The Joint United Nations Program on HIV/AIDS (UNAIDS) estimated that there were 22.5 million people living with HIV in sub-Saharan Africa in 2007 . Though large progress has been made in implementing HIV prevention programs, existing efforts have been insufficient to reduce HIV incidence in some countries . Historically, HIV prevention programs have focused on reducing HIV acquisition risk among those not infected or those with unknown HIV status rather than on reducing transmission risk from those already infected. However, with increased antiretroviral therapy (ART) access, HIV-infected people are living longer and have healthier lives, raising concerns of a potential increase in transmission risk from HIV-infected [2–4]. UNAIDS recommends positive prevention (prevention programs for HIV-infected to reduce transmission of HIV from those already infected) as an essential component of comprehensive HIV prevention [5,6].
Several studies have documented substantial transmission risk from people living with HIV/AIDS (PLWHAs) [7,8]. In Mozambique, patients initiating ART in a public hospital clinic reported that 65% of their sexual acts in the past 3 months were unprotected . A number of interventions with PLWHAs, based on increasing awareness and offering prevention and partner counseling, are associated with reduced HIV transmission risk [9–11]. In Uganda, knowledge of HIV status was associated with a three-fold increase in condom use in a nationally representative survey , and among patients on ART, prevention counseling and partner voluntary counseling and testing (VCT) was associated with a 70% reduction in sexual risk behavior and 98% reduction in HIV transmission risk .
The World Health Organization (WHO) lists preventing unintended pregnancies among PLWHAs as an important component of preventing mother-to-child transmission of HIV . In Abidjan, Cote d'Ivoire, among postpartum HIV-infected women, 51% experienced an unwanted pregnancy over 18 months . Nevertheless, studies have demonstrated success in reducing unwanted pregnancies among HIV-infected women. In Malawi, receipt of HIV test results among HIV-infected women was associated with an increase in contraceptive use from 38 to 46% and pregnancy incidence was lower among women not desiring more children .
Although positive prevention has been recommended for Africa [15,16] and is being expanded, interventions will be most effective when based on country-specific and cultural-specific data on PLWHAs. We analyzed nationally representative survey data, prior to the expansion supported by the President's Emergency Plan for AIDS Relief (PEPFAR) and the Global Fund to fight AIDS, Tuberculosis and Malaria (GFATM), from Kenya (2003) and Malawi (2004–2005) to determine sexual risk behaviors, awareness of HIV status and contraceptive use among HIV-infected persons.
We analyzed two nationally representative cross-sectional surveys: the 2003 Kenya Demographic and Health Survey (KDHS) and the 2004–2005 Malawi Demographic and Health Survey (MDHS) .
Survey designs and ethical review
The 2003 KDHS was designed to be representative at the national and provincial levels and for urban and rural areas . From the households selected for sampling, all women (15–49 years of age) in the selected households and all men (15–54 years of age) in every second household were eligible for interview. The 2004–2005 MDHS was designed to be representative at the national and regional levels, rural and urban areas and selected large districts . From the households selected for sampling, all women (15–49 years of age) in the selected households and all men (15–54 years of age) in every third household were eligible for the interview.
In KDHS and MDHS, female and male participants were eligible for HIV testing only in those households selected for men's interview. Both surveys collected blood samples for HIV testing from eligible and consenting survey participants. Survey participants had the right to refuse blood collection and participate only in the survey. Separate informed consents documented voluntary participation for the interview and blood collection for each participant. Personal identifiers were removed during merging of laboratory, household and individual data. All participants were provided information and referral vouchers for VCT services for free HIV testing. Additional details of the survey design are presented elsewhere [18,19]. KDHS was approved by the Macro International, Incorporation Institutional Review Board (IRB) and Scientific Steering and National Ethical Committee of Kenya Medical Research Institute. MDHS was approved by the National Health Sciences Research Committee of the Ministry of Health and Population. Both were reviewed and approved by the US Centers for Disease Control and Prevention (CDC).
In KDHS, interviews were completed by 8195 women (94% of eligible) and 3578 men (86% of eligible), and 76% of 4303 eligible women and 70% of 4183 eligible men agreed to participate in HIV testing. In MDHS, interviews were completed with 11 698 women (95.7% of eligible) and 3261 men (85.9% of eligible), and 70% of 4071 eligible women and 63% of 3797 eligible men agreed to participate in HIV testing. On the basis of the survey, national HIV prevalence among 15–49-year olds was 6.7% (women: 8.7%; men: 4.6%) in Kenya and 11.8% (women: 13.3%; men: 10.2%) in Malawi [18,19]. The following analysis was limited to HIV-infected persons identified in the survey (weighted total: Kenya: 412; Malawi: 664) based on HIV test results from the survey.
Measures and analysis
Analysis was performed using the survey feature (svy) of Stata Release 9.2  after specifying the design of the survey. We used individual sampling weight of HIV testing as the analysis was restricted to PLWHAs. PLWHAs who reported having had an HIV test prior to the survey and had received their test results were classified as aware of their HIV status. Participants who reported they had sex in the past month were classified as recently sexually active. For each country, we constructed four separate multivariate logistic regression models to determine factors independently associated with awareness of HIV status, recent sexual activity, condom use at last sexual intercourse and use of modern contraceptives. We followed a three-step process in constructing multivariate models. Variables with univariate P values of 0.1 or less were included in the initial multivariate models. Only those variables that were found to be significant were included in the final multivariate models. All models were adjusted for age, sex and residence. In all models, we tested the interaction of sex with all other variables in the model but the interaction term was included in the final multivariate model only if it was significant. The presented tables list variables included in the final multivariate model. Presented confidence intervals (CIs) are 95% CIs.
Characteristics of people living with HIV/AIDS
The majority of PLWHAs were either married or cohabitating (Kenya: 62%; Malawi: 78%) (Table 1).
Sexual risk behaviors
Among PLWHAs sexually active in the last month, 86% in Kenya and 96% in Malawi reported their spouse or cohabiting partner as their most recent partner. In multivariate analysis of being sexually active in the past month, Kenyan women residing in rural areas [adjusted odds ratio (AOR): 0.18; CI: 0.07, 0.49] and Kenyan men aware of their HIV status (AOR: 0.17; CI: 0.06, 0.49) were significantly less likely to be sexually active (Table 2). Also in multivariate analysis, married or cohabiting HIV-infected participants were significantly more likely to be sexually active as compared with never married participants in Kenya (AOR: 36.49; CI: 13.55, 98.27) as well as in Malawi (AOR: 13.40; CI: 3.62, 49.65). In addition, in Kenya, widowed and divorced PLWHAs were significantly more likely to be sexually active than never married PLWHAs (AOR: 5.96; CI: 1.59, 22.39).
Of the PLWHAs whose last intercourse was with their spouse or cohabiting partner, 3% in Kenya and 6% in Malawi used a condom at their last intercourse. In multivariate analysis of condom use at last sexual encounter, Kenyan PLWHAs who were married or had cohabiting partners (AOR: 0.04; CI: 0.01, 0.13) or those who were widowed or divorced (AOR: 0.10; CI: 0.01, 0.89) were less likely to use condom at last sexual encounter (Table 3). In Malawi, PLWHAs who were married or living with a partner (AOR: 0.04; CI: 0.01, 0.18) were significantly less likely to use a condom at last sexual intercourse.
Awareness of HIV status
Overall, 20 and 16% of PLWHAs in Kenya and Malawi, respectively, were aware of their HIV status. Among those tested prior to the survey, 89% in Kenya and 87% in Malawi reported receiving their test results. In Kenya and Malawi, among HIV-infected persons not aware of their status, 64 and 86% knew a place to obtain an HIV test, respectively. In Kenya, 72% of PLWHAs believed that they were at no or small risk of acquiring HIV (data not available for Malawi). In Kenya, PLWHAs with at least secondary education (AOR: 2.50; CI: 1.33, 4.68) were significantly more likely to be aware of their HIV status (Table 4).
Pregnancy, contraception and desire for more children
Thirty-three percent and 66% of HIV-infected women in Kenya and Malawi, respectively, had received HIV/AIDS information at last pregnancy from a health worker (Table 1). In Malawi, only 14% were tested for HIV at their last pregnancy. Fifty-four percent of HIV-infected women in Kenya and 40% in Malawi reported that their last child was either unplanned or unwanted. Overall contraceptive use was low with 26% of HIV-infected women in Kenya and 19% in Malawi using modern contraceptives. Of the nearly three-quarters of HIV-infected women who did not want a child ever or in the next 2 years, less than one-third in Kenya and less than one-fifth in Malawi were using modern contraceptives.
