High Uptake of Home-Based,District-Wide,HIV Counseling and Testing in Uganda

Abstract

More than 80% of the people infected with HIV in low-income countries of sub-Saharan Africa do not know their HIV serostatus. Innovative measures of increasing access to HIV counseling and testing (HCT) are urgently needed so as to improve care and prevention. We implemented a home-based HCT program in Bushenyi District from September 2004 to March 2007, in Uganda where approximately 90% of people aged older than 14 years had never tested for HIV to gauge whether it was acceptable and increased uptake of HCT. Twenty-nine teams comprising a counselor and a laboratory assistant systematically visited homes offering HCT for all people older than 14 years of age and at-risk children (mother deceased or HIV infected) using a rapid HIV testing three-test algorithm. HIV-infected people received cotrimoxazole prophylaxis, were supplied with long-lasting insecticide-treated bed nets and equipment for treatment of drinking water at home, and were referred for assessment for antiretroviral therapy. The program reached 92,984 (63%) of all the homes in the district. Of these, 32,3621 people were eligible for HCT, and 28,2857 (87%) were present at home and were offered pretest counseling. A total of 264,966 (94%) accepted testing and received their results, of whom 11,359 (4.3%) were HIV-infected. Ninety percent of those testing had never tested before. The cost of testing was $7.83 per previously untested client. Ninety-seven percent of HIV-infected people initiated cotrimoxazole prophylaxis, 74% received bed nets, 70% received water treatment equipment, and 11% began antiretroviral therapy. Forty-four percent of people who were in an HIV-discordant relationship were infected. These results demonstrate that home-based HCT was well-accepted, feasible, and effective in identifying HIV-infected individuals who did not know their HIV status in rural Uganda.

Introduction

I n Uganda, 6.4% of the adult population is HIV infected.¹ As HIV/AIDS epidemics mature, the proportion of infections transmitted through casual sexual partnerships decreases and HIV is increasingly spread through regular sexual partnerships within HIV-discordant couples, the majority of whom are not aware of their HIV status.^2

–9 In this situation, without testing, HIV prevention messages regarding sexual abstinence, faithfulness, and condom use have limited efficacy, as most married or cohabiting couples assume they are HIV-concordant (either negative or positive) and are unlikely to use condoms or be abstinent within their relationship.^8

–15

HIV counseling and testing (HCT) of individuals and couples is a cost-effective primary HIV prevention strategy.¹⁶ For HIV-infected adults, knowledge of HIV status has been associated with a 63% reduction in risk behavior,⁸ and in HIV-discordant couples, HIV testing and condom provision has been associated with an 80% reduction in HIV transmission.⁹ In order to strengthen HIV prevention, there is a need to increase HCT so that people are aware of their own HIV status and that of their sexual partners, can implement efforts to prevent HIV transmission or acquisition, and can access effective HIV-specific care.¹⁷ For HIV-infected people, HCT provides the opportunity to access antiretroviral therapy (ART), prevention of mother-to-child transmission services, and basic preventive care, such as cotrimoxazole and isoniazid prophylaxis. In addition, people with HIV who have received their test results have useful information for important life decisions such as reproduction and long-term planning.

In Uganda in 2005, only 12% of women and 11% of men aged 15–59 years reported that they had ever received HCT.¹ The major barriers to HCT include perceived negative consequences of a test result, influences from a sexual partner, cost of HCT, physical accessibility of HCT, awareness, risk of HIV infection, the need for linking HCT with care, and the perceived low quality of care of HCT services.^1,18,19 Home-based HCT has acceptance rates of over 90%, thus overcoming many of the perceived obstacles.²⁰ However, it has not been fully implemented in a population-based program outside of a study context. If HCT could be offered to an entire population over a relatively short period of time, this could not only increase access to an important prevention tool, but could potentially reduce morbidity and mortality by early identification of HIV infection and provision of care. In high HIV prevalence settings, e.g., in Uganda, and many other countries in sub-Saharan Africa, it would be ideal for every adult to know their HIV status. Providing this service on a community-wide basis might help to reduce stigma that can be associated with testing, educate large populations, and empower the community to take charge of its own health.²¹ We implemented a district-wide, door-to-door, HCT and care program in Uganda and assessed the uptake of HIV testing and care services.

