Home-based HIV counseling and testing: Client experiences and perceptions in Eastern Uganda

HIV testing and counseling is the gateway to prevention, care and treatment since these interventions depend on individuals seeking HIV testing and knowing their HIV status [1]. Additionally, studies have suggested that early initiation of HIV treatment may have important prevention benefits [2‐4]. Despite the importance of HIV Counseling and Testing (HCT), its uptake remains low, particularly in developing countries. HIV testing in fact remains one of the greatest challenges to current HIV/AIDS policy [5, 6]. Globally, over 60% of HIV-infected individuals remain unaware of their sero-status [7]. In Sub-Saharan Africa, it is estimated that 80% are unaware of their status and nearly 90% are unaware of their partner’s status [8].

Global concern over the gulf between the needs and the reality has led to urgent calls for increased access to HCT services [9]. Recent years have witnessed new initiatives to increase access, including incorporating HCT into routine healthcare and providing HCT within people’s homes (Home-Based HIV Counseling and Testing or HBHCT). National and international policies have also been revised to incorporate provider-initiated testing and counseling (PITC) [10]. In Uganda options for getting tested are no longer restricted to stand-alone Voluntary Counseling and Testing (VCT) facilities since HCT services were integrated within broader healthcare and HBHCT was adopted in 2005 [11]. The number of people ever tested for HIV duly increased from 23% in 2006; to 38% in 2008, and 57% in 2010/11 to 57% in 2010/11 [12‐15].

Within HBHCT, providers reach out to the community, providing counseling and testing within clients’ homes. In Uganda, HBHCT comes in three different forms. First, some research and surveillance programs include HBHCT as part of their research [14, 15]. The second approach is family-based or targeted ‘index client’ HCT, mainly used by HIV treatment programs, where the homes of HIV-infected patients registered into care are identified, visited and household members tested for HIV. The third approach, the one used in the area where this study was conducted, is door-to-door HBHCT, where mobile teams of counselors and testers mobilize entire communities and provide HCT door-to-door [11].

HBHCT has been lauded for making testing convenient. It has been commended for its success in increasing uptake, with acceptance rates of over 90% [16‐19]. The high uptake of HBHCT is associated with the elimination of costs for clients traveling to VCT centers and the removal of stigma associated with going to such centers [16, 17, 20]. Menzies and colleagues noted the importance of door-to-door HBHCT for reaching many first-time testers, including couples and children, and for the early identification of HIV infection. They also noted that HBHCT, at US$ 8.29 per client, is comparatively cheaper than hospital-based HCT (US$ 11.68) and stand-alone VCT (US$ 19.26) [17]. Other benefits of HBHCT include the opportunity to reach the entire household with HIV interventions [21].

While there are many advantages to HBHCT, our limited understanding of its processes necessitates further study [22, 23]. More specifically, concerns surround: 1) privacy – due to the general lack of counseling space at home [18]; 2) confidentiality – whether health workers can keep clients’ results from being revealed to household members and neighbors; 3) consent – whether testing at home is really voluntary and whether the right to refuse is not down-played by providers [24]; and 4) whether clients diagnosed with HIV infection are referred and successfully linked to appropriate medical care. Some observers remain cautious about the potential for negative effects in HBHCT, including stigma, discrimination and violence that may come with disclosure of HIV-positive status [25‐27]. Similarly, there are concerns about consent and the tension between safeguarding individual rights and protecting public health [9, 28, 29].

This paper examines the experiences of HBHCT clients in Kumi district in eastern Uganda. We asked clients about: 1) the process of mobilization; 2) counseling, consent, privacy and confidentiality; 3) disclosure and 4) referral to care.

This study was conducted as part of the Multi-country African study on Testing and Counseling for HIV (MATCH), designed to compare modes of testing for HIV across four countries: Burkina Faso, Kenya, Malawi and Uganda. The study was conducted in 2008–2009 and included a main survey of clients and providers at healthcare facilities as well as smaller surveys of people involved in home-based testing in Kenya and Uganda, and testing campaigns in Burkina Faso and Malawi. In 2007 and 2008, HBHCT was implemented in selected districts in Kenya and Uganda.

Kumi district (2,821 square kilometers) is located in eastern Uganda and has an estimated population of 370,800 (projected from the 2002 census). HIV prevalence in eastern Uganda is estimated at 5.3% (MoH and ORC Macro, 2006), while Kumi district authorities have estimated district prevalence to be 3.6%. Kumi has three hospitals (two public and one private) and 27 health centres. A number of these facilities provide HIV-related services, for example Prevention of Mother to Child Transmission (PMTCT). Kumi is the second district in Uganda to provide district-wide door-to-door HBHCT, after Bushenyi [30]. The program was implemented in 2007 by district health services with support from the President’s Emergency Plan for AIDS Relief (PEPFAR), through the US Centers for Disease Control and prevention (CDC). HBHCT has been implemented in more than 10 other districts in Uganda, including Bushenyi, Tororo and Kalangala, among others [31].