In multivariate analysis, HIV-infected women in Kenya who did not desire more children (AOR: 3.73; CI: 1.45, 9.64) were significantly more likely to be using modern contraceptives as compared with those who wanted a child within 2 years (Table 5). In Malawi, HIV-infected women who had three to five living children (AOR: 3.84; CI: 1.69, 8.72) or those who had more than five living children (AOR: 7.87; CI: 2.62, 23.62) were significantly more likely to be using modern contraceptives.
Our results from Kenya and Malawi demonstrate the urgent need to improve HIV testing uptake, which is a necessary first step to access prevention, care and treatment services. Over 80% of PLWHAs in Kenya and Malawi were unaware of their HIV status. Because the majority of PLWHAs who were not aware of their status already knew of a place to obtain a HIV test and, in Kenya, perceived themselves to be at low-risk of HIV acquisition , HIV program planners may need to implement innovative approaches to increase utilization of HIV testing services.
Several testing approaches have been associated with high uptake in other countries such as home-based HIV testing and routine HIV testing (RHT) in clinical facilities [22–25]. In Bushenyi district of Uganda, a home-based VCT program started in 2005 increased knowledge of HIV status from less than 10 to 84% in less than a year . In another study in rural south-western Uganda, in which blood for HIV testing was collected at participant's home, home-delivery of HIV testing results was associated with a two-to-five-fold increase in HIV test uptake than that observed when participants had to visit sites to obtain test results . In 2004, Botswana introduced RHT, which has rapidly improved HIV test uptake (from 36/1000 persons in 2004 to 93/1000 persons in 2005) and since the start of the program, the proportion of patients starting ART with CD4 cell counts of 100 cells/μl or less has decreased .
We found that one-third of HIV-infected women in Kenya and two-thirds in Malawi had received information on HIV/AIDS in their last pregnancy. Of the pregnant women, about 88% in Kenya and 93% in Malawi receive antenatal care from medical professionals [18,19]; the majority receive care from nurse/mid-wife, who should be trained to provide HIV testing and counseling along with antenatal care.
We found that being married or having a cohabiting partner was strongly associated with sexual activity and lack of condom use, with over 90% of last sexual acts being unprotected. Of all married or cohabiting HIV-infected persons, 50% in Kenya and 41% in Malawi have an HIV-negative spouse [18,19], highlighting the importance of ensuring that couples receive HIV prevention programs. Although it may be challenging to design prevention strategies for discordant couples , they represent a significant group as most new infections in sub-Saharan Africa occur in discordant couples [29,30]. A study in Uganda found that disclosure of status to partners promoted safer sex and increased care-seeking behavior . Studies in discordant couples have reported increased condom use and safer sexual behavior with male-focused or joint couple counseling [32,33].
Our findings underscore the need for integrating family planning services with HIV services. There was a large unmet need for temporary and permanent contraceptive methods. Family planning services should be tailored to the specific requirements of HIV-infected persons and emphasize the need for barrier contraceptives in discordant couples.
Our findings have several important limitations. First, these findings reflect the situation in Kenya in 2003 and in Malawi in 2003–2004. Since that time, there has been a large scale-up of HIV prevention, care and treatment services in part due to increased support from the GFATM, PEPFAR and other donors. In Kenya, the number of VCT sites has increased from three in 2000 to 680 in 2005 . Second, because population-survey data are self-reported, it may be affected by social desirability bias. Participants may underreport risky sexual behaviors and may not be comfortable disclosing if they have been tested for HIV. The results may also be biased due to nonresponse in the survey (refusal to participate in the survey and refusal to participate in HIV testing). We may have underestimated the effect of knowledge of HIV status, as we assumed that PLWHAs who had received test results in the past were aware of their HIV infection. It is possible that they were tested prior to seroconversion. Because the survey is cross-sectional, the temporal sequence between independent and dependent variables in the regression models cannot be determined. The survey did not collect information on whether the participants were currently sick or incapacitated due to HIV disease progression, an important confounder for several possible associations in our regression models. Finally, the survey was not adequately powered for this subset analysis of HIV-infected persons, which led to small cell sizes and wide CIs.
In summary, the most recent publicly available population-based data for Kenya and Malawi suggest that most HIV-infected persons are unaware of their HIV infection, do not use condoms and presumably are not receiving appropriate HIV-related care and treatment services. To reduce the risk of transmission from HIV-infected persons, not only does HIV testing coverage need to increase, but also testing and counseling services that incorporate prevention strategies with proven efficacy for couples, especially married or cohabiting discordant couples, need to be promoted [35,36]. An urgent need is to reduce risk behaviors within discordant couples and the design of these prevention interventions should be a high priority. It is likely that the situation is similar in many other countries . Implementation of comprehensive positive prevention interventions could have a large impact on HIV incidence in sub-Saharan Africa.