Methods

Program setting

The home-based HCT program (named “Everyone Know Your Status”) was implemented by the Integrated Community Based Initiatives (ICOBI), an indigenous nongovernmental organization, in collaboration with the District Health Services of Bushenyi District, in southwestern Uganda from September 1, 2004 to March 31, 2007.

Bushenyi District has a population of approximately 800,000 people (projected from the 2002 census) in an area of 5396 square kilometers. At least 20% of the population is younger than 5 years of age; 44% are between 15 and 49 years of age, and approximately 95% of the population is rural with subsistence agriculture as the main source of income. Administratively, Bushenyi District is divided into 5 counties, 29 subcounties, 170 parishes, 2034 villages, and 142,832 households. The district is served by 3 hospitals, 89 health centers, and 215 immunization posts. In 2004, facility-based HCT, including provision of services for prevention of mother-to-child transmission of HIV, was available at 15 of the health units. Provision of antiretroviral therapy (ART) was available at 5 health units and CD4 cell count testing was available at one health unit.

HIV counseling and testing

All people (older than 14 years of age) found at home were eligible for HCT. Children younger than 15 years and older than 18 months were eligible if considered at risk of HIV infection (mother dead or HIV infected). HCT was provided by 29 outreach teams consisting of a laboratory assistant and a counselor who systematically visited every home offering HIV education, counseling, and testing. The teams prepared workspace behind or in the home and followed a national serial rapid testing algorithm (Determine™, Abbott Laboratories, Abbott Park, IL) for screening, Statpak™ (Chembio Diagnostics, Medford, NY) for confirmation and Unigold™ (Trinity Biotech [Bray, Ireland] as tie breaker) on finger-stick blood collected in ethylenediaminetetraacetic acid (EDTA)-coated capillary tubes. Results were provided in a confidential setting in or near the home; participants could choose to test and receive results as individuals or as couples. The mean household composition was seven people of whom three were eligible for HCT. The testing teams spent approximately 2 h in each household.

Dry blood spots were prepared for retrospective quality control from all HIV-positive and 5% of all HIV-negative samples and retested in the Centers for Disease Control and Prevention (CDC) Reference Laboratory in Entebbe. In the reference laboratory, Genetic Systems HIV-1/HIV-2 EIA (Redmond, WA) and Vironostika Uniform II + O (bioMerieux, Durham, NC) were used in parallel for retesting. There was over 99% agreement between the field results and those of the reference laboratory.

Throughout the district, 170 resident parish mobilizers (RPMs), one in each parish, and village chairmen mobilized communities, supported outreach teams to carry out home-based HCT, provided follow-up posttest support, encouraged the formation of parish-based HIV posttest clubs, and referred those diagnosed with HIV infection to relevant service organizations. The RPMs were community-owned resource persons trained by the program and paid a minimal stipend; criteria for being chosen as an RPM included being able to read and write in English, permanent residence in the parish, and having their communities' trust for this role. Persons with HIV were encouraged to apply. The RPMs were trained in a 2-week specially tailored course to suit their expected roles. The course included basic counseling skills, family and community based care, HIV/AIDS prevention, treatment and care, HIV/AIDS education including peer education. The course also included sessions on voluntary counseling and testing (VCT) and prevention of mother-to-child transmission of HIV (PMTCT). The RPMs were also trained to distribute condoms, mosquito nets, and safe water vessels.

