Background

The South African National Strategic Plan on HIV, STIs and TB aims to get 80% of the population to know their HIV status by 2016 [1]. However, available HIV counselling and testing (HCT) strategies like client or provider-initiated counselling and testing will not be sufficient to achieve universal access to HIV testing and treatment [2] because of challenges that limit their effect and scope [3, 4]. Additionally, HIV self-testing (HIVST) is globally accepted as an important complement to existing HCT approaches, and several countries including South Africa have already introduced or are considering the introduction of HIVST as part of their testing strategies [2, 5, 6]. Moreover, HIVST has the potential to facilitate more people getting tested because it has advantages of convenience, speed, privacy, anonymity, confidentiality, and accessibility. Further, HIVST can also help eradicate most of the barriers that stand in the way of people getting tested. It has the potential to be an acceptable option for high risk populations who would otherwise not test for HIV using currently available HCT services for various reasons, including stigma [6, 7]. All of these benefits of HIVST are aligned to recent World Health Organization (WHO) HIV testing guidelines that highlighted the potential contribution of HIVST to close critical gaps in HIV testing coverage worldwide [8].

The adoption and implementation of HIVST as a testing option in many African countries as well as globally did not come without challenges, criticism, and controversies from health experts and government stakeholders [2]. The main arguments against HIVST include the risks of inaccurate results, potential psychological risks posed by lack of adequate counselling, and uncertainty over unsupervised linkage to care for individuals who test positive for HIV [9, 10]. There is evidence that the arguments against self-testing are largely based on vague fears since HIVST is a new modality [11]. A recent review of articles showed little evidence of potential harm from selected self-tests as well as HIV self-tests [12]. However, the absence of face-to-face counselling, which is viewed as an essential component of all HCT models in South Africa, remains the most significant argument against HIVST [2]. It is therefore crucial that research on the acceptability and feasibility of HIVST should also investigate the provision of pre- and post-counselling.

South Africa has lagged behind in the adoption and creation of awareness of HIVST as a testing option to increase access to HIV testing mainly because of the constrains and arguments against HIVST. It is only recently that key stakeholders, engaged in HIV prevention and treatment, research, and policy formulation in South Africa, have indicated significant support for the introduction of HIVST in the country’s HCT portfolios [2]. Although HIVST is now acceptable in South Africa as an option to increase HIV testing [13], its acceptability among population groups with poor uptake of HCT remains limited. Therefore, future research on HIV testing acceptability is of significance to facilitate improvement of the uptake of HCT [14]. To investigate the acceptability of HIVST among students in Technical Vocational Education and Training (TVET)colleges, the study determined their HIV testing practices using the traditional provider and client initiated services, assessed their acceptability and willingness to use HIVST, and determined their perceptions on the provision of pre- and post-test counselling in the context of HIVST. South African youth are considered as one of the HIV high risk groups and TVET colleges enrol a large number of youth (at most 75%) of the youth from 16 to 24 years of age from poverty stricken backgrounds. TVET colleges are also a relatively new types of institutions (started after 2010) compared to generic institutions of higher learning such as Universities and hence they are under-researched while they enrol a cohort of students who are educationally and economically deprived like the majority of the people in South African communities [][15].

The study was conceptualised and conducted at a time when South Africa has not yet taken a position to promote and adopt HIVST. The findings of the study remain relevant even after the adoption of HIVST in the country, given the limited data on HIVST. The findings will highlight the gaps that still exist for scaling up HIVST and will also be used as a tool to increase social awareness of HIVST to further promote its adoption and uptake.

Materials and Methods

Results

The description of study participants in Table 1 illustrates the demographics, sexual behaviours and the HCT practices of the students. A total of 3,662 students participated, of which 68.28% were in years 1 to 3, while 31.72% were in years 4 to 6 of training. There were slightly more female students 56.6% than males. The mean age of the students was 21.9 years (SD: 2.7, age range 18–32 years). The majority of the students (n = 2,541, 70.4%) were between 20 and 24 years old.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographics, sexual behaviour and testing practices of TVET students.

https://doi.org/10.1371/journal.pone.0169765.t001

More than three-quarters, 81.7% were had sexual relationships, 66.2% had one sexual partner, 19.4% had two sexual partners and 14.4% had more than two sexual partners. The mean age of sexual debut was 17.7 (SD: 2.4 years). Up to66.5% reported having used condoms in the last sexual act. Data also show that 72.2% of the participants had ever tested for HIV and 60.2% reported that they knew the HIV status of their sexual partners.

