Background
In South Africa, 12 % of the population (6.4 million) are living with HIV [
1]. The AIDS epidemic is generalised with primarily heterosexual transmission [
2]. Overall HIV prevalence is highest among females (14 %) compared to males (10 %) and among urban informal dwellers (20 %) compared to those living in rural informal areas (13 %) [
1]. More than half (56 %) of the burden lies within the poorest 40 % of the population [
3]. Around Cape Town the antenatal prevalence ranges between 9 and 37 % across health sub-districts [
4]. Access to HIV services needs to be in line with demographic and socioeconomic factors.
HIV counselling and testing (HCT) is a fundamental component of HIV care and prevention. In 2012, 65 % of South Africans reported to have ever tested for HIV of which 66 % tested in the previous 12 months [
1]. Although indicative of fairly good testing coverage, it falls short of the Government’s call for all to test annually [
5] and many individuals, including those most likely to be living with HIV, still do not know their HIV status [
6]. Public health facilities cannot test everyone as some populations for example males do not readily access public health facilities [
7]. In striving toward universal access to HCT, a better understanding of the factors that enable and constrain access to HCT is important and diverse testing opportunities should be considered in an attempt to increase access to HCT.
“Access” is an important concept in health services research [
8]. However, it is a multidimensional and complex term [
9] that has not been precisely defined [
10] nor measured in a consistent manner [
9]. Although it is widely accepted that access to health services, refers to the opportunity to use a health service [
3], utilization of health services is only one indicator of access having been realized [
11] and is insufficient on its own to determine access. A broader interpretation includes utilization of a health service being dependent on three dimensions of access that are conceptually clear and separate [
12]; availability, acceptability and affordability [
13,
14]. Within this framework availability is concerned with health services being supplied in the right place and time to meet the needs of the population [
12], acceptability is concerned with the provider and patient attitudes and expectations of each other and affordability measures the relationship between the cost to the patient to use the service and their ability to pay [
14].
This study focuses on the dimensions of availability and acceptability and investigated reasons why clients choose to have an HIV test, why they chose either a public funded primary health care facility or a non-governmental mobile facility as service provider and compared their experiences of HCT. As HCT is offered free of charge at both service providers, affordability was excluded from this study. This study therefore compares participants who accessed either one of two HCT service types.
Methods
Setting
In the Cape Town district of the Western Cape Province of South Africa, HCT is routinely provided at publically funded primary health care facilities, district hospitals, at secondary and tertiary level care. There are approximately 142 public primary health care (PHC) facilities offering free HCT services. HCT services are also provided free of charge through non-governmental organizations (NGO).
Since 2008, the Desmond Tutu TB Centre (DTTC) at Stellenbosch University has worked in partnership with non-governmental organizations to establish Community HCT centres in 8 communities around the Cape Town metropole in the Western Cape Province of South Africa. These services are provided either from community-based centers or on a mobile basis, whereby non-permanent structures are set up in appropriate spaces within communities to specifically target groups that do not typically access facility-based services [
15].
Stellenbosch University provides overall management, technical assistance and quality assurance and contracts local NGOs to provide HCT services from either the community HCT centres or on an outreach, mobile basis. Mobile services are provided from pop-up tents and a caravan (mobile van) set up in an appropriate open space, including public transport hubs such as taxi ranks, train stations or open fields.
This study was embedded within a larger study, which compared the characteristics of clients who accessed HIV testing across public PHC facilities and NGO mobile services [
16]. This larger study took place in the above-mentioned eight communities on the outskirts of the city of Cape Town. These communities are under-developed, densely populated, have a mixture of formal and informal dwellings and are characterised by high unemployment, socio-economic challenges and a high HIV and TB burden.
The larger study mapped a geographical area within each of these 8 communities, by plotting a 2 km radius around each of eight pre-existing NGO community-based HCT centres. All public PHC facilities that fell within this 2 km radius were included in the study. In each of the 8 areas, between one and four public PHC facilities fell within this 2 km radius. There was no overlap between areas. The study focused on comparing the HCT services of mobile funded NGO providers and public PHC facilities within these areas. Services were comparable in terms of operating hours and neither provider type charged a fee.
Design and sampling
This is a descriptive qualitative study. For the purposes of this study, two of the eight geographical areas were purposively sampled to ensure the inclusion of an ethnically diverse set of respondents. In area one, 1 NGO mobile service and two public PHC facilities were included and in area two, 1 NGO mobile service and 4 public PHC facilities were included.
