Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
HIV Med. Author manuscript; available in PMC 2013 October 21.
Published in final edited form as:
PMCID: PMC3801091

Incentivized recruitment of a population sample to a mobile HIV testing service increases the yield of newly diagnosed cases, including those in need of antiretroviral therapy



The aim of the study was to compare the yields of newly diagnosed cases of HIV infection and advanced immunodeficiency between individuals attending a mobile HIV counselling and testing (HCT) service as participants in a population-based HIV seroprevalence survey and those accessing the same service as volunteers for routine testing.


The study was conducted in a peri-urban township within the Cape Metropolitan Region, South Africa. Survey participants (recruited testers) were randomly selected, visited at home and invited to attend the mobile HCT service. They received 70 South African Rand food vouchers for participating in the survey, but could choose to test anonymously. The yield of HIV diagnoses was compared with that detected in members of the community who voluntarily attended the same HIV testing facility prior to the survey and did not receive incentives (voluntary testers).


A total of 1813 individuals were included in the analysis (936 recruited and 877 voluntary testers). The prevalence of newly diagnosed HIV infection was 10.9% [95% confidence interval (CI) 9.0–13.1%] among recruited testers and 5.0% (3.7–6.7%) among voluntary testers. The prevalence of severe immune deficiency (CD4 count ≤200 cells/μL) among recruited and voluntary testers was 17.8% (10.9–26.7%) and 4.6% (0.0–15.4%), respectively. Linkage to HIV care in recruited testers with CD4 counts ≤350 cells/μL was 78.8%.


Compared with routine voluntary HCT, selection and invitation in combination with incentives doubled the yield of newly diagnosed HIV infections and increased the yield almost fourfold of individuals needing antiretroviral therapy. This may be an important strategy to increase community-based HIV diagnosis and access to care.

Keywords: HIV testing, South Africa, incentives


Uptake of HIV counselling and testing (HCT) is still <50% among adults in sub-Saharan Africa, despite a considerable expansion of HCT services over the past decade [1]. HCT scale-up needs to be met with an equal growth in demand for universal access to be achieved. Demand for HCT is driven by distance, costs, knowledge of available services and health-seeking behaviour, which in turn is influenced by income, education and social and cultural characteristics [2,3].

Work-place, mobile and home-based HCT services overcome structural barriers by offering testing in near distance [47]. Studies from sub-Saharan Africa have shown that most people do know where to test for HIV [2,8,9]. The major challenge today is how to enhance health-seeking behaviour and extend HCT coverage to population groups with limited access to existing services.

The success of home-based HCT services might rely on the combination of convenience (bringing the health services to people's doorstep) and personal invitation [5,8,10]. Personal invitation has also been successful in promoting HCT among couples [11,12].

Conditional cash transfer programmes in South America increased health service use and preventive behaviours mainly in the context of child and maternal health [13]. A study from Malawi found that monetary incentives increased the uptake of HIV tests by 27% [14]. More widespread implementation of incentivized testing will need careful consideration of operational, technical and ethical issues. Furthermore, the effect of incentives on health-seeking behaviour and linkage to HIV care following a positive HIV test result will need to be assessed.

We compared the yields of cases of newly diagnosed HIV infection and low CD4 counts (≤200 cells/μL) in individuals recruited and tested as part of a community-based HIV seroprevalence survey and individuals tested on their own initiative at a mobile HCT service in a peri-urban community in Cape Town, South Africa. We also assessed the proportion of newly diagnosed HIV-infected individuals tested following active recruitment who subsequently linked to HIV care.



The study was based in a peri-urban township in Cape Town, South Africa, with 17 000 residents and an adult HIV prevalence of 23% measured in the latest population-based seroprevalence survey in 2010. HCT has been available to all individuals accessing either the local clinic or the hospital since 2001 and antiretroviral therapy (ART) has been provided free of charge since 2004.