Before the outreach teams visited a village, the RPM and the village chairman would visit every home and inform the household members about the day and the purpose of the visit by HCT outreach teams. In addition to these visits, informal and formal community mobilization sensitization meetings were held in every parish. Informal information was passed through gathering such as weddings, funerals, village meetings, or religious congregations. In addition weekly talk shows on a local radio station with a phone-in facility were organized. Listeners could call and ask questions related to HIV in general and HCT in specific. Those unable to call could send written questions through outreach teams. In Bushenyi District, an estimated 85% of households own a radio and 10% a mobile phone.²²

Provision of care and prevention services

Along with referral to HIV care and treatment services, RPMs were trained to directly provide every household with an HIV-infected person with psychosocial support (like counseling the person infected and those affected in coping skills, ongoing counseling, and referral for services), two long-lasting insecticide-treated nets, condoms, a plastic water vessel, and an ongoing supply of a dilute chlorine solution.²³ In addition every HIV-infected person was assessed for initiation of cotrimoxazole prophylaxis²⁴ by either ICOBI clinicians or from the network of health units in Bushenyi District. Replenishing of cotrimoxazole was continued with the help of RPMs. HIV-infected persons who had been coughing for 2 weeks and or were chronically sick were referred to health units for tuberculosis (TB) diagnosis, treatment of opportunistic infections, and evaluation for antiretroviral therapy (ART). Women who were pregnant were referred for PMTCT services. For the last 796 people (>14 years) who tested HIV-positive, a venous blood draw was performed in the home and CD4 cell count enumeration conducted at the health center. CD4 results were delivered to the people in their homes. In addition, special support groups were formed for HIV-discordant couples and HIV-infected adults to promote prevention with positives.

Adult participants provided informed oral consent for receiving HIV testing. For children and other people who were not able to consent, consent was provided by parents/guardians or caregivers according to the existing guidelines of the Uganda Ministry of Health.²⁵ A private place in the house or within the garden was used for posttest counseling and giving of results. Disclosure of results to a third party was the decision of the individual or couple. The evaluation protocol was approved by the Institutional Review Boards of the Makerere University Institute of Public Health, and the Uganda National Council for Science and Technology.

Data management and analysis

Essential program data were collected daily, compiled at the end of the week, and entered in a computer database using Access (Microsoft, Seattle, WA). We calculated the proportion of people who were found at home, who tested for HIV, who tested HIV-positive, and those who received specific care and prevention services. We also calculated the proportion of couple members who opted to receive results together and the rate of HIV-discordance. In order to calculate program coverage consistent with other national and international indicators²⁶ we estimated the proportion of all people aged 15 to 49 years who tested for HIV within this program, using population data from the 2002 census.

Basic demographic information, e.g., age, gender, education, and relationship to the head of the household, were collected using a household-level questionnaire (answered by the head of household or spouse) and individual questionnaires answered by all people aged older than 14 years. The household questionnaire was used to list the members in each household. Visitors were defined as people staying in the house that do not normally reside there. Although visitors were offered HCT they were not included in analyses. On the individual questionnaire we collected data on age, gender, marital/cohabiting status, occupation, sexual behavior in the previous 3 months, previous testing for HIV, and pregnancy status. The aim of this information was to help the counselor in tailoring the counseling and prevention to the needs of each person.

Program costs

A cost analysis of the home-based HCT program was conducted in order to assess program efficiency and success at targeting priority populations. The economic costs of the program were assessed over a 6 month period between March 1 and August 31, 2005. A programmatic perspective was adopted, collecting data on recurrent and capital expenses associated with HCT provision. Cost data were collected from project inventories and accounts and through interviews with key project personnel. Donated or subsidized goods were valued at market rates. The costs of capital items were annualized over a useful life of 4 years for initial training and 5 years for vehicles and other equipment, with a discount rate of 3%. Costs were converted to U.S. dollars at market exchange rates and inflated to current prices (2007 U.S. dollars). Additional data on total clients and priority populations seen during the evaluation period were derived from program monitoring records.

