Secondary Logo

Journal Logo

Research Letters

The contribution of HIV-discordant relationships to new HIV infections in Rakai, Uganda

Gray, Rona; Ssempiija, Victorb; Shelton, Jamesc; Serwadda, Davidb,d; Nalugoda, Fredb; Kagaayi, Josephb; Kigozi, Godfreyb; Wawer, Maria Ja

Author Information
doi: 10.1097/QAD.0b013e3283448790
  • Free



Understanding the transmission dynamics of generalized HIV epidemics is a key to prevention strategies. One model estimated that the majority of infections occurred within long-term stable HIV-serodiscordant couples, suggesting that couples voluntary counseling and testing (CVCT) could prevent transmission, mitigating the epidemic [1,2]. An alternative approach estimated that upwards of two-thirds of infections occurred via multiple partnerships, predominantly outside of stable relationships, and thus would not be amenable to a CVCT intervention [3]. We examined this issue using data on incident infections from Rakai, Uganda before and after the availability of antiretroviral therapy (ART).


The Rakai Community Cohort Study conducts census and serosurveys among consenting residents aged 15–49 years in 50 rural communities of rural Rakai District [4]. For respondents who are married (including stable consensual relationships), we request information on their spouse or partner which allows linkages of couple data to partners residing in the study areas, and approximately 68% of married or consensual individuals (hereafter referred to as ‘married’) have an identifiable linked partner.

We assessed incident HIV infections among initially HIV-negative cohort participants observed during two intersurvey intervals; one intersurvey interval prior to the availability of ART and one after ART services were available. The pre-ART intersurvey interval used data collected from two surveys; the first conducted from July 2002 to August 2003 with follow-up between September 2003 and November 2004. For the post-ART interval, we used data from a survey conducted between January 2005 and June 2006 with a follow-up survey between July 2006 and January 2008. HIV status was determined by two separate enzyme-linked immunosorbent assays and confirmed by HIV-1 western blot, as previously described.

We divided the population into HIV-negative participants who were not currently married at the beginning of an intersurvey interval (i.e. never married, or previously married but currently separated, divorced or widowed) and persons who were currently married at the beginning of an intersurvey interval. The currently married were further disaggregated into couples in which both partners were concordantly HIV-negative at the start of the interval, couples who were HIV-discordant, and married persons whose partner's HIV status was unknown (e.g. the partner was absent or refused participation).

We then tabulated the proportion of all new HIV infections observed during the intersurvey intervals by the marital and HIV status of the participants at the initial survey. Our main focus was on the proportion of total incident infections that occurred among couples whose primary relationship was HIV-discordant. In Rakai, most people eventually marry, and divorce is relatively infrequent [5]. More than 90% of participants agree to HIV testing [4].


The distribution of the populations at the beginning of the intersurvey intervals and the number of incident HIV infections during follow-up are shown in Table 1. The proportions of all new HIV infections that occurred within identifiable HIV-discordant couples were 18.3% during the pre-ART period and 13.7% during the post-ART period (P = 0.8). Most new infections occurred among persons who were not currently married (29.3% pre-ART and 42.7% post-ART, P = 0.08), