A risk score was developed using 4 demographic variables that remained independent predictors of HIV in logistic regression analysis. A score of 0 was given to a couple who had no demographic risk factors, with 1 point assigned to each risk factor and summed for a score (range: 0 to 4). Risk score as a predictor of concordant positive, discordant, or concordant negative serostatus was assessed with the objective of providing a way for service providers to target high-risk couples for CVCT (Table 3).
Participation in the CVCT Program Over Time
Figure 1 presents the distribution of couples that attended ZEHRP by couple HIV serostatus between August 1994 and March 2000. In the first 10 months of CVCT, when promotion was limited to mass media and counselor visits to local organizations, an average of 76 couples per month tested.
Rapid HIV testing with same-day counseling was implemented in June 1995, along with recruitment of community workers to promote couples' testing in their neighborhoods; 230 couples per month were tested between June and December 1995, increasing to 309 couples per month during 1996. In January 1997, the CVCT program was reduced from 6 days to 3 days per week to allow time for other follow-up activities. Between January and June of that year, the average number of couples dropped to 196 per month, and then to 152 per month from July to December 1997. When recruitment for prospective studies was completed in July 1998, active community promotion of couples' testing ceased.
Although radio announcements and newspaper advertisements continued, the average number of CVCT couples dropped to 33 per month between August and November 1998. The center was closed from December 1998 until June 1999 and reopened with CVCT services provided to walk-in couples until March 2000. With promotion limited to radio and newspaper announcements, only 20 couples per month sought CVCT during this time. Almost 80% of couples had at least 1 HIV-positive partner in this small group, possibly due to self-selection among symptomatic individuals or referral of patients seeking care from local clinics.
Demographic Distributions Stratified by Couple Serostatus
Of the couples tested, 54% were concordant negative for HIV (M−F−), 26% were concordant positive (M+F+), 11% had an HIV-positive man and an HIV-negative woman (M+F−), and 12% had an HIV-negative man and an HIV-positive woman (F+M−). The mean age was 34 years for men and 27 years for women, and mean duration of union was 7 years. On average, women had delivered 3 times and couples had 2 living children. Overall, 15% of women were pregnant and 36% were breast-feeding. The prevalence of HIV in men and women stratified by demographic characteristics is shown in the first 2 columns of Table 1.
Interestingly, only 2% of men <20 years old were HIV positive, compared with 21% of women <18 years of age. HIV prevalence was highest in men aged 30 to 39 and women aged 25 to 29 and 30 to 34. In >95% of couples, the man was older than his wife, with more than half having at least a 5-year age difference. HIV prevalence was not strongly associated with age difference, except in the 2.5% of couples in which the woman was older than the man. Unless otherwise noted, these and subsequent comparisons are statistically significant at a 95% confidence interval (CI).
The relationship of HIV status to length of union was not linear and differed in men and women. Among women, prevalence was stable at 40% to 43% for unions 1 to 8 years and dropped to 27% for longer unions. Among men, prevalence increased steadily with duration of union from 36% (<3 years) to 43% (6 to 8 years), then, as with women, dropped to 30% for unions >8 years.
Distributions of demographic characteristics within couple serostatus are given in the columns to the right in Table 1. Among concordant HIV-negative couples, 34% had cohabited for more than 8 years, compared with 16% to 26% of couples with 1 or both partners HIV positive. Two thirds of concordant positive, concordant negative, and M+F− couples had cohabited for ≥3 years, compared to less than half of F+M− couples. F+M− couples were not younger than other couples; the reason for this difference is not clear and warrants further investigation.