Results

HIV counseling and testing uptake and results

Of a total of 146,532 households in the district, 92,984 (63%) were reached with HCT during the program period. The total number of people who were eligible for HCT in these households was 323,621, of whom 282,857 (87%) were found at home and provided pretest counseling. Of those counseled, 264,966 (94%) accepted testing and 264,953 (100%) received their test results (Table 1). A total of 11,359 (4.3%) were HIV infected. Of those who tested, 90% (89% of the women and 91% of the men) were testing for the first time ever. Among the people tested, 26,949 (10%) were below 15 years of age, and 204,979 (77%) were aged 15–49 years. Of children (<15 years) tested, 1747 (7%) were aged 18 months to 4 years and 21,950 (81%) were above 10 years. The prevalence of HIV among children less than 5 years was 7.6% (8.0% for males and 7.2% for females), 5.1% for those aged 5–9 years (5.1% for males and 5.2% for females), and 0.7% for those aged 10–14 years (0.7% for females and 0.7% for males). There was no difference in HIV prevalence between females and males for children less than 15 years (crude odds ratio [OR] 1.06, 95% confidence interval [CI] 0.88–1.28, p = 0.56). Of all the people tested, 32,140 (16,243 men and 15,897 women) were above 49 years. Men above 49 years were more likely to be HIV infected (2.8%) than women (1.8%) of the same age group (OR 1.64, CI 1.42–1.85, p < 0.0001). This is probably because men above 49 years are more likely to be sexually active compared to women of the same age group.¹

Table 1.

Coverage of Home-Based HIV Counseling and Testing in Bushenyi District Uganda

	Male number (%)	Female number (%)	Total number (%)
Coverage for the whole population
Eligible	161,208	162,413	32,3621
At home	135,689 (84)	147,168 (91)	282,857 (87)
Tested	123,501 (91)	141,465 (96)	264,966 (94)
Testing for first time	112,386 (91)	125,904 (89)	238,290 (90)
Received results	123,491 (100)	141,462 (100)	264,953 (100)
Coverage and test results among individuals 15–49 years
Found at home and pretest counseled	101,866 (84)	117,825 (93)	219,691 (89)
Accepted testing and tested	93,991 (92)	110,988 (94)	204,979 (93)
Received results	93,983 (100)	110,987 (100)	204,970 (100)
HIV-infected	3,341 (3.6)	6,671 (6.0)	10,012 (4.9)

Table 1 also shows the number and proportion of people aged 15–49 years reached with HCT during the program period. Eighty-nine percent of all household members in this age group were found at home at the time of the visit. More females (93%) than males (84%) were likely to be found at home (OR 2.70, CI 2.63–2.78, p < 0.0001). All people 15–49 years found at home were offered HCT and 93% accepted testing. Females (94%) were more likely to accept testing than males (92%; OR 1.36, 95% CI 1.32–1.41, p < 0.0001). Overall, in the households reached, of 248,190 people aged 15–49 years, 204,979 (83%) tested for HIV during the program with females aged 15–49 years (88%) being more likely to test during the program compared to males (77%; OR 2.13, 95% CI 2.08–2.17, and p < 0.0001). The HIV prevalence was higher among females (6.0%) than among males (3.6%; OR 1.74, CI 1.69–1.76, p < 0.0001). The higher prevalence of HIV among women is attributed to earlier initiation of sex, low empowerment as well as high-risk male partners being more likely to have sex with low-risk females.^1,3

–7

To calculate the proportion of people in the district aged 15–49 years who tested during the program, we used district population estimates from the 2002 national census (interpolated to 2005 figures). There were 204,979 (58%) of an estimated 352,000 people aged 15–49 years who tested for HIV. The proportion of females who tested was 110,988 of 179,520 (61.8%) compared to 93991 of 172,480 (54.5%) of the males.