The mobile HCT service used in this study has been described elsewhere [15]. HIV testing in combination with screening for other chronic conditions was provided free of charge. Individuals who tested HIV positive were staged according to the World Health Organization (WHO) staging manual and underwent a CD4 cell count test.

Study design

This study used data collected as part of a population-based seroprevalence survey conducted between September and December 2010 [16]. A house-to-house enumeration of the community in August 2010 provided a database of 12 520 residents aged 15 years or older of whom 1300 residents were randomly selected for inclusion in the study (10% of the community). Field workers invited these selected individuals to attend the mobile HIV testing service.

Participant characteristics, HIV prevalence and CD4 cell counts in this group were compared with those of individuals who had voluntarily attended the mobile HCT service since May 2009 up to the time of the survey. Informed consent was obtained from all individuals participating in the survey and those participating in the linkage to care component of the study. Data collection and analysis were approved by the University of Cape Town Research Ethics Committee.

Study population

The HIV seroprevalence survey among recruited participants was conducted over a period of 3 months from September to November 2010. Community awareness was raised before and during the survey through pamphlets and meetings with the community advisory board and church women's groups. Field workers subsequently visited individuals selected for the survey to invite them to participate and provide information. Survey participants were invited to test at the mobile HCT service and could choose (i) to test and receive their HIV result together with screening for chronic diseases, (ii) to provide blood and not receive their HIV result, but undergo screening for chronic disease or (iii) to only provide blood and not receive their HIV result.

All survey participants received 70 South African Rand vouchers (approximately US$9.6) regardless of which testing option they chose. The vouchers were printed using a biometric system that unlocked the voucher on the basis of the participant's fingerprint (Fig. 1). This was done for security purposes and to ensure that participants did not retest and subsequently receive vouchers more than once. The vouchers were redeemable for food at a national supermarket chain. Cigarettes and alcohol could not be purchased with the vouchers.

Fig. 1
Electronic biometric fingerprint reader and voucher. The client places his/her index finger on the biometric fingerprint reader. This machine electronically captures and stores the fingerprint image and then unlocks the printing of the voucher. The system ...

The mobile HIV testing service operated 1–2 days per month in this community prior to the seroprevalence survey. It parked at a township shopping centre or a parking lot in front of the primary school. The service was not formally advertised, but the vehicles were brightly coloured and educators and counsellors invited passers-by to attend the service. Clients attended the free service voluntarily without reimbursement in cash or kind.

Known HIV-positive individuals, survey participants who chose to test anonymously and individuals <15 years of age were excluded from the analysis, as the survey only included adults aged ≥15 years.

Data collection

Data on age, sex, previous test experience, HIV status, clinical stage and CD4 cell count were routinely collected in all individuals testing at the mobile service. Linkage to care was assessed by telephonic or face-to-face interviews in recruited testers with CD4 counts ≤200 and 201–350 cells/μL at 4 and 12 weeks post-diagnosis, respectively. Linkage to care was defined as having attended a healthcare facility for HIV-related care.


For the purpose of this analysis, individuals who tested at the mobile HCT services as part of the randomized population sero-survey and who were therefore personally invited to test and received a voucher were defined as ‘recruited testers’. Individuals who accessed the same mobile testing unit before the survey on their own initiative were defined as ‘voluntary testers’.

All analyses were carried out using stata version 11.0 (Stata Corp. LP, College Station, TX, USA). Characteristics of recruited and voluntary testers were compared using cross-tabulation and the χ2 test.


A total of 2066 individuals attended the mobile HCT service, including 1144 (88%) of the 1300 randomly selected actively recruited survey participants and 922 voluntary testers. A total of 208 recruited and 45 voluntary testers were excluded from the analysis: 66 tested anonymously and 187 were known to be HIV positive. Therefore, 936 recruited and 877 voluntary testers were eligible for inclusion in the analysis. The mobile HCT service visited the study community on 27 days, seeing a median of 35 clients [interquartile range (IQR) 25–42] per day, prior to the survey. The same unit conducted the sero-survey over a 40-day period, seeing a median of 47 clients (IQR 38.5–55) each day.