The highest HIV prevalence was observed among women with fewer than 2 children. The proportion of women who were pregnant was also lower among HIV-positive women (12%) than HIV-negative women (16%) (not shown). In comparison, 12% of women surveyed in the 1996 Zambia Demographic and Health Survey were pregnant.19
In concordant HIV-positive couples, the prevalence of syphilis in both partners was high (27% to 30%; not shown), whereas the corresponding prevalence in concordant negative couples was much lower (13% in men and women). Interestingly, the prevalence of positive RPR was similar in discordant couples with HIV-positive men (21% in men and 18% in their HIV-negative spouses), whereas HIV-positive women in discordant couples had a substantially higher prevalence of syphilis (32%) than their HIV-negative partners (18%). A positive syphilis serology was associated with high HIV prevalence, with an OR of 2.6 (95% CI: 2.3 to 2.8) for women and corresponding OR of 2.1 (95% CI: 1.9 to 2.4) for men.
Demographic Risk Factors for Unions With 1 or Both Partners Positive
We combined M+F+ couples and discordant couples and compared them to concordant negative couples using the demographic variables as risk factors for being in an HIV-positive union (Table 2). Both the man's age (30 to 39) and the woman's age (25 to 34) were independently associated with being in a positive union. This effect was neither enhanced nor lost by logistic regression. A union of <3 years remained independently predictive of HIV, though the OR was reduced in the multivariate analysis. Couples with fewer than 2 children were more likely to be in an HIV-positive relationship, independent of age or duration of union.
A risk score was created that assigned 1 point for each of the 4 demographic variables that were independent predictors of HIV in the couple. Table 3 and Figure 2 illustrate that there is an increased risk of being in a discordant or concordant positive union for every increase in score. Whereas a score of 0 was associated with a comparatively lower prevalence of having 1 or both partners HIV positive (29%), only 17% of couples had this low score. A risk score of 3 or 4 was associated with a substantially higher prevalence of HIV (72% to 78% with 1 or both partners HIV positive), but again, only 17% of couples met these criteria. Two thirds of couples had a risk score of 1 or 2, and 45% of these couples had at least 1 HIV-positive partner. There are clearly demographic factors associated with elevated risk of HIV. In this population with high HIV prevalence overall, however, it is not possible to develop a sensitive and specific screening tool based only on demographic characteristics.
CVCT is the most effective HIV prevention intervention for cohabitating couples in Africa, the largest group at risk for HIV infection in the world.20 Although this has been known for many years,5,21,22 very few couples in high-prevalence areas have been tested together. VCT is becoming increasingly available with the advent of perinatal prevention efforts and antiretroviral medication provision programs, but these programs prioritize the individual and generally do not accommodate couples.23-25 Disclosure is a primary challenge in this context,26,27 which can be overcome if both partners are counseled jointly. Most serosurveys in high-prevalence areas present risk factors for individuals,28 with the few that analyze couples presenting relatively small numbers.18,29 Couples are the most appropriate unit of analysis and intervention in reproductive health;5,9,10,13-15,24-26,30-34 our experiences have demonstrated that couples will come together for testing if financial and logistical obstacles are overcome.18,35
Previous studies found knowledge about CVCT to be poor among prospective clients,18,35 prompting us to use a variety of promotional strategies to publicize the CVCT program. Although our overtures to local employers and organizations were warmly received, they did not result in increased attendance at our CVCT program. This may have been due to couples' fear that test results would be disclosed to employers if they received services from a group recommended by their supervisors; further investigation is needed to confirm this. Similarly, radio announcements and newspaper advertisements alone did not result in an increase in couple attendance. There are many radio stations in Lusaka representing several of the 73 Zambian languages. Newspapers are published in English, which relatively few adults read. Mass media may be more effective where major-language channels and print media are popular. Door-to-door efforts with community workers resulted in substantial increases in attendance, but this peer model was labor and cost intensive and the impact was not sustained. When research targets were achieved, outreach by community workers was discontinued and attendance dropped by more than 90%. Saturation is not a likely explanation; the 17,000 individuals tested during this time represented only 1% of Lusaka's 1.7 million inhabitants. Successful promotion of CVCT must include widespread advocacy from community and nongovernmental organizations, support and endorsement from business and political leaders, and efforts by community leaders to reduce the stigma surrounding HIV.