Couple testing

Table 2 shows the results of HIV testing by sexual partnerships. As seen in the table among the 238,017 people older than 14 years who tested for HIV, 139,395 (58.6%) were either married or cohabitating. 34,497 (54.5%) of the married or cohabiting males and 35,634 (46.8%) of the married or cohabiting females received results as a couple. In all, 35,634 (47%) of the sexual partnerships received results as a couple and 2.6% of all sexual partnerships were in an HIV discordant relationship. Among the sexual partnerships where at least one member was infected, 61.6% were in a discordant relationship.

Table 2.

Couple Testing Among People > 14 Years in Bushenyi District

Variable	Number (percent)
Married or cohabitating couples	76,117
Received results as a couple	35,634 (46.8%)
HIV test results
Concordant negative	34,133 (95.8%)
Concordant positive	576 (1.6%)
Discordant	925 (2.6%)
Male-infected discordant couples	616 (66.6%)
Female-infected discordant couples	309 (33.4%)
Couples with at least one HIV-infected partner who are discordant	925 (61.6%)

All the individual females who received results as a couple did so with only one sexual partner whereas 33,362 (96.7%) of the males received results with one sexual partner and 1136 (3.3%) received results with two sexual partners. A total of 1421 (4.1%) of the males who received results as a couple were HIV infected compared to 1501 (4.2%) of the females who received results as a couple. Among the 1785 individuals (860 men and 925 women) who were in discordant relationships 782 (44%) were HIV infected. Males (55%) were more likely than females (33%) to be the infected partner in an HIV-discordant relationship (OR 2.44, 95% CI 2.00–2.94, p < 0.0001).

Provision of care and prevention services

Table 3 shows the proportion of people receiving services. The majority of the people testing positive were initiated on cotrimaxzole prophylaxis, and received bed nets and water purification supplies. In Uganda these services are recommended for all HIV-infected persons irrespective or their CD4 counts.²⁴ In addition, approximately 11% of the HIV infected were initiated on ART and 10% referred for TB assessment. Of those evaluated for ART, 81% were initiated on ART. This high level of initiation is because ill people were more likely to be referred and evaluated. Generally females were more likely to receive services. Among the 796 people above 14 years of age who were tested for CD4 cell count enumeration, 31% had CD4 counts less than 200 cells per microliter, 23% between 200 and 350 cells per microliter, and 45% greater than 350 cells per microliter.

Table 3.

Number and Proportion of HIV-Infected People Receiving Services

Indicator	Male N = 4042 Number (%)	Female N = 7317 Number (%)	Total N = 11359 Number (%)
Initiated on cotrimoxazole prophylaxis	3790 (94)	7243 (99)	11,033 (97)
Received bed nets	2865 (71)	5562 (76)	8427 (74)
Received water purification system	2695 (67)	5232 (72)	7927 (70)
Referred to health units for TB diagnosis	411 (10)	692 (9.5)	1106 (9.7)
Evaluated for ART	523 (13)	941 (13)	1464 (13)
Initiated on ART	415 (10.3)	775 (10.6)	1190 (10.5)

Program costs

During the period of the economic evaluation 52,342 clients were offered HCT, at a cost of US$367,792. The major cost components included personnel (63% of total), HIV testing supplies (21% of total), and transportation (8% of total). HCT was found to cost $7.05 per client completing counseling and testing, $7.83 per previously untested client completing counseling and testing, and $139.32 per HIV-positive individual identified.

Discussion

Home-based HIV counseling and testing was well received; 94% of the population at home accepted testing, 58% of the total district population between 15 and 49 years received HIV testing, and 97% of those diagnosed with HIV initiated cotrimoxazole prophylaxis. The program provided test results to 11,359 HIV-infected people, 90% of whom had not known they were infected, and provided test results to 925 HIV-discordant couples.