Opinions of HIV counselling and testing and the reasons for uptake and non-uptake

Of the students who up took HCT, the services were accessed at the Clinic (48.6%), HCT Campaign (27.9%), at a hospital (11.7%), at a General practitioner (9.8%) and other unspecified service providers 2.0%). Data also show that 79.2% reported satisfaction with HCT services, and a majority of 91.1% considered HIV testing to be important for young people. In addition, 87.9% cited that HIV pre-test counselling was necessary (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Opinions on HCT and reasons for testing and not testing.

https://doi.org/10.1371/journal.pone.0169765.t002

From a sample of n = 2,752 who responded to reasons for the uptake of HCT, the student cited the reasons as wanting to know their status (76.2%), testing due to pregnancy (7.1%), being sexually active 128 (4.7%); tested with partner 178 (6.5%); and offered HCT by a healthcare worker 152 (5.5%). The reasons for the non-uptake of HCT from a sample of (n = 1048) were cited as not feeling at risk of contracting HIV 265 (23.5%); not being sexually active 220 (19.5) fear of stigma if the test results come out positive 168 (14.9%); not wanting to know once HIV status 152 (13.5%); not feeling comfortable testing in hospital settings 133 (11.8%); and not sure if the results will be kept confidential 110 (9.78%).

Opinions and acceptability regarding HIVST

A proportion slightly less than half of the students who participated in the study (46.2%) we aware of HIVST prior to the survey. The acceptability of HIVST as a testing option was 87.1%. The proportion of students who indicated the necessity for pre-test counselling for HIVST was 60.7% and the necessity for post-test counselling for HIVST was 60.7%. Moreover, the willingness of participants to purchase a self-test kit was 74.7%, while the willingness to submit HIVST results at a local clinic for inclusion in the HCT statistics which is necessary for the targets of the HCT uptake campaign was at 14.8%, and the willingness to confirm HIV positive self-test results at a local health facility was 75.6%. In terms of the use of the toll-free hotline for provision of HIVST post-test counselling, only 40.0% indicated that hotlines were appropriate, while 45.7 were not sure and 14, 3% did not approve. Appropriateness of using information leaflets to provide information on how to use a self-test kit was accepted by47.9% of the students, 12.5% disapproved and 39.6% were not sure (Table 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Opinions and acceptability of HIVST and counselling by gender.

https://doi.org/10.1371/journal.pone.0169765.t003

There were significance gender differences for acceptability of HIVST among the students, with more females showing acceptability versus males (p = 0.000), more female students willing to confirm HIV-positive test results at a health facility, more female students citing willingness to submit HIVST test results at the nearest health facility (p = 0.03), more females willing to use HIVST with their sexual partners (p = 0.000), and willing to buy HIVST kits (p = 0.000).

Factors associated with the uptake of HIVST

The outcome measure for this study was the acceptability of HIV self-testing. Factors significant at the p < 0.05 level in unadjusted bivariate analysis were the following: being tested for HIV in the past year, being sexually active, number of sexual partners, ever tested for HIV, willing to buy HIVST, submit test results at the local health facility for HCT statistics, confirm test results at the local health facility and uptake HIVST with a sexual partner.

We first analysed computed the bivariate analysis followed by logistic regression to confirm the association. The outcome variable for this analysis was acceptability of HIVST. Logistics regression using forward elimination of the statistically significant variable for bivariate analysis was computed for the acceptability of HIVST.