Age, sex, body mass index and prevalence of tuberculosis symptoms were not significantly different between recruited and voluntary testers (Table 1). Significantly more voluntary testers had been tested before (72.3%) compared with recruited testers (66.9%). The proportion of individuals who had had an HIV test within the last 12 months was higher among voluntary testers (45.6%) compared with recruited testers (35.8%).

Table 1
Comparision between recruited and voluntary testers

HIV prevalence

The yield of cases of newly diagnosed HIV infection was significantly higher among recruited testers [10.9%; 95% confidence interval (CI) 9.0–13.1%] compared with voluntary testers (5.0%; 95% CI 3.7– 6.7%) (Table 1). CD4 count distributions were different, with a larger proportion of individuals with advanced immune suppression (CD4 count ≤200 cell/μL) among recruited testers (17.8%) compared with voluntary testers (4.6%). The median CD4 count was 385 cells/μL (IQR 267–602 cells/μL) among the recruited testers and 414.5 cells/μL (IQR 309–680 cells/μL) among voluntary testers.

Linkage to care

Linkage to care was assessed in 33 (80.5%) out of 41 recruited testers with a CD4 count ≤350 cells/μL We were unable to contact the remaining eight individuals (19.5%) despite seven attempts to establish contact telephonically and/or through home visits. Twenty-six out of 33 individuals (78.8%) reported having linked to HIV care: 11 (73.3%) individuals with CD4 counts ≤200 cells/μL and 15 (83.3%) individuals with CD4 counts of 201–350 cells/μL.


This study shows that active recruitment combined with incentives was associated with twice the yield of cases of newly diagnosed HIV infection compared with voluntary testing at the same mobile HCT service in the same community. The proportion of individuals with advanced HIV infection was more than three times higher in recruited testers compared with voluntary testers. In addition, the proportion of first-time testers and individuals who tested more than 12 months ago was higher in recruited testers compared with voluntary testers, which might explain the differences in CD4 cell count distribution. Use of incentives and active recruitment may be important strategies to increase community-based HIV diagnosis and access to care and treatment.

HCT aims to identify individuals infected with HIV, in particular individuals in need of ART. Twice as many HIV infections and four times more individuals in need of ART were identified through the combination of personal invitation, the provision of targeted information and the offer of a food voucher. In addition, this intervention resulted in a higher number of first-time testers consenting to undergo HIV testing. While the intervention was successful in reaching a particularly vulnerable sector of the population, it is unclear which part of the intervention – personal invitation or incentivization – was more important or if the two parts worked synergistically. In addition, HCT was more frequently available during the sero-survey as compared with the period of routine testing. This might have influenced awareness and test uptake.

Studies on incentivized testing are scarce. In a randomized controlled trial from Malawi, respondents were given vouchers with values ranging between US$0 and 3 at the time they provided blood. The vouchers were redeemable when individuals returned to receive their test result 2 months later. Eighty per cent of those who received any incentive returned for their result compared with 39% of those who did not receive an incentive [14]. In our study, individuals received vouchers for attending the mobile HCT service, but could opt to test anonymously. The majority of clients (94%) tested and chose to receive their result. Thus, the incentive may have encouraged individuals to attend the service and, once they had initiated that step, the majority agreed to be tested.

The effect of active recruitment together with a personal invitation on test uptake has not previously been formally investigated. Personal invitation is a component of home-based HCT services, which have been shown to be effective in increasing test uptake and identification of HIV-infected individuals [5]. However, the impact of personal invitation alone has never been assessed.