The uptake of the program was high, reflecting a large unmet need for HIV testing and care services. Before the implementation of this program we had identified barriers associated with uptake of HCT in this district, including cost, accessibility, perceived low risk of HIV infection, and poor linkage to HIV care.¹⁹ WHO guidelines for HCT emphasize the removal of barriers in order to increase the proportion of persons undertaking HCT.²⁷ In this program, there was no cost to participants for receiving an HIV test, and physical accessibility was guaranteed by reaching people in their homes. The use of an intensive education program through local leaders and radio during which everyone was encouraged to test rather than those perceived to be at high risk probably reduced stigma. Linking HIV-infected people to care was also encouraged through provision of a basic care package, ongoing counseling and support, and referral for ART.

The shortage of health workers in Uganda and other low-income countries is often cited as one of the limiting factors influencing HIV program outcomes.^28
–30 As a result some authorities have argued that nonlaboratory personnel such as counselors may test clients for HIV.^25,31 In this program we employed laboratory qualified health workers (with a certificate in medical laboratory technology after 12 years of formal education) to do the HIV testing. Such workers were not recruited from formal health services. Most of these health workers were unemployed due to a freeze on recruitment of health workers in the country and due to the salary ceilings of public health workers in Uganda. All of the counselors employed had a nonmedical background. Although the counselors were trained in HIV testing and were allowed to do HCT, and laboratory workers were trained in counseling and allowed to do counseling, it later emerged that most of the time counselors did most of the pretest and posttest counseling while the laboratory assistants did the HIV testing.

During the implementation of this program, there was a high level of public-private cooperation between a nongovernmental organization (ICOBI) and the Bushenyi District Health Services. ICOBI was able to access funds very quickly and do procurement and recruitment of staff in a relatively short time; something that may not have been possible in the government health system. However, the public Bushenyi district health system was able to plan for and provide care for a considerable number of newly identified HIV-infected adults and children. Most of the patients who needed ART were provided with it free of charge. However, it is likely that a proportion of people who were clinically eligible for ART did not visit the health units for evaluation despite being encouraged to do so. The program assisted the public district health system in terms of supplies (such as cotrimoxazole), renovation of space for attending to the referred clients, and improving the skills of the health workers in attending to the referred clients.

The cost of providing home-based HCT in this program was relatively modest making home-based HCT strategy very inexpensive and effective at reaching population groups with low rates of prior testing. The costs of home based HCT are comparable to those of other HCT strategies reported in various other settings in low income countries.^32
–34 In the Ugandan setting where home-based district wide HCT was compared to other three HCT strategies (stand-alone HCT, hospital-based HCT, household-member HCT), home-based district wide HCT was the least expensive strategy per client tested.³² In addition to these program savings, home-based district wide HCT leads to considerable savings for clients if the services come to their home.

Although the number of people participating in the program was large, 100% district coverage was not achieved as the program funding had a finite timeline. In addition, a district census was not done meaning that methods used to calculate district coverage could only use projected population. Therefore the results, particularly HIV prevalence, may not be fully generalizable. People not found at home and those who refused testing (although a relatively small proportion of the population) may be different from those who participated. Moreover, people were informed when the testing would be done through multiple communication channels. It is possible that subjects who perceived themselves to be at risk or those that already knew their status and did not want it known within the family could absent themselves. It is even more difficult to generalize the high HIV prevalence in young children compared to adults because only children at risk were tested and the lower prevalence in early teens may be as a result of death in HIV infected children previously not identified and/or treated. Furthermore, due to shortages of water vessels and long-lasting insecticide-treated bed nets it was not possible to supply a basic care package to every HIV-infected person during the lifetime of the program.

This is the first report of district-wide program of home-based HCT. These data suggest that large-scale, home-based HCT can be implemented at a relatively modest cost with extremely high uptake. The importance of universal access and increased coverage of HIV testing is increasingly recognized for access to both prevention and care.^26,35,36 Home-based, district-wide programs may present an opportunity to help reach that goal. This large scale program found that 90% of HIV-infected people in this setting have never been tested. Thus, there is need for the public health community to be proactive in identifying people who are HIV infected.