Logistic regression (Table 4) showed that college students who were willing to uptake HIVST were at most four time more likely test using HIVST with their sexual partners at unadjusted odd ratio (OR = 3.99; CI: 2.84–5.62). At adjusted odd ration statistics show that college students who had ever tested for HIV within the HCT program were twice as likely to accept HIVST (OR = 2.13; CI:1.72–2.63).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Factors associated with the acceptability of HIV self-testing.

https://doi.org/10.1371/journal.pone.0169765.t004

Discussion

The study assessed the acceptability of HIV self-testing among students in TVET colleges in rural and urban sub-districts of two provinces in South Africa in the wake of the adoption by many countries of HIV self-testing as an alternative HIV testing strategy. Since HIV self-testing is an alternative HIV testing option, we also assessed the students’ HIV testing practices using the available HCT services. The uptake of HCT was high among the students, with almost three-quarters (72%) indicating that they had been tested in the past year. The uptake of HCT is relatively high compared to that of students in colleges and universities in other countries [19–21]. Over three-quarters (76%) indicated that they tested to know their HIV status. Of the students (n = 1,048) who had not tested, about a quarter did not perceive themselves at risk of contracting HIV. This should be viewed in the context of reported risky sexual behaviours among the students. Over a third (33.3%) reported multiple sexual partners, and of the sexually active students, 33.5% did not use condoms the last time they had sex. Other reasons cited for not having been tested included not being sexually active, fear of stigma, not wanting to know one’s HIV status, not feeling comfortable testing in hospital settings, and not being sure that the results would be kept confidential. These findings are consistent with reasons cited in other studies [22, 23].

Even though less than half of the students were aware of HIVST prior to the survey, the majority (87.1%) considered HIVST to be acceptable, and 84% would uptake HIV self-testing with partners. Being sexually active, having multiple sexual partners, and having been tested for HIV were associated with acceptability of HIVST. A randomised control trial (RCT) by Choko and colleagues [24]. also showed high (84%) uptake of HIVST among people over the age of 16 years. This compares to the level of acceptability in our study. High HIVST acceptability has also been illustrated with different population groups in other studies in Sub-Saharan countries [25–27].

The cost of an HIV self-test kit has been identified as a potential barrier to adoption, willingness to use, purchase, and the uptake of HIVST, particularly among people in poor-resource settings [13, 28]. Despite the students being from poor-resource settings, we found that three-quarters (75%) reported the willingness to purchase self-testing kits. Willingness to self-test was also reported in a feasibility study conducted with university students in Canada; the acceptability of HIVST was also high among these students [29]. Nonetheless, about a quarter of the students in the current study were not willing to purchase the self-testing kit. This suggests that if there was no cost involved in the acquisition of self-testing kits, the uptake of HIVST might greatly increase [13]. While most trainee nurses in a South African study were willing to purchase the self-testing kits, some felt that it should be provided for free because the cost will be a potential barrier for the uptake of HIVST [30].

We found that 59%% of the students indicated that they would submit their HIVST test results at a local clinic for the purpose of health statistics even though less than half of the sample participating in the study did not know about HIVST prior to the survey. The HIVST is regarded as safe if implemented correctly to ensure the accuracy of the tests, and there is confirmation testing and referral for counselling and care [9, 29, 30]. According to Estem and colleagues [13], setting up effective strategies for measuring HIV incidence and prevalence and the extent to which people who have conducted the HIVST report their status is one of the conditions that should be addressed before the scale up of HIVST.

In terms of confirming positive test results, over three-quarters (75%) reported the intention to confirm their HIV positive self-test results at a local health facility. While it is commendable that the majority wanted to be linked to a health facility after self-testing, currently, there are no referral mechanisms to link people who self-test to treatment and care [2]. Some of the suggestions on the linkage to confirmation testing included that persons with positive HIVST results return to their pharmacist where the self-testing kits were purchased to seek advice [30, 31]. Makusha and colleagues [2] suggested that the self-testing instruction kits should provide adequate information on what do after self-testing. Despite high intentions to confirm HIVST results, a quarter (25%) of the students did not intend to confirm their self-test results. Willingness to confirm HIV results and seek treatment after a self-test is key to effective control of the HIV epidemic [2]. The role of educating people on what to do after self-testing cannot be over emphasised, particularly because HIVST requires individuals to be more proactive than when testing using facility-based HCT approaches [31].

In order to ensure proper implementation of HIVST, post-test counselling should be accessible to people who conduct HIVST [2, 6, 30, 32], and the reported satisfaction level of 79.4% with the current HCT among the students explains the importance of face-to-face counselling. Makusha and colleagues [2] argued that the absence of face-to-face counselling remains a significant challenge for the adoption and implementation of HIVST. This is true of the findings in the current study; we found that the suggestion that pre-and post-counselling be provided through an anonymous toll-free hotline service was acceptable to only 40% of the students. Pai and colleagues [33] found diversity in opinions as to the desired approach to counselling. Only 16% of the students preferred anonymity, and 41% were comfortable with either anonymity or face-to-face counselling, while 39% preferred face-to-face counselling. In our study, similar preferences for face-to-face counselling were reported as in other studies [26, 32].

In spite of recommendations and suggestions that self-testing instruction kits should provide information about self-testing in the form of leaflets [2, 4, 27–31]], only half (50%) of the students in the current study felt that information contained in the leaflet would be sufficient to provide information on how to use self-testing. This implies that other strategies and approaches should be explored to educate and promote self-testing. Besides pamphlets or leaflets, a link to a YouTube video demonstration was suggested as one of the ways to show people how to use self-testing kits [30]. The use of instructional videos and pamphlets in a feasibility study resulted in successful conducting of self-tests among students in an unsupervised HIVST programme [29].

Limitations

Several shortfalls of this study should be considered when interpreting the findings. The study employed a cross-sectional survey design with only college students in two provinces included in the sample; hence, there is a level of selection bias even though probability sampling was used to select the sites (TVET colleges), and random sampling was done to address the bias. Since probability sampling was used and the sample size was considerable, the findings can be generalised to other similar contexts but not to the general population of young people in South Africa. There were incomplete responses in some of the questions, which resulted in missing values, but the sample size in the analysis was adjusted to account for the missing values.

Implications

Currently, HIV self-testing is acceptable in South Africa as an HIV testing option, and self-tests kit are available over the counter in pharmacies. Nonetheless, the high acceptability of HIVST among the students calls for extensive planning and preparation for the scaling up of HIVST beyond its availability in pharmacies. Even though the students intend to confirm their HIV positive results with a local clinic, there is currently no plan or system for linking HIVST to health facilities for confirmatory testing and HIV care as recommended by WHO. In planning for linkages, the country should guard against supervised HIVST because the fundamental principal for HIVST is to allow choice for users to test without the need for a health worker to be present [32]. For a country with such diverse populations, the provision of HIVST should take into consideration the context of the target populations and tailor the distribution site, cost, and support accordingly. This will address the needs of population groups who may specifically choose HIVST to avoid the need for counselling that does not meet their needs, who may afford to pay for self-testing kits, and who might know when and how to access treatment and care. A one-size-fits-all approach may not be appropriate for South Africa. The attempt to follow up on the usage of HIVST at government clinics should not represent a barrier that could undermine the principal motivation for those seeking self-testing, namely full confidentiality [32].

There is no doubt that pre- and post-test counselling remains a challenge for the scaling up of HIVST. The remote services that will complement the convenience of HIVST, such as toll-free hotlines, are vital for the scaling up and implementation of HIVST. However, the telephone hotline as a counselling approach has not been tested in South Africa in an unsupervised HIVST model, and our findings suggest that this is not the preferred method of counselling for the students. The appropriateness of this counselling approach should be tested in unsupervised feasibility studies to explore user perspectives on the counselling and linkage to care requirements in the self-testing model [6].

The current results should be taken in the context that these are college students and the telephone hotlines might work for other population groups. Nevertheless, there is a need for further studies to explore provision of counselling for effective implementation of HIVST in South Africa and other African countries. Awareness campaigns similar to those conducted to promote provider-initiated testing and counselling should be considered to educate communities about HIVST.

Conclusions

A majority of the TVET college students already utilise HCT in public health facilities, and a vast majority have high levels of acceptability for HIVST and intention to test with partners. The results suggest that scaling up HIVST in South Africa is feasible, considering the majority of students are willing to purchase HIVST kits despite being from low socioeconomic settings and to confirm HIV positive test results at a local health facility. It is also evident that HIVST as a new modality is more acceptable to the students who had ever tested for HIV in the existing HCT program. This means that there is a need for programme designers to device innovative ways of making new modalities such as HIVST more appealing to those who never up took HCT for various reasons.

Supporting Information