Recruited testers received food vouchers amounting to 70 South African Rand (approximately US$9.6), which was more than 10 times the minimum hourly wage (6.31 South African Rand) of a South African farm worker in 2010 [17]. The reimbursement represented a significant amount of money in the context of this community, with 47% unemployment (excluding part-time and informal employment) and a median household income of 1600 South African Rand (IQR 1000-2435 South African Rand) in 2008 [18]. Fraud and security were the two concerns before starting the study. Participants could fraudulently access generic vouchers repeatedly and cash incentives at research sites are a focus for criminal activity. The use of a biometric system allowed attempts at fraud to be limited by identifying individuals who had already received a voucher. The unlocking of the printer by the participant's fingerprint and the fact that it was impossible to print more than one voucher per person reduced the risk of theft and armed robbery. Three attempts at fraud were detected during the study. There was no incidence of theft or robbery.

There are concerns that individuals tested through active recruitment might not show the same level of health-seeking behaviour as individuals testing on their own initiative. This could jeopardize linkage to HIV and ART services. However, in this study, linkage to care was 73.3% in ART-eligible individuals. These results compare favourably with those of a recent study from the same community reporting 67% of linkage among ART-eligible individuals tested at stationary voluntary HCT services [19] and other studies from sub-Saharan Africa [20,21]. A linkage to care study performed at the same mobile testing unit including patients not only from this community, but from the greater area of Cape Town, showed an overall linkage of 52.5% in all newly diagnosed patients. Linkage was highest (100%) in patients with CD4 counts <200 cells/μL, but numbers were very small (n=13), and 66.7% and 36.4% in those with CD4 counts of 201–350 cells/μL and >350 cells/μL [22].

This study has several limitations. First, previous HIV testing experience and linkage to HIV care were both determined by self-report and could be subject to bias. Secondly, the extent to which home visits and/or incentives influenced test uptake could not be determined, but the combination of the two increased the yield of cases of newly diagnosed HIV infection. Thirdly, confounding and changes over time could explain some of the differences between recruited and voluntary testers.

Accessing the harder-to-reach populations that do not necessarily access routine HCT poses a challenge. Active recruitment and incentives might help to extend HCT coverage in previously untested clients and marginalized populations. The magnitude of the incentive, the right target population and the cost-effectiveness of incentivized testing need further research.


We wish to thank all study participants and the dedicated staff of the Desmond Tutu HIV Foundation, in particular the Tutu Tester team and the community field workers.

Funding: KK and SDL have received funding from the Wellcome Trust, London, UK. RW has received funding from IEDEAA (5U01AI069924-02), CEPAC (5 R01 AI058736-02), USAID Right to Care (CA 674 A 00 08 0000 700) and CIPRA (IU19AI53217-07). LGB has received funding from the NIH CIPRA (1U19AI053217). The study was funded by the Wellcome Trust and the Desmond Tutu HIV Foundation. The HIV testing was made possible by the support of the American People through the United States Agency for International Development (USAID).