A.A., R.W.S. and R.E.B. designed the concept and the analytical plan. L.H.M., J.N.M., K.K. and J.M.A. participated in data collection. All authors participated in data analysis, interpretation of results and preparation of article. All authors have approved the final version.
The authors thank all participants in Kenya Demographic and Health Survey (KDHS) and Malawi Demographic and Health Survey (MDHS). The authors are also grateful to the Ministry of Health of Kenya and Malawi and to KDHS and MDHS staff who conducted data collection, laboratory testing and data management.
Support for this analysis was provided by the Epidemic Intelligence Service Program in Atlanta, USA and the President's Emergency Plan for AIDS Relief in Atlanta, USA and Nairobi, Kenya.
The findings and conclusions in this paper are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
The authors declare no conflicts of interest.
1. Joint United Nations Program on HIV/AIDS. 2007 AIDS epidemic update
. Geneva, Switzerland: UNAIDS; 2007.
2. Scheer S, Chu PL, Klausner JD, Katz MH, Schwarcz SK. Effect of highly active antiretroviral therapy on diagnoses of sexually transmitted diseases in people with AIDS. Lancet 2001; 357:432–435.
3. Vlahov D, Safaien M, Lai S, Strathdee SA, Johnson L, Sterling T, et al
. Sexual and drug risk-related behaviours after initiating highly active antiretroviral therapy among injection drug users. AIDS 2001; 15:2311–2316.
4. Auvert B, Males S, Puren A, Taljaard D, Carael M, Williams B. Can highly active antiretroviral therapy reduce the spread of HIV?: a study in a township of South Africa. J Acquir Immune Defic Syndr 2004; 36:613–621.
5. UNAIDS. Intensifying HIV prevention – UNAIDS policy position paper. http://data.unaids.org/publications/irc-pub06/jc1165-intensif_hiv-newstyle_en.pdf
. [Accessed 05 June 2007].
6. UNAIDS. 2006 Report on the global AIDS epidemic. http://www.unaids.org/en/HIV_data/2006GlobalReport/default.asp
. [Accessed 24 September 2007].
7. Pearson CR, Kurth AE, Cassels S, Martin DP, Simoni JM, Hoff P, et al
. Modeling HIV transmission risk among Mozambicans prior to their initiating highly active antiretroviral therapy. AIDS Care 2007; 19:594–604.
8. Simbayi LC, Kalichman SC, Strebel A, Cloete A, Henda N, Mqeketo A. Disclosure of HIV status to sex partners and sexual risk behaviours among HIV-positive men and women, Cape Town, South Africa. Sex Transm Infect 2007; 83:29–34.
9. Farquhar C, Kiarie JN, Richardson BA, Kabura MN, John FN, Nduati RW, et al
. Antenatal couple counseling increases uptake of interventions to prevent HIV-1 transmission. J Acquir Immune Defic Syndr 2004; 37:1620–1626.
10. Bunnell R, Ekwaru JP, Solberg P, Wamai N, Bikaako-Kajura W, Were W, et al
. Changes in sexual behavior and risk of HIV transmission after antiretroviral therapy and prevention interventions in rural Uganda. AIDS 2006; 20:85–92.
11. Bunnell R, Opio A, Musinguzi J, Kirungi W, Ekwaru P, Mishra V, et al
. HIV transmission risk behavior among HIV-infected adults in Uganda: results of a nationally representative survey. AIDS 2008; 22:617–624.
12. World Health Organization. Strategic approaches to the prevention of HIV infection in infants
. Switzerland, Geneva; 20–22 March 2002.
13. Desgrees-Du-Lou A, Msellati P, Viho I, Yao A, Yapi D, Kassi P, et al
. Contraceptive use, protected sexual intercourse and incidence of pregnancies among African HIV-infected women. DITRAME ANRS 049 Project, Abidjan 1995–2000. Int J STD AIDS 2002; 13:462–468.
14. Hoffman IF, Martinson FE, Powers KA, Chilongozi DA, Msiska ED, Kachipapa EI, et al
. The year-long effect of HIV-positive test results on pregnancy intentions, contraceptive use, and pregnancy incidence among Malawian women. J Acquir Immune Defic Syndr 2008; 47:477–483.