This study was designed to specifically compare issues around availability and acceptability of HCT services at two existing HCT service types, as reported by those who accessed HCT. Therefore sampling included only those who accessed HCT at either of these two service types.
A total of sixteen adult participants (>18 years) were enrolled in the study. Lists containing the participant’s barcoded study identity number from the larger study were used to systematically sample every tenth participant from those who had attended the mobile services and each PHC facility within the selected areas until 8 participants were enrolled from each service type. In area one, 4 participants were sampled from the mobile and two from each of the two PHC facilities. In area two, 4 participants were sampled from the mobile and one from each of the four PHC facilities. Including participants from different facilities allowed for client experience across several health facilities to be explored.
Each sampled individual was contacted telephonically and a date set for the interview. If the individual was not able to be contacted or declined to be interviewed, the next individual on the list was contacted until for each area, the required number of participants gave permission to be interviewed. In order to obtain permission from 8 participants who had attended the NGO mobile services, 24 were sampled from the lists of which 15 were not contactable (no answer, had moved or had missing contact details) and 1 declined. In order to obtain permission from 8 participants who had attended the public PHC facilities, 19 were sampled: 10 were not contactable and 1declined.
Data collection
Semi-structured interviews were conducted between March and June 2011, approximately 1 month after the participants’ HIV testing experience. One male and two female interviewers, with experience in qualitative interviewing and from similar ethnic groups to the participants did the interviews in the preferred language of the participant (Afrikaans, isiXhosa or English). To minimise potential bias during data collection the interviewers were not part of the larger initial study and were not aware of the participants’ HIV status.
All interviewers were trained on the overall objective of the study and in using the interview guide as a tool. This guide comprised 3 global topics; participant demographics, health seeking behaviour and HCT experience. The semi-structured interview tool was developed in English and was not translated into any other language. The interviewers could speak and understand English well. The training they received allowed them to translate any questions if required as they conducted the interviews. Interviews were conducted in the communities, predominantly in participants’ homes or a private space elsewhere, including in the interviewer’s car. The decision regarding where to hold the interview was a mutual one between interviewer and participant based on safety and privacy. Interviews varied between twenty and sixty minutes, were digitally recorded and transcribed by each interviewer. Interviews conducted in a language other than English were translated during transcription.
Analysis
Each interview transcription was read and reread by SM, to ensure familiarity with the content. A constant comparative analysis was done to identify common and divergent responses in relation to the key questions and categorised and abstracted to form common themes. A sample of four transcripts (two from PHC facilities and two from mobile services) were independently analysed by a co-author (NL). There were no discrepancies in the analysis between the two researchers. Those involved in data analysis were blinded to participant HIV status.
Two identical matrixes, one each for NGO mobile and public PHC facilities, were developed in Microsoft Excel 2010. These were used to categorise and compare data across provider types. Health seeking behaviour and HCT experience were entered as pre-determined global themes in the first column. The main interview questions related to each of these global themes. Participant responses to these global themes were organised around sub-categories, which were entered underneath the relevant global theme. These sub-categories were aspects of the global themes that the researchers were interested in and formed minor questions (prompts) on the interview schedule, meaning that they were only used if the participant provided insufficient detail when responding to the main interview questions. Within each of these sub-categories, responses were compared across participants and themes emerged within each sub-category. The matrix is illustrated in Table
1. A general comments column was included to note consistent and divergent responses across HCT provider types.
Table 1
Illustration of the data analysis matrix
Reason for testing | Opportunity | “…on the way we saw the tents on the side of the road with a sign that says HCT”. | | |
Affected by HIV | “I have a sister in XXX who is HIV positive…” | | “have a cousin who is HIV positive…she died in January..” |
Perceived personal risk for HIV | | “… I had a boyfriend whom I did not trust, and then I thought for my sake I needed to know my status”. | “My boyfriend doesn’t want to use a condom…” |
Ethics approval
The study was approved by the Health Research Ethics Committee of Stellenbosch University (N10/09/288) and The International Union against Tuberculosis and Lung Disease Ethics Advisory Group (EAG 58/10). All participants were part of the larger study and had already provided written consent, which included the possibility of being contacted for face to face interviews. When contacted telephonically, participants could decline to be interviewed and those interviewed were free to end the interview at any point. Participants were not given any incentives for taking part.
