. Grocery stores were also considered suitable, although in non-healthcare settings some informants felt that health authority representatives should be present to advise individuals on follow-up actions to take if their self-test was positive.

Informants considered that self-testing kits should contain guidance about their proper storage, use, and disposal; how to interpret test results; and what to do upon receiving a positive result. The instructions should include illustrations and be provided in all official South African languages. Three HCWs considered the instructions should explain that, in the presence of COVID-19-compatible symptoms, a negative result does not necessarily indicate that the user is SARS-CoV-2-negative.

Let’s got the presumption that it’s 90% sensitive, saying that in 10 people 1 people will be missed. So: ‘please do this and if it’s positive, it’s positive. But if it’s negative and you have ongoing symptoms for another 2 or 3 days, please access a health care provider’. (HCW, 37, male)

High costs for self-testing were considered a potential barrier to uptake, particularly for economically vulnerable groups who already experience barriers to seeking traditional testing. While some informants felt that self-tests should cost less than 200 South African rand (R) (∼13 USD), others felt they should cost no more than a pregnancy test or an HIV self-test (R20–50, 1.3–3.3 USD). It was recommended that homeless persons or individuals receiving social grants should have access to subsidised or free self-testing.

It was felt that most people would want to self-test at home. HCWs indicated this would be essential to reduce the influx of patients to healthcare facilities. Relatives and friends were identified as people that self-test users would be most likely to trust if they needed assistance with self-testing. The majority of HCWs, and a few PIs and RCSs, felt that some individuals may also trust trained health professionals to provide assistance.

Potential actions following COVID-19 self-testing

Informants’ responses regarding the provision of post-test counselling varied. RCSs and PIs indicated that the public would prefer to receive counselling from healthcare workers. In contrast, the HCWs suggested that, as medical care is not necessary for many SARS-CoV-2-infected individuals, non-healthcare professionals could staff a counselling hotline. HCWs also felt that staff who distributed self-tests, such as pharmacists, could provide counselling. In settings such as King Sabata Dalindyebo, where health professions are lacking and there is a strong platform for community outreach, community health workers would be trusted to provide counselling. Local community leaders and healthcare providers were also seen as having a symbiotic relationship and would be able to cross-refer according to the need emerging from a positive test result:

I think they could want [guidance] from people who stand for people in the community. For example, from the Chief. The Chief has his own way of guiding his people … For example, from the Chief he/she can ask for advice. Then if the Chief doesn’t have the right words to say, he will refer the person to the clinic. (RCS, 37, female)

Regarding methods for counselling, informants tended not to agree on a single practice. Hotlines were identified as an acceptable counselling method. Electronic tools, such as apps, QR codes, and links on social media, were considered feasible options to provide information about where to seek confirmatory testing and care. However, there were concerns that rural and older users would be unable to use digital tools.

Information campaigns were thought necessary to increase public awareness of self-testing availability, to build knowledge of how self-testing should be used, and to instruct the public on what actions should be performed following a reactive result. Radio, television, social messaging, and social media were commonly recommended means for spreading this information. However, some HCWs and RCSs were adamant that self-testing be delivered alongside information campaigns.

The possible harms caused by a positive self-test were considered to mirror the harms that could follow a positive provider-initiated test. Potential negative consequences of a positive self-test included anxiety, fear, and shock. One rural RCS feared that the psychological impact of a positive self-test may lead some people to self-harm. Negative responses were expected to be particularly severe among people who had lost a loved one to COVID-19 and those with underlying health conditions. Social stigma resulting from a COVID-19 diagnosis was also described. Beyond the psychological impacts, informants identified significant economic consequences of a positive self-test, in the form of lost income or dismissal from one’s job.

Likelihood of communicating results and isolating

There was agreement that the public currently struggles to disclose their test results and self-isolate with traditional testing. Hence, there was disagreement whether a positive self-test would lead to any behaviour changes compared with current testing forms. Informants were unclear whether individuals would report their results or self-isolate following a positive self-test. Several RCSs and PIs expressed frustration that shortcomings in the healthcare system may render the communication of positive self-test results to health authorities challenging.

HCWs tended to feel that ‘responsible’ people would report their results and disclose them to their contacts, but this would also be dependent on the perceived consequences a positive self-test would have for people’s finances and employment. They also felt that symptomatic individuals would be more likely to report. A male, rural physician described how the actions required for a positive self-test may compromise an individual’s freedom and how fear of this consequence could deter people from reporting. In contrast, a rural, male, traditional leader felt that people would readily communicate with clinics because they have experience communicating with healthcare providers for other diseases.

People don’t like any of their freedom to be taken away from them. People feel that if they have to isolate, their freedoms are being taken from them … They don’t want that information to be spread to the general population of the clinic because it may cause people to ask of them to stay at home … (HCW, 37, male)

As with reporting self-test results, informants were unsure whether people would disclose their results to contacts, whether due to fear of stigma or being blamed for another person’s illness or death. Several informants indicated the need for community education campaigns aimed at destigmatising COVID-19 and informing the public about resources available following a positive self-test.

Informants felt that individuals’ willingness to isolate would be determined by whether they would lose income should they take sick leave. RCSs believed that isolation would be seen by some as a way to protect ones’ family and friends, and therefore people would be willing to isolate. Some RCSs and PIs recognised the challenge of self-isolation in densely populated homes. A rural-based, female RCS, a facilitator of community development programs, thought the roll-out of self-testing should be accompanied by services that allow people to self-isolate, while a female, Durban-based RCS, a health education officer, believed that these services should be available at minimal cost.

Overall, informants felt that it was unclear whether or how self-testing could be delivered to facilitate self-isolation, disclosure, contact tracing, and linkage-to-care following a positive self-test result. Several recommendations for supporting these actions involved the creation of a register in which self-test users’ information could be collected and which could be used to follow-up individuals who purchase a self-test. Electronic tools, such as QR codes included with self-testing kits or hotlines via WhatsApp, were recommended as alternatives to a paper register physically located at sites of self-test purchase.

Hotlines. So, if your test result says red, which means stop and isolate, uhm, WhatsApp lines, the Facebook messages, the QR code, will tell you what’s the next step. Because lots of people don’t know what’s next. (RCW, 41, female)

Discussion

This qualitative study suggests that South Africans would accept self-testing as a complement to provider-initiated SARS-CoV-2 testing to increase the detection of SARS-CoV-2 carriers and decrease COVID-19-attributable morbidity and mortality rates in their communities. Roll-out of self-testing was largely considered an opportunity to increase access to testing, especially among vulnerabilised groups but also among all men and young persons, should health authorities-endorsed affordable and easy-to-use self-testing devices be made widely available.

Self-testing, as per our informants’ narratives, would empower local rural and urban inhabitants with knowledge and personal experience in the use of novel diagnostics as technological means to invest in their own health and in the wellbeing of their communities. De-medicalizing their experiences with self-testing could shift public focus from lack of trust in the public’s sense of responsibility towards the pandemic to the potential in educating the public in the use of self-tests to fight off the COVID-19 pandemic and future infectious diseases outbreaks in the country. Self-testing, beyond its advantages for public health and to decongest health facilities country-wide, has the potential to serve as an educational tool to fuel societal harmony, sorority and brotherhood, and transform the ways in which local communities are engaged in health emergencies response since their very outset in South Africa.

Additionally, self-testing creates circumstances that would directly benefit vulnerabilised populations who cannot otherwise access facility-based SARS-CoV-2 testing in many South African settings. The direct cost of tests as well as indirect costs, such as transportation and lost wages, disproportionately burden low-income and vulnerabilised populations. Coincidentally, some of the groups identified as ‘vulnerabilized groups’ in this study are groups considered as at-increased risk of seeing their rights to TB and HIV detection and care neglected in South Africa. Consequently, by recognising the complex dynamics within educational, socio-economical, and the knowledge and experience generated by decades of community TB and HIV programs, a more nuanced understanding of SARS-CoV-2 testing needs within specific communities which could potentially benefit from the introduction of self-testing can be gained.

In our study, self-testing was described as an optimum form of screening for SARS-CoV-2 infection that would promote individuals’ privacy and autonomy, particularly given fears of stigma and limited agency in health decision-making, which were important for people residing in close-knit communities. Should self-testing be delivered in ways that are acceptable to the general population and within a system that enables individuals to undertake appropriate responses to a reactive self-test result, informants indicated that self-testing could be an important component of a strategy to reduce community transmission of SARS-CoV-2. Given the temporality of data collection, it is important to consider how the role of community expectations related to COVID-19 detection may have changed as the epidemiological profile of the disease has changed towards endemic patterns, and social pressures and stigma have become less acute.

The roll-out of self-testing must involve broad grassroots, civil society organisations and public institutional engagement. In Durban, mistrust in government institutions is an important consideration, although trust in healthcare providers remains steady among the public, while trust in non-profit organisations varies across informants. This ambivalence may be the result of two characteristics of urban life: urban clinics tend to experience high patient volumes, requiring extended wait times, and urban populations are generally more critical of services. In contrast, rural informants tended to trust local authorities and existing health systems, which have a history of community outreach with the support of, normally, traditional, religious, and women and youth leaders. These varying degrees of trust in institutions must likely be translated to differentiated modes of delivery. For example, in Durban, it may be more acceptable for self-testing distribution to involve healthcare professionals, whereas in areas such as King Sabata Dalindyebo, distribution may be assisted with the support of existing, trusted community leaders and non-profits with established community outreach platforms.

Traditional leaders may play a role in shaping attitudes related to actions following self-testing usage, although this relationship requires further anthropo-epidemiological explorations. Regarding traditional leaders, communities’ chiefs should be engaged throughout the self-testing introduction process, particularly in education campaigns. These leaders may be important facilitators in increasing the acceptance of self-testing. Traditional healers also retain a high degree of trust, even among educated populations. The roll-out of self-testing should be complementary – rather than antagonistic – to traditional practices to detect a disease that is harmful to the person’s physical and spiritual, and to traditional cosmogonies of health and disease.

Beyond South Africans’ preferences for self-testing, it is important to consider how their values around self-testing would affect their likelihood of using this approach. Questions of potential economic harm and psychological impact caused by a reactive self-test are perhaps best considered in the context of high unemployment rates and a shrinking economy in South Africa, both of which were exacerbated by the lockdowns in 2020 (Posel, Oyenubi, & Kollamparambil, 2021). These factors may decrease individuals’ willingness to self-test, report results, and self-isolate, due to fears of risking a loss in income or employment. For individuals whose income is earned in the informal economy or who rely on hourly wage labour, fears of further economic damage are particularly acute (Thulare & Moyo, 2021).

Some informants suggested that people who would use self-testing are self-selecting, with those choosing to self-test being more likely to undertake further action following a positive result. Others suggested that the likelihood of undertaking these behaviours falls along socioeconomic lines, with those who have the resources and social capital to undertake these behaviours being the most likely to do so. Some informants stated that individuals would use self-testing but would prefer not to report their results, choosing instead to treat their symptoms and continue working for fear of losing income. These opinions reflect the high rates of unemployment in South Africa, the reliance on income from work in the informal labour sector, and the lack of worker protections for those in the informal economy. Together, these factors translate to a high percentage of the population who may be unlikely to use self-testing or report self-testing results should they understand it to impose undue risk to their income.

The history of HIV/AIDS in South Africa also likely informs individuals’ concerns around stigma, isolation, and loss of autonomy (Skinner & Mfecane, 2004; Visser, Makin, Vandormael, Sikkema, & Forsyth, 2009). Rather than being compelled to self-isolate, some individuals may prefer to treat their symptoms at home until there is no other choice but to seek care. The idea of testing for SARS-CoV-2, and of being positive, still resonates strongly with experiences of HIV. The collective trauma of the HIV/AIDS crisis has not yet healed, and its stigma still affects a significant portion of the population (Naidu, 2020). Parallels were drawn between the history of stigma around HIV/AIDS in the country and the social consequences of receiving a reactive self-test for SARS-CoV-2. Both experiences hold a foundation of neglect by systems of power that lead to publicly enforced stigma, harm and social isolation. Similarly to HIV, COVID-19 put the brunt of the responsibility of safety on the hands of the public. The failures of health systems and governments exist in both realities of COVID-19 and HIV. Yet, in time, both narratives shifted focus onto the importance of self-empowerment as an individualised experience fostered through knowledge, testing and prevention measures. Therefore, it is critical that self-testing is not perceived as a risk to individuals’ income or ability to make choices. Conversely, ‘knowing your status’ as it relates to SARS-CoV-2 infection could be a message that resonates with individuals and encourages testing, although it is important to consider the inherent differences in infectiousness of people living with HIV, for whom infection is in almost all cases lifelong, compared with the infectiousness of people with SARS-CoV-2, whose window of infectiousness is temporally constrained. Emphasising the risks to the community from one individual infected with SARS-CoV-2 may also encourage self-testing.

It must be considered how self-testing will be integrated with conventional forms of testing, with the existing public health system, particularly for people at high risk of severe COVID-19 disease. Confirmatory PCR or RAT may be necessary for subsets of the population who receive a non-reactive self-test and are experiencing symptoms. It is also important to consider the added value or unintended costs of self-testing in the absence or presence of COVID-19-like symptoms or risk of possible exposure. Best practices for integrating self-testing into existing diagnostic and care pathways should undergo further research.

A supportive system that enables individuals to undertake the requisite actions following a positive self-test is crucial. We identified concerns about whether self-testing can facilitate improved access to care, ability to isolate, and confidence to disclose self-test results to contacts. The lack of oversight for self-testing and lack of centralised, systematic guidance for people receiving a positive result were among the reasons some informants would be hesitant to recommend self-testing. However, self-testing may be a core component of, and in some ways even facilitate, a decentralised approach to COVID-19 care. This would involve the deployment of non-healthcare professionals to provide information and assist with linkage to care, training for healthcare workers to provide counselling to individuals who self-test positive, and, with the support of local NGOs and media, facilitate community awareness. For example, peer supporters and educators could be trained by the Department of Social Development and the Department of Health, in partnership with NGOs and universities, to promote self-testing in remote communities and among the most vulnerable. Peer support and home-based care imperatives could provide new pathways for upskilling and employment in the form of public works. A coordinated and supportive response may in fact have the added benefit of rebuilding public trust in government institutions.

This investigation had some limitations. First, data collection was performed in the urban setting of Durban, KwaZulu-Natal and in the rural setting of King Sabata Dalindyebo sub-district, Eastern Cape. We sought to include diverse informants, and these localities are typical of urban and rural settings across the country; however, we acknowledge that perspectives reflected here may not be representative of the entire country. Second, given that SARS-CoV-2 self-testing was not available in South Africa at the time this study was conducted, most informants were unfamiliar with self-testing. Limited familiarity with self-testing is reflected by the varying degrees of feasibility observed among the recommendations for implementation and delivery of self-testing that informants offered. Finally, data were collected and analysed in July 2021, a time following the significant social upheaval in South Africa (i.e. the ‘Zuma riots’) (Makoni, 2021), after high case numbers of SARS-CoV-2 infections and before the onset of a new epidemiologic wave of infections caused by the SARS-CoV-2 Omicron variant. The general turmoil in South Africa at the time of this research might have increased caution and concerns.

Conclusion

South Africans are generally positive about the concept of self-testing for SARS-CoV-2 infection. For any roll-out of self-testing to be successful, trusted stakeholders, institutions, and community members, including lay people, will be required. These include government institutions, local pharmacies, private healthcare providers, and public health NGOs. In planning this roll-out, the scepticism that urban South Africans have toward government systems and institutions, particularly regarding the COVID-19 response, must be addressed.

Focus should be placed on young people and vulnerable population groupings, as they have expressed concern about the various information platforms they are exposed to and their confusion about what information they should be paying attention to. Taking guidance from previous public health campaigns, civic groupings need to partner with the government in the implementation of these campaigns. The possibility of self-testing offers new opportunities for South Africans; a prospective way of increasing faith in the government and the reach of public health initiatives. In a country plagued by extremely high unemployment rates, low levels of trust in government, and insufficient healthcare-seeking behaviours, SARS-CoV-2 self-testing offers real transformative potential for a range of public goods.

Ethics approval and consent to participate

Ethics approval was obtained from the Institutional Research Ethics Committee at the Durban University of Technology (Ref. IREC 165/21). All informants provided written informed consent. All methods were carried out in accordance with relevant guidelines and regulations.

Authors’ contributions

GZMP and SS developed the initial research project. MWW, ANB, NX, GBB, MMM, JKA, and KL adapted the research protocol. GBB, KL, and MWW led the implementation of the study in South Africa. ANB and GBB performed the data processing and analyses. ANB, GBB, JT, and GZMP wrote the manuscript. All authors have reviewed the final version of the manuscript.

Acknowledgments

The authors would like to thank all of the informants in KwaZulu-Natal and Eastern Cape, South Africa, who contributed to the interviews and focus group discussions. The authors thank Adam Bodley for editing the manuscript.

Data availability statement

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was funded by the Government of Germany. The funder played no role in the study design; in the collection, management, analysis, or interpretation of the data; in writing the report; or in the decision to submit the report for publication

Notes on contributors

Amanda N. Brumwell

Amanda N. Brumwell is the managing director of Advance Access & Delivery.

Gbotemi B. Babatunde

Dr Gbotemi B. Babatunde served as the project manager for the COVID-19 ST study through a public health NGO – Advanced Access and Delivery, South Africa.

Sonjelle Shilton

Sonjelle Shilton is currently working with FIND, Global Alliance for Diagnostics (Switzerland) as Deputy Director of Operational and Implementation Research. She is the Lead of the multi-country Covid-19 Self-testing Values & Preferences research.

Jade Tso

Jade Tso is Program Manager at Advance Access & Delivery.

Michael W. Wilson

Michael W. Wilson serves as a director and co-founder of Advance Access & Delivery, and is Adjunct Assistant Professor in the Department of Health Behaviour at the Gillings School of Global Public Health.

Noeline Xulu

Noeline Xulu works as Programme and Development Director for Advanced Access and Delivery.

Jamila K. Adam

Professor Jamila K. Adam is the Chairperson of the Institutional Research Ethics Committee at the Durban University of Technology (DUT), and she is also appointed as an Honorary Research Professor in the Faculty of Health Sciences at DUT.

Monique M. Marks

Professor Monique M. Marks currently heads up the Urban Futures Centre (UFC) at the Durban University of Technology. The UFC, together with Advance Access and Delivery, manages the Bellhaven Harm Reduction Centre.

Guillermo Z. Martínez-Pérez

Dr Guillermo Z. Martínez-Pérez works with FIND, Global Alliance for Diagnostics (Switzerland) as senior social scientist in a range of diagnostics acceptability studies in low/middle-income sites.