Most couples that came for HIV counseling and testing were young, had been married for more than 3 years, and had at least 2 children. The prevalence of HIV in women was 38% and 36% in men. Forty-five percent of couples had at least 1 HIV-positive partner, and 23% were HIV discordant. The HIV prevalence among young wives (16 to 17 years old) in this study was 90% higher than among men of a slightly older age group (18 to 19 years old). This is similar to the previously published finding that HIV prevalence in women was 6 times that in men among sexually active 15- to 19-year-olds in Kisumu, Kenya, and Ndola, Zambia.36
HIV-discordant couples with HIV-positive women had a unique profile in several ways. The women were more likely to be older or close in age to their partner than other couples. Over half had been in unions for <3 years, and 70% had fewer than 2 children. In addition, although men and women in concordant unions and discordant unions with HIV-positive men had similar prevalence of syphilis infection, in discordant couples with HIV-positive women, a significantly higher prevalence of syphilis was found in the female partner (32%) compared with their partners (18%). This suggests that HIV-positive women in F+M− couples have latent syphilis and had passed the contagious period when they entered their current union.
Our research goal was to recruit discordant couples, and we had hoped to develop a demographic profile that would allow us to target promotional strategies to couples at highest risk of HIV. Although having at least 3 of 4 risk factors-man's age, woman's age, duration of union, and number of children-was associated with a substantially elevated prevalence of HIV, only 1 in 6 couples had such a profile. Two-thirds of tested couples had an intermediate risk score of 1 or 2, which was associated with a 43% to 47% prevalence of HIV in 1 or both partners (range: 0 to 4). Conversely, although couples with none of the analyzed risk factors were at lower risk than the others, they could not accurately be classified as “low risk,” because 29% had at least 1 HIV-positive partner. Risk profiles may be more useful in prioritizing couples for joint testing in a region with lower HIV prevalence, where demographic risk groups are more distinctive.
The generalizability of our findings may be limited by selection bias. Although we know how many couples came for CVCT, CWs did not record how many couples they invited; thus we could not calculate acceptance rates or compare attendees to nonattendees. However, the selection bias is unlikely to be substantial because couples seeking testing were demographically similar to urban couples in the Zambian Demographic and Health Survey19,37 and the prevalence of HIV parallels that in previous cluster sampling surveys in Lusaka.38
Though ZEHRP's initial objective was to gather clinical trial information, our research has clear implications for public health policy and interventions. In urban sub-Saharan Africa, where the epidemic is generalized and the prevalence among antenatal clinic clients ranges from 10% to 32%, all sexually active adults are at risk.38 “Abstinence” and “Be faithful” are primary components of the ABC HIV prevention model.39 Although A and B may be appropriate for single men and women and are thought to have contributed substantially to the decline in HIV prevalence in Uganda,40 abstinence is not feasible in married couples and fidelity has no protective effect in discordant couples. In high-prevalence areas, any promotional message that implies that being faithful to one's spouse is protective is misleading unless both partners are aware that they have the same serostatus and remain monogamous. Until protective biomedical interventions such as noncondom barrier methods, preventive vaccines, or therapeutic strategies are found, “Condoms” will remain the cornerstone of HIV prevention in couples.
CVCT programs and risk-reduction strategies are feasible in a high-prevalence area and present opportunities to reduce HIV transmission.3,7,12,20 The benefits of CVCT have been extensively demonstrated in the peer-reviewed literature cited above, which reflects the experience of several research groups working with couples in Africa. Although the first publications about discordant couples date from the early 1990s, until recently translation of these research findings into public health practice has been slow. Research projects are limited by the mandate of their specific aims; when recruitment has ended and comparable services are not available elsewhere, the repercussions can be damaging.41 Fortunately, in 2006, the President's Emergency Plan for AIDS Relief in Africa added couples' counseling to the priorities for VCT.42 This indicates the growing recognition of the need for a focus on couples, particularly in high-prevalence regions where heterosexual transmission predominates. Although the effectiveness of CVCT in reducing new infections is clear, cost-effectiveness studies comparing promotional strategies and assessing the relative costs of providing CVCT in standalone versus integrated service venues are currently lacking. Politicians, development agencies, and community leaders must be studied to determine the factors that will elicit their tangible support for CVCT; community leaders in the faith-based and nongovernmental sectors must be recruited to endorse couples' testing, and their effectiveness in that role must be quantified; and lastly, HIV service providers in the VCT, PMTCT, and HAART sectors must come together to research best practices in adding the “C” to VCT.
We thank the study participants, staff, interns, and Project Management Group members of the Zambia-Emory HIV Research Project in Lusaka, Zambia.
Dr. Chomba is the senior co-investigator of the Zambia Emory HIV Research Project. Susan Allen is the principal investigator for the Rwanda Zambia HIV Research Group. Dr. Kanweka wrote the first draft of the manuscript. Amanda Tichacek participated in all aspects of the study, from study design to creation of instruments and data analysis, and reviewed the manuscript. Garret Cox and Erin Shutes performed data analysis procedures. Isaac Zulu, Moses Sinkala, and Nzali Kancheya oversaw all recruitment efforts in the field and took part in the paper revision process. Dr. Stephenson participated in manuscript editing and critical revision. Dr. Haworth assisted with manuscript preparation and revision. The Rwanda Zambia HIV Research Group encompasses all US and Zambian staff members who assisted in the research for this project and provision of CVCT services.
1. UNAIDS, WHO. AIDS Epidemic Update: December 2006
. Geneva: The Joint United Nations Programme on HIV/AIDS and the World Health Organization; 2006.
2. Allen S, Van de Perre P, Serufilira A, et al. Human immunodeficiency virus and malaria in a representative sample of childbearing women in Kigali, Rwanda. J Infect Dis
3. Allen S, Tice J, Van de Perre P, et al. Effect of serotesting with counselling on condom use and seroconversion among HIV discordant couples in Africa. BMJ
4. Jenkins RA, Chinaworapong S, Morgan PA, et al. Motivation, recruitment, and screening of volunteers for a phase I/II HIV preventive vaccine trial in Thailand. J Acquir Immune Defic Syndr Hum Retrovirol
5. Painter TM. Voluntary counseling and testing for couples: a high-leverage intervention for HIV/AIDS prevention in sub-Saharan Africa. Soc Sci Med
6. Piot P, Bartos M, Ghys PD, et al. The global impact of HIV/AIDS. Nature
7. Allen S, Serufilira A, Bogaerts J, et al. Confidential HIV testing and condom promotion in Africa. Impact on HIV and gonorrhea rates. JAMA
8. Bentley ME, Spratt K, Shepherd ME, et al. HIV testing and counseling among men attending sexually transmitted disease clinics in Pune, India: changes in condom use and sexual behavior over time. AIDS
9. Sweat M, Gregorich S, Sangiwa G, et al. Cost-effectiveness of voluntary HIV-1 counselling and testing in reducing sexual transmission of HIV-1 in Kenya and Tanzania. Lancet
10. The Voluntary HIV-1 Counseling and Testing Efficacy Study Group. Efficacy of voluntary HIV-1 counselling and testing in individuals and couples in Kenya, Tanzania, and Trinidad: a randomised trial. Lancet
11. Skurnick JH, Kennedy CA, Perez G, et al. Behavioral and demographic risk factors for transmission of human immunodeficiency virus type 1 in heterosexual couples: report from the Heterosexual HIV Transmission Study. Clin Infect Dis
12. Fideli US, Allen SA, Musonda R, et al. Virologic and immunologic determinants of heterosexual transmission of human immunodeficiency virus type 1 in Africa. AIDS Res Hum Retroviruses
13. Kamenga M, Ryder RW, Jingu M, 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
14. Padian NS, O'Brien TR, Chang Y, et al. Prevention of heterosexual transmission of human immunodeficiency virus through couple counseling. J Acquir Immune Defic Syndr
15. Gray RH, Wawer MJ, Brookmeyer R, et al. Probability of HIV-1 transmission per coital act in monogamous, heterosexual, HIV-1-discordant couples in Rakai, Uganda. Lancet
16. Allen SA, Karita E, N'gandu N, et al. The evolution of voluntary testing and counseling as an HIV prevention strategy. In: Gibney L, DiClemente RJ, Vermund SH, eds. Preventing HIV in Developing Countries: Biomedical and Behavioral Approaches
. New York: Plenum Press; 1999:87-108.