Footnotes

Acknowledgments

We thank the staff of ICOBI and of the Bushenyi District Health services that took part in the data collection. This publication was supported by Cooperative Agreement Number U62/CCU024535-01 from the Emergency Plan for AIDS Relief (PEPFAR), Centers for Disease Control and Prevention (CDC). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of CDC.

Author Disclosure Statement

No competing financial interests exist.

References

Ministry of Health [Uganda] and ORC Macro. Uganda HIV/AIDS Sero-behavioural Survey 2004–2005. Uganda Ministry of Health and ORC Macro: Calverton, 2006. www.measuredhs.com/hivdata/. 2010 May 25.

Robinson

, Mulder

, Auvert

, Whitworth

, Hayes

. Type of partnership and heterosexual spread of HIV infection in rural Uganda: Results from simulation modelling. Int J STD AIDS, 1999; 10:718–725.

Nuwaha

, Faxelid

, Hojer

. Predictors for condom use among patients with sexually transmitted diseases in Uganda. Sex Transm Dis, 1999; 26:490–495.

Nuwaha

, Faxelid

, Neema

, Hojer

. Lay people's perception of sexually transmitted infections in Uganda. Int J STD AIDS, 1999; 10:709–717.

Serwadda

, Gray

, Wawer

et al. The social dynamics of HIV transmission as reflected through discordant couples in rural Uganda. AIDS, 1995; 9:745–750.

Hugonnet

, Mosha

, Todd

et al. Incidence of HIV infection in stable sexual partnerships: a retrospective cohort study of 1802 couples in Mwanza Region, Tanzania. J Acquir Immune Defic Syndr, 2002; 30:73–80.

Carpenter

, Kamali

, Ruberantwari

, Malamba

, Whitworth

. Rates of HIV-1 transmission within marriage in rural Uganda in relation to the HIV sero-status of the partners. AIDS, 1999; 13:1083–1089.

Marks

, Crepaz

, Senterfitt

, Janssen

. Meta-analysis of high-risk sexual behavior in persons aware and unaware they are infected with HIV in the United States: Implications for HIV prevention programs. J Acquir Immune Defic Syndr, 2005; 39:446–453.

Allen

, Serufilira

, Bogaerts

et al. Confidential HIV testing and condom promotion in Africa. Impact on HIV and gonorrhea rates. JAMA, 1992; 268:3338–3343.

10.

Allen

, Tice

, Van de Perre

et al. Effect of serotesting with counselling on condom use and seroconversion among HIV discordant couples in Africa. BMJ, 1992; 304:1605–1609.

11.

Were

, Mermin

, Wamai

et al. Undiagnosed HIV infection and couple HIV discordance among household members of HIV-infected people receiving antiretroviral therapy in Uganda. J Acquir Immune Defic Syndr, 2006; 43:91–95.

12.

Kamb

, Fishbein

, Douglas

Jr.

et al. Efficacy of risk-reduction counselling to prevent human immunodeficiency virus and sexually transmitted diseases: A randomized controlled trial. Project RESPECT Study Group. JAMA, 1998; 280:1161–1167.

13.

The VCT Efficacy Study Group. Efficacy of voluntary HIV-1 counselling and testing in individuals and couples in Kenya, Tanzania, and Trinidad: A randomised trial. The Voluntary HIV-1 Counselling and Testing Efficacy Study Group. Lancet, 2000; 356:103–112.

14.

Allen

, Meinzen-Derr

, Kautzman

et al. Sexual behavior of HIV discordant couples after HIV counselling and testing. AIDS, 2003; 17:733–740.

15.

Kamenga

, Ryder

, Jingu

et al. Evidence of marked sexual behavior change associated with low HIV-1 seroconversion in 149 married couples with discordant HIV-1 serostatus: Experience at an HIV counselling center in Zaire. AIDS, 1991; 5:61–67.

16.

Sweat

, Gregorich

, Sangiwa

et al. Cost-effectiveness of voluntary HIV-1 counselling and testing in reducing sexual transmission of HIV-1 in Kenya and Tanzania. Lancet, 2000; 356:113–121.