Discussion
This study compared client health seeking behaviour and experiences at two types of HCT services; those provided by non-governmental organisations that provided mobile services and public primary health care facilities, offering HCT services from fixed sites.
Irrespective of where participants obtained their HIV test, health seeking behaviour was related to the opportunity to test, being affected by HIV and a perceived personal risk for HIV. These findings are consistent with other studies in sub-Saharan Africa which showed that having an HIV test was associated with knowing the location of a test site [
17,
18], personally knowing someone with HIV [
17,
19,
20], and a perceived sense of being at risk for HIV infection [
17,
18,
21,
22].
The concept of availability is concerned with an adequate supply of services [
9] and the opportunity to obtain an HCT service by those who need it when they need it [
23]. This study has highlighted that opportunity to obtain an HIV test does play a role in the utilization of HCT services. Providing an HCT opportunity, whether through a mobile service provided at a busy transport hub or an integrated health service at a public health facility, makes HCT physically available. Mobile services appear to have ‘opportunistic’ clients compared to health facilities, as they are able to attract those who are just walking past and who would otherwise not have tested for HIV at that particular time. Mobile HCT plays a role in increasing access for those who were not contemplating testing until this opportunity arose.
Opportunity to test is closely aligned with accessibility, which emerged as an important consideration for participants regardless of which service they attended. Participants who attended the NGO mobile services spoke about “just passing by”. The accessibility of mobile services may lie in the increased proximity of the testing site to the surrounding community; reducing travel time and cost for those who do not stay close to a clinic. This is in line with studies in sub-Saharan Africa which have shown that distance to testing site and travel cost are barriers to HIV testing [
19,
24,
25]. Those who attended public PHC facilities reported that the facility was in close proximity to where they live. Health facilities are able to provide HIV testing through integrated health care, for example reproductive health services, making HIV testing very accessible to those who attend these services. For these clients, there is no additional travel time or cost to acquire an HIV test.
The convenience of having physical access to the opportunity to test for HIV played a role in participants’ decisions to test for HIV. The accessible opportunity that mobile HCT offers should be considered when aiming to increase access to HCT overall, as mobile HCT can provide services to those who are passing by and would not have accessed an HIV test at that particular time. Currently, mobile services are limited in the number of related health services they offer compared to public primary health care facilities. Future research could determine if a wider variety of health services at mobile HCT could attract a higher number of people.
Improving availability of an HCT service requires consideration around the manner in which the service is organised to meet the needs of the client. Within this study, waiting time emerged as a differentiating factor between the two service types. Longer waiting times were consistently reported as a source of dissatisfaction in health facilities and may be potential barriers to access. Many South African studies have reported client dissatisfaction with long waiting times with other health services [
26‐
29]. Shorter waiting time encouraged the use of NGO mobile services and may therefore provide an avenue to reduce the time at HCT and thereby expand uptake. However, a higher utilisation of mobile services may put a strain on waiting times. It is also important to consider that shorter waiting times is not a unique feature of the NGO mobile service and could also be achieved within a public clinic setting by fast-tracking clients who only want to access HCT services.
Availability is just one dimension of access. Understanding client experiences is also important for developing strategies to strengthen health systems and improve access to services. Acceptability indicates a match between service provision and the expectation of the user [
23]. Although acceptability is critical to ensuring that an individual uses a service [
14], it is a subjective concept, heavily dependent on the user’s expectations. Although this study did not explicitly measure satisfaction, the reported experiences indicate levels of satisfaction, which determine future utilisation of health care services and are ultimately linked to access [
27].
Participants reported overall positive experiences irrespective of where they accessed services. Staff attitude, competence and trust emerged as factors influencing acceptability. Participants who accessed mobile services experienced friendly competent staff that they felt they could trust with keeping their HIV results confidential. Reports regarding the friendliness of health facility staff were contradictory and instances of poor staff demeanour were noted. While some South African studies reveal high levels of satisfaction with public health care providers [
26,
30], others have reported that staff did not treat patients with sufficient respect [
27,
31]. Anticipated disrespectful treatment has been shown to be partly accountable for delayed care seeking [
13]. Conduct of nurses is a core element by which clients judge health services [
32]. Although this study did not identify reasons for poor staff demeanour, shortage of staff and management, inadequate resources, high workloads and stress and burnout have been identified as major challenges faced by nurses [
33,
34] and may have played a role in the cases where participants reported incidents of poor staff demeanour.
Healthier working environments, with adequate resources and reduced stress have been linked to positive client experiences [
35]. Interventions that strengthen management capacity [
36], ensure teamwork among healthcare providers [
37] and enhance trust between health care providers and patients [
38] are required for a healthier working environment. Within the NGO mobile HCT services, professional nurses are employed in a management capacity and receive continual training in performance management skills. This includes developing skills in data evaluation, which allows the professional nurses to better understand their HCT data, identify gaps in the services and generate and implement plans that address these gaps. In this manner professional nurses are able to provide strategic direction to their teams. General staff wellness is addressed through regular debriefing sessions. These sessions are hosted bi-monthly by psychologists and social workers and are attended by the entire mobile HCT team. The sessions provide psychosocial support and in addition aim to develop communication and teamwork. These interventions may have played a role in the friendly and competent service reported by participants who accessed mobile HCT.
Participants did not have concerns about the cleanliness of the HCT setting, but were concerned about the risk of being stigmatized in both the health facility and mobile settings. This aligns with previous work in high HIV prevalence settings that showed that concerns about stigma did not differ between HCT services in integrated or non-integrated health facility settings [
39]. In health facilities, perceived stigma was associated with overcrowding and lack of private spaces, whilst at the mobile service; it was due to the public placement of the tents. Further investigation into the experience of stigma and ways to reduce stigma is required to help inform health authorities on ways to limit real or perceived stigma associated with HCT.
This study compared two HCT service providers across two dimensions of access; availability and acceptability. Both dimensions are important considerations when aiming to increase access to HCT services at these health service providers. Future operational research could explore the dimensions of access further within the South African context. Research is required to assess the feasibility of government outreach services, and whether these can increase availability of services, by accessing different populations to those presenting at health facilities. Future studies are also needed to determine the impact of interventions designed to produce a healthy working environment. Randomized control trials could be utilised to determine if such interventions impact on the acceptability of health services within this context. Affordability is a third dimension along which access is measured and should also be taken into consideration in future studies.
Strengths and limitations
The study was conducted within a limited geographical area, which limited the number of PHC facilities and mobile services included. The sample of participants was relatively small. However, the findings align with those in large representative sample surveys [
31] and provide significant insight into individual perspectives and circumstances motivating HCT utilisation and provider choice.
The translation process was not checked for quality and no back translation was done, which may have limited the quality and depth of the English transcription - some nuances in the language use of patients may not have been sufficiently captured. Not translating the interview guide may have limited the consistency of the translation of questions between interviewers, but all interviewers spoke and understood English well.
The length of the interviews varied, with half of the interviews taking 20 to 30 min, and half 30 to 60 min. This is directly attributed to the amount of probing and may have limited the depth and quality of the data collected. Whilst all interviews addressed the main guiding questions, depth and quality of data was potentially limited in shorter interviews.
The study only takes into consideration the health seeking behaviour of those who had an HIV test. In order to better understand barriers to accessing HCT, future studies should include participants who have never tested.
The study does provide a unique comparative analysis of NGO mobile services and public sector primary health care facilities providing HCT services within a specific geographical area that is representative of densely populated, low socio-economic urban settings, where the highest levels of HIV infection are found. It also highlights some of the key issues that affect utilisation of HCT services from a client perspective in such a setting, which provides insights to guide policy makers and other stakeholders in exploring strategies that can bring us closer to the goal of universal HCT coverage.
Acknowledgements
This research study and publication was supported by a United States Agency for International Development (USAID) Cooperative Agreement (TREAT TB – Agreement No. GHN‐A‐00‐08‐00004‐00). The contents are the responsibility of the author(s) and do not necessarily reflect the views of USAID.”
The TB/HIV Integration project, that provides mobile HCT services, was funded by PEPFAR through the cooperative agreement number 5U2GPS000739 from the Centres for Disease Control and Prevention. The content of this publication is solely the responsibility of the authors and do not necessarily represent the official views of the Centres for Disease Control and Prevention.
The support of the City of Cape Town Health Directorate, Western Cape Provincial Department of Health and participating NGOs is acknowledged.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
Conceived and designed the research: SM, PN. Acquisition of data: SM. Analysis and interpretation of data: SM, NL. Wrote the paper: SM, NL, PN, KJ, RB, NB. All authors read and approved the final manuscript.