17. Weinhardt LS, Carey MP, Johnson BT, et al. Effects of HIV counseling and testing on sexual risk behavior: a meta-analytic review of published research, 1985-1997. Am J Public Health
18. McKenna SL, Muyinda GK, Roth D, et al. Rapid HIV testing and counseling for voluntary testing centers in Africa. AIDS
. 1997;11(Suppl 1):S103-S110.
19. Central Statistical Office (Zambia), Central Board of Health (Zambia) and ORC Macro. 2003. Zambia Demographic and Health Survey 2001-2002
. Calverton, MD: Central Statistical Office, Central Board of Health, and ORC Macro.
20. Allen S, Meinzen-Derr J, Kautzman M, et al. Sexual behavior of HIV discordant couples after HIV counseling and testing. AIDS
21. Higgins DL, Galavotti C, O'Reilly KR, et al. Evidence for the effects of HIV antibody counseling and testing on risk behaviors. JAMA
22. Colebunders R, Ndumbe P. Priorities for HIV testing in developing countries? Lancet
24. 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
25. Were WA, Mermin JH, Wamai N, 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
26. Antelman G, Smith Fawzi MC, Kaaya S, et al. Predictors of HIV-1 serostatus disclosure: a prospective study among HIV-infected pregnant women in Dar es Salaam, Tanzania. AIDS
27. King R, Allen S, Serufilira A, et al. Voluntary confidential HIV testing for couples in Kigali, Rwanda. AIDS
28. Glynn JR, Carael M, Buve A, et al. HIV risk in relation to marriage in areas with high prevalence of HIV infection. J Acquir Immune Defic Syndr
29. Carael M, Van de Perre PH, Lepage PH, et al. Human immunodeficiency virus transmission among heterosexual couples in Central Africa. AIDS
30. Becker S. Couples and reproductive health: a review of couple studies. Stud Fam Plann
31. Hugonnet S, Mosha F, Todd J, 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
32. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med
33. Serwadda D, Gray RH, Wawer MJ, et al. The social dynamics of HIV transmission as reflected through discordant couples in rural Uganda. AIDS
34. Wawer MJ, Gray RH, Sewankambo NK, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis
35. Bakari JP, McKenna S, Myrick A, et al. Rapid voluntary testing and counseling for HIV. Acceptability and feasibility in Zambian antenatal care clinics. Ann N Y Acad Sci
36. Glynn JR, Carael M, Auvert B, et al. Why do young women have a much higher prevalence of HIV than young men? A study in Kisumu, Kenya and Ndola, Zambia. AIDS
. 2001;15(Suppl 4):S51-S60.
37. Stephenson R, Barker J, Cramer R, et al. The demographic profile of sero-discordant couples enrolled in clinical research in Rwanda and Zambia. AIDS Care
. 2007. In press.
38. Fylkesnes K, Ndhlovu Z, Kasumba K, et al. Studying dynamics of the HIV epidemic: population-based data compared with sentinel surveillance in Zambia. AIDS
39. Shelton JD, Halperin DT, Nantulya V, et al. Partner reduction is crucial for balanced “ABC” approach to HIV prevention. BMJ
40. Murphy EM, Greene ME, Mihailovic A, et al. Was the “ABC” approach (abstinence, being faithful, using condoms) responsible for Uganda's decline in HIV? PLoS Med
41. Stephenson R, Shutes E, Allen S, et al. The impact of project closure on HIV incidence and mortality in a cohort of couples in Lusaka, Zambia. AIDS Care
. 2007. In press.
Keywords:© 2008 Lippincott Williams & Wilkins, Inc.
voluntary counseling and testing; discordant couples; HIV; sub-Saharan Africa; heterosexual HIV prevention