The district health team trained 29 teams of providers, each with a counselor, a laboratory technician, and mobilizers. The 29 teams moved simultaneously from house to house in the district’s 16 sub-counties, providing HCT services free of charge, using rapid test kits (Determine, Uni-Gold and STAT-PAK) and immediately disclosing results. The teams mobilized a specific number of homes each day, commensurate to their personnel capacity. This meant that they were able to allocate considerable time to conduct comprehensive individual counseling – something that would not be possible in public facilities where providers have additional duties and no control over the number of clients coming to seek services.

Prior to the HBHCT team’s visit, community mobilizers ensured that all households were informed of its coming. Mobilizers mainly consisted of local community leaders, though other community resource persons were also trained for this role. Outreach activities were organized in public places such as churches and mosques while the district health team held talk shows on the local FM radio stations to mobilize communities for HBHCT. People who were married or cohabiting were encouraged to test as couples rather than individuals. Over 95% of people who were approached in Kumi district agreed to be tested with a reported prevalence of 4% [22].

Between April and June 2008, we conducted 395 individual interviews in Kumi district. Kumi was selected because it had recently completed the HBHCT exercise (within six months of this study). This was intended to minimize recall bias.

We adopted the WHO 30 x 7 cluster sampling strategy (27). Previously used in studies to estimate coverage levels in the Expanded Program on Immunization (EPI), it employs a simplified cluster sampling method based on the random selection of 210 respondents in 30 clusters (7 respondents per cluster).

In this study, 32 clusters were selected from four (two rural and two urban) sub-counties. For each stratum (rural and urban sub-county), 16 villages (clusters) were randomly selected to provide a total of 32 clusters. Seven households were selected in each village (cluster). In the rural areas where households are scattered, adjacent households were visited. In the urban and peri-urban areas where households are close to each other, we selected every fifth household to avoid selecting all respondents from the same vicinity. In each household, we interviewed up to two eligible respondents. All adults (male and female) in a household (18 years and older) were eligible for participation. The minimum number of respondents would be 224 (32x7) if only one respondent was interviewed per household. However, the total number of respondents was 395 because we targeted to interview two respondents in every household (except those households where only one person was available at the time of the interview) [32]. Using the rotary method, respondents were randomly selected from those adults present in their homes at the time of the interviews. Overall, 65% of our respondents were women. Men were probably under-represented in the population that we sampled, because they work more often outside the community (the interviews were done during the day).

We used semi-structured questionnaires to elicit information on the HBHCT processes, from mobilization to counseling, testing, receipt of results and post-test experiences (disclosure and access to services). All respondents provided written informed consent either by signature or thumb print (for those who were unable to write). Participants were asked about the information that they received before testing; whether they consented to or declined to be tested; their perceptions of consent and counseling processes pre- and post-test; disclosure of HIV status and the outcomes of disclosure; and referral to care for those who were found to be HIV-positive.

As our goal was to examine the HBHCT processes, we excluded those respondents (114) who reported testing before the HBHCT program started in 2007. Clients’ experiences were analyzed following the trajectory of HBHCT: from the point of community mobilization to counseling before the test, the consent procedure, post-test counseling and receipt of results, disclosure dynamics and for those found to be HIV-positive, referral to care. We used largely descriptive statistics to document respondents’ experiences. The analysis was done by gender to check for gender variations in experiences (p < 0.05). We also analyzed the open-ended questions that were embedded in our questionnaires using ‘cloud analysis’ software to identify common themes, which describe client experiences and perceptions across the HBHCT trajectory. Quantitative data were analyzed using the Stata statistical package (2010).

The study was approved the ethics committee of the Amsterdam Medical Center, the Makerere University Child Health and Development Centre, and the Uganda National Council for Science and Technology.

Out of the 395 individuals interviewed, 376 (95%) had ever tested for HIV. Among the 19 who had never tested, six said they had been approached for testing through HBHCT but declined to test. The majority of those who had never tested (12 out of 19) were men. Out of the 376 respondents who had ever tested for HIV, 281 (75%) had done so through HBHCT in or after 2007. Of this latter group, 12 (4%) reported being HIV-positive while 262 (96%) stated that they were not infected; seven declined to declare their HIV status. The experiences of testing at home described below concern the 281 respondents who tested through HBHCT in or after 2007.

Most HBHCT clients (65%) were married or cohabiting and had little formal education. Cultivation was the leading occupation (51%), with only 10% in professional employment and 18% in trade(see Table 1).

In Table 2 the experiences of HBHCT are presented by gender, and in Table 3 the referral experiences are given for the 12 respondents who tested HIV positive.

This study revealed high levels of uptake of HIV testing and counseling and overall satisfaction with HBHCT. The percentage of respondents who had ever tested was much higher than the national average of 38% at the time [13]. All individuals who tested received their results immediately, facilitated by the use of rapid test kits. Similar results have been reported by other HBHCT studies andprograms [16‐18, 30, 33], suggesting that if expanded, HBHCT could increase uptake of HIV testing and reach many first-time testers.

The quality of services and the attitude of healthcare providers are often cited as limiting factors in the use of HIV counseling and testing services [31, 34]. In our study, both male and female clients reported largely positive experiences with HBHCT: the information they received during pre- and post-test counseling, the consent procedure, and general handling by providers. Most respondents were counseled before the test and felt that the information they received from providers was sufficient. Allowing clients to ask questions was highly valued.

We found high adherence to consent requirements on the part of the HBHCT providers, in particular explanation of the opt-out option. Nearly all (94%) respondents (both male and female) acknowledged being given the chance to opt out of the test; and they valued this. The Kumi HBHCT teams were able to allocate considerable time to conduct comprehensive individual counseling since they controlled the number of households and individuals served on each day.

Contrary to our expectation that privacy within homes would be problematic due to limited space and the presence of other family members, clients considered their own homes as more private than healthcare facilities. Most respondents were satisfied with the privacy offered by HBHCT. Other studies have described the involvement of families in seeking private counseling space within and outside the home [20]. The participation of clients in identifying private spaces within the home was experienced as empowering, in contrast to facilities where users have no say over the setting of testing and counseling.

While most clients were satisfied with the level of privacy, more than a quarter cast doubt on whether their test results would be safeguarded after providers left their homes. In this study, no direct question was asked about breach of confidentiality by the provider. Nonetheless, in order to build trust and confidence, HBHCT clients should be informed about the practical issues of confidentiality beyond the test process and what happens to records when providers leave their homes.

It is widely assumed that individuals do not want others to know that they have tested for HIV [35]. In the context of HBHCT, anonymity may be impossible. Family members and neighbors knew about the visiting HBHCT team and tried to find out whether others had taken the test. Almost half of the respondents were asked by other community members if they had taken the HIV test. Nevertheless, we found that anonymity was not a major concern among respondents as there was no stigma attached to testing: most household and community members were in the same boat. Those who declined to test were the minority. This is unlike testing in a voluntary testing and counseling facility where lone individuals who go for testing prefer to hide their identity [36]. HBHCT opened up space within the home to talk about HIV and testing. To some extent, interactions and discussions among partners and other family members influenced individual decisions to test at home, though others stated, that testing was a personal responsibility. The dynamics of these discussions varied between respondents; patterns of influence often did not conform to the traditional hierarchy of many African families (e.g. parents over children, husbands over wives). For example, we found sons and daughters to have encouraged their parents to take the test.

Prior mobilization allows potential clients to think about, discuss and make a decision before the team arrives. This is similar to VCT where an individual can take time to consider whether or not to be tested. Individuals who do not wish to be tested could, for example, leave home before the arrival of the HBHCT team. The involvement of local leaders in the mobilization process can influence individual decisions to take the test; their involvement encouraged trust in the program and community cooperation. Although not a focus of this study, the widespread mobilization and discussions around HIV within the homes and community may change the norms around HIV testing and impact on HIV stigma, discrimination and access to services.

Disclosure levels were found to be high: 78% of HBHCT clients had disclosed their HIV test results to someone, often to more than one person. However, the disclosure was selective as reported in other studies [37, 38]. Men were found to disclose significantly more to others in the household and to friends than women. This is an interesting finding, as other studies suggest that men are more secretive about their status than women. It is important to note here that 96% of our informants were negative. For those who test negative, disclosing is perhaps not a big deal. Disclosure problems are more likely to occur for those who test positive. However, some previous studies have demonstrated no increases in domestic violence or other negative social outcomes after HBHCT [39]. Our findings show that all of the HIV positive respondents had disclosed their results to someone. Many clients tested (69%) were counseled with their partners, suggesting that home-based testing is a good way to promote couple-counseling. Couple counseling and testing may also explain the high disclosure rates to sexual partners, in comparison to other studies [30, 40].

The findings from this study may have some limitations: 1) Recall bias (some respondents had tested more than a year prior to the interview); 2) This study gathered information on only individuals who were present at home at the time of the interview, which may exclude the views of those who were out (at work); 3) The study generally relied on respondent accounts which may be subject to bias due to social desirability. However, in real life, it is users’ perceptions that trigger many health actions including care, and it is thus important to explore the client perspectives.

The number of HIV infected individuals was too small to exhaustively explore linkage to and access to HIV care and treatment as well as outcomes of HIV status disclosure among the HIV infected individuals. However, most of the HIV infected respondents confirmed referral to medical care and support groups, similar to another study of HBHCT in western Uganda [30]. Even with the linkage to care challenges, HBHCT can still play a major role since several studies demonstrate that knowing one’s status if a person is HIV-infected, has a substantial benefit in terms of risk reduction [41, 42].

Several studies show that HBHCT increases uptake of HIV testing, identifies infected individuals earlier, and reaches more couples and children, in comparison to other HCT approaches [21, 30]. Despite the ethical concerns, our study shows good adherence to standard HIV counseling and testing recommendations (consent, counseling, confidentiality and referral to care) and appreciation of the HBHCT approach by the community.

This study also shows high coverage of HIV testing within the Kumi district community after HBHCT, with many individuals testing as couples. HBHCT can play a significant role in rapidly increasing access to HIV testing, care and treatment as well as prevention services.