15. Bunnell R, Mermin J, De Cock KM. HIV prevention for a threatened continent: implementing positive prevention
in Africa. JAMA 2006; 296:855–858.
16. International HIV/AIDS Alliance. Positive prevention: HIV prevention with people living with HIV
. Brighton, England: International HIV/AIDS Alliance; 2007.
17. Macro International Incorporation. Measure DHS home. http://www.measuredhs.com/
. [Accessed 03 March 2008].
18. Central Bureau of Statistics (CBS) [Kenya] Ministry of Health (MOH) [Kenya] and ORC Macro. Kenya Demographic and Health Survey 2003
. Calverton, Maryland: CBS, MOH and ORC Macro; 2004.
19. National Statistical Office (NSO) [Malawi] and ORC Macro. Malawi Demographic and Health Survey 2004
. Calverton, Maryland: NSO and ORC Macro; 2005.
20. Stata Statistical Software [computer program]. Version 9.2. College Station, Texas: StataCorp.; 2005.
21. Johnson K, Way A. Risk factors for HIV infection in a national adult population: evidence from the 2003 Kenya Demographic and Health Survey. J Acquir Immune Defic Syndr 2006; 42:627–636.
22. Wolff B, Nyanzi B, Katongole G, Ssesanga D, Ruberantwari A, Whitworth J. Evaluation of a home-based voluntary counselling and testing intervention in rural Uganda. Health Policy Plan 2005; 20:109–116.
23. Were WA, Mermin JH, Wamai N, Awor AC, Bechange S, Moss S, 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.
24. Matovu JK, Kigozi G, Nalugoda F, Wabwire-Mangen F, Gray RH. The Rakai project counselling programme experience. Trop Med Int Health 2002; 7:1064–1067.
25. Were W, Mermin J, Bunnell R, Ekwaru JP, Kaharuza F. Home-based model for HIV voluntary counselling and testing. Lancet 2003; 361:1569.
26. Nuwaha F, Muganzi E, Kasasa S, Achom M, Bunnell R, Kabatesi D, et al
. District-wide, door-to-door, home-based HIV voluntary counselling and testing in rural Uganda.
In: XVI International AIDS Conference
; Toronto, Canada; 2006.
27. Steen TW, Seipone K, Gomez Fde L, Anderson MG, Kejelepula M, Keapoletswe K, et al
. Two and a half years of routine HIV testing in Botswana. J Acquir Immune Defic Syndr 2007; 44:484–488.
28. Bunnell RE, Nassozi J, Marum E, Mubangizi J, Malamba S, Dillon B, et al
. Living with discordance: knowledge, challenges, and prevention strategies of HIV-discordant couples in Uganda. AIDS Care 2005; 17:999–1012.
29. Guthrie BL, de Bruyn G, Farquhar C. HIV-1-discordant couples in sub-Saharan Africa: explanations and implications for high rates of discordancy. Curr HIV Res 2007; 5:416–429.
30. Dunkle KL, Stephenson R, Karita E, Chomba E, Kayitenkore K, Vwalika C, et al
. New heterosexually transmitted HIV infections in married or cohabiting couples in urban Zambia and Rwanda: an analysis of survey and clinical data. Lancet 2008; 371:2183–2191.
31. King R, Katuntu D, Lifshay J, Packel L, Batamwita R, Nakayiwa S, et al
. Processes and outcomes of HIV serostatus disclosure to sexual partners among people living with HIV in Uganda. AIDS Behav 2007; 12:232–243.
32. Roth DL, Stewart KE, Clay OJ, van Der Straten A, Karita E, Allen S. Sexual practices of HIV discordant and concordant couples in Rwanda: effects of a testing and counselling programme for men. Int J STD AIDS 2001; 12:181–188.
33. Allen S, Meinzen-Derr J, Kautzman M, Zulu I, Trask S, Fideli U, et al
. Sexual behavior of HIV discordant couples after HIV counseling and testing. AIDS 2003; 17:733–740.
34. Marum E, Taegtmeyer M, Chebet K. Scale-up of voluntary HIV counseling and testing in Kenya. JAMA 2006; 296:859–862.
35. Karita E, Chomba E, Roth DL, Telfair J, Zulu I, Clark L, 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.
36. Chomba E, Allen S, Kanweka W, Tichacek A, Cox G, Shutes E, et al
. Evolution of couples' voluntary counseling and testing for HIV in Lusaka, Zambia. J Acquir Immune Defic Syndr 2008; 47:108–115.