17.

Bunnell

, Mermin

, De Cock

. HIV prevention for a threatened continent: Implementing positive prevention in Africa. JAMA, 2006; 296:855–858.

18.

Uganda Bureau of Statistics (UBOS) and Macro International Inc. Uganda demographic health survey 2006. Calverton, MD: UBOS and Macro International Inc, 2007. www.measuredhs.com/hivdata/. 2010 May .

19.

Nuwaha

, Kabatesi

, Muganwa

, Whalen

. Factors influencing acceptability of voluntary counselling and testing for HIV in Bushenyi district of Uganda. East Afr Med J, 2002; 79:626–632.

20.

Were

, Mermin

, Bunnell

, Ekwaru

, Kaharuza

. Home-based model for HIV voluntary counselling and testing. Lancet, 2003; 361:1569.

21.

De Cock

, Mbori-Ngacha

, Marum

. Shadow on the continent: public health and HIV/AIDS in Africa in the 21 st century. Lancet, 2002; 360:67–72.

22.

Uganda Bureau of Statistics (UBOS). The Uganda National Household Survey 2005/06: Report on the Socio-economic module. Kampala: Uganda, 2006.

23.

Lule

, Mermin

, Ekwaru

et al. Effect of home-based water chlorination and safe storage on diarrhea among persons with human immunodeficiency virus in Uganda. Am J Trop Med Hyg, 2005; 73:926–933.

24.

Mermin

, Lule

, Ekwaru

et al. Effect of co-trimoxazole prophylaxis on morbidity, mortality, CD4-cell count, and viral load in HIV infection in rural Uganda. Lancet, 2004; 364:1428–1434.

25.

Ministry of Health [Uganda]. Uganda National Policy on HIV Counselling and Testing. Kampala: Uganda, 2005.

26.

World Health Organization (WHO). National AIDS programmes: A guide to indicators for monitoring and evaluating national HIV prevention programmes. Geneva, Switzerland, 2004.

27.

UNAIDS. WHO and UNAIDS Secretariat Statement on HIV Counselling and Testing, August 14, 2006 Geneva, Switzerland, 2006.

28.

Chen

, Evans

, Anand

et al. Human resources for health: overcoming the crisis. Lancet, 2004; 364:1984–1990.

29.

Anyangwe

, Mtonga

. Inequities in the global health workforce: The greatest impediment to health in sub-Saharan Africa. Int J Environ Res Public Health, 2007; 4:93–100.

30.

Hongoro

, McPake

. How to bridge the gap in human resources for health. Lancet, 2004; 364:1451–1456.

31.

Schneider

, Blaauw

, Gilson

, Chabikuli

, Goudge

. Health systems and access to antiretroviral drugs for HIV in Southern Africa: Service delivery and human resources challenges. Reprod Health Matters, 2006; 14:12–23.

32.

Menzies

, Abang

, Wanyeze

et al. The costs and effectiveness of four HIV counselling and testing strategies in Uganda. AIDS, 2009; 23:395–401.

33.

Thielman

, Chu

, Ostermann

et al. Cost-effectiveness of free HIV voluntary counselling and testing through a community-based AIDS service organization in Northern Tanzania. Am J Pub Health, 2006; 96:114–119.

34.

Aisu

, Raviglione

, van Praag

et al. Preventive chemotherapy for HIV-associated tuberculosis in Uganda: an operational assessment at a voluntary counselling and testing centre. AIDS, 1995; 9:267–273.

35.

Appiah

, Havers

, Gibson

, Kay

, Sarfo

, Chadwick

. Efficacy and acceptability of rapid, point-of-care HIV testing in two clinical settings in Ghana. AIDS Patient Care STDs, 2009; 23:365–369.

36.

Kiene

, Bateganya

, Wanyenze

, Lule

, Nantaba

, Stein

. AIDS Patient Care and STDs, 2010; 24:117–126.