1. Towards universal access: scaling up priority HIV/AIDS interventions in the health sector: progress report. World Health Organization; Geneva, Switzerland: 2009.
2. Peltzer K, Matseke G, Mzolo T, et al. Determinants of knowledge of HIV status in South Africa: results from a population-based HIV survey. BMC Public Health. 2009;9:174. [PMC free article] [PubMed]
3. Kranzer K, McGrath N, Saul J, et al. Individual, household and community factors associated with HIV test refusal in rural Malawi. Trop Med Int Health. 2008;13:1341–1350. [PubMed]
4. Khumalo-Sakutukwa G, Morin SF, Fritz K, et al. Project Accept (HPTN 043): a community-based intervention to reduce HIV incidence in populations at risk for HIV in sub-Saharan Africa and Thailand. J Acquir Immune Defic Syndr. 2008;49:422–431. [PMC free article] [PubMed]
5. Lugada E, Levin J, Abang B, et al. Comparison of home and clinic-based HIV testing among household members of persons taking antiretroviral therapy in Uganda: results from a randomized trial. J Acquir Immune Defic Syndr. 2010;55:245–252. [PubMed]
6. Corbett EL, Dauya E, Matambo R, et al. Uptake of workplace HIV counselling and testing: a cluster-randomised trial in Zimbabwe. PLoS Med. 2006;3:e238. [PubMed]
7. Grabbe KL, Menzies N, Taegtmeyer M, et al. Increasing access to HIV counseling and testing through mobile services in Kenya: strategies, utilization, and cost-effectiveness. J Acquir Immune Defic Syndr. 2010;54:317–323. [PMC free article] [PubMed]
8. Wringe A, Isingo R, Urassa M, et al. Uptake of HIV voluntary counselling and testing services in rural Tanzania: implications for effective HIV prevention and equitable access to treatment. Trop Med Int Health. 2008;13:319–327. [PubMed]
9. Ostermann J, Reddy EA, Shorter M, et al. Who Tests, Who Doesn't, and Why? Uptake of Mobile HIV Counseling and Testing in the Kilimanjaro Region of Tanzania. PLoS One. 2011;6:e16488. [PMC free article] [PubMed]
10. Mutale W, Michelo C, Jurgensen M, et al. Home-based voluntary HIV counselling and testing found [PMC free article] [PubMed]
11. Allen S, Karita E, Chomba E, et al. Promotion of couples' voluntary counselling and testing for HIV through influential networks in two African capital cities. BMC Public Health. 2007;7:349. [PMC free article] [PubMed]
12. Chomba E, Allen S, Kanweka W, et al. Evolution of couples' voluntary counseling and testing for HIV in Lusaka, Zambia. J Acquir Immune Defic Syndr. 2008;47:108–115. [PubMed]
13. Lagarde M, Haines A, Palmer N. The impact of conditional cash transfers on health outcomes and use of health services in low and middle income countries. Cochrane Database Syst Rev. 2009;7:CD008137. [PubMed]
14. Thornton R. The Impact of Incentives on Learning HIV Status: Evidence from a Field Experiment. Boston: Harvard University; 2005.
15. Van Schaik N, Kranzer K, Wood R, et al. Earlier HIV diagnosis – are mobile services the answer? S Afr Med J. 2010;100:671–674. [PubMed]
16. Kranzer K, van Schaik N, Karmue U, et al. High prevalence of undiagnosed HIV despite high coverage of HIV testing: a randomized population based sero-survey in a South African community; 6th IAS Conference on HIV Pathogenesis and Treatment; Rome, Italy. July 2011; Abstract MOPE 293. [PMC free article] [PubMed]
17. South African Department of Labour. [accessed 10 May 2011];Minimum wages for farm workers in the Republic of South Africa. Available at
18. Middelkoop K, Bekker LG, Myer L, et al. Antiretroviral program associated with reduction in untreated prevalent tuberculosis in a South African township. Am J Respir Crit Care Med. 2010;182:1080–1085. [PMC free article] [PubMed]
19. Kranzer K, Zeinecker J, Ginsberg P, et al. Linkage to HIV care and antiretroviral therapy in Cape Town, South Africa. PLoS One. 2010;5:e13801. [PMC free article] [PubMed]
20. Losina E, Bassett IV, Giddy J, et al. The ART of linkage: pre-treatment loss to care after HIV diagnosis at two PEPFAR sites in Durban, South Africa. PLoS One. 5:e9538. [PMC free article] [PubMed]
21. Amuron B, Namara G, Birungi J, et al. Mortality and loss-to-follow-up during the pre-treatment period in an antiretroviral therapy programme under normal health service conditions in Uganda. BMC Public Health. 2009;9:290. [PMC free article] [PubMed]
22. Govindasamy D, van Schaik N, Kranzer K, et al. Linkage to HIV care from a mobile testing unit in South Africa by different CD4 count strata; 18th Conference on Retroviruses and Opportunistic Infections; Boston, MA. February 2011; Abstract MOPE 293. [PMC free article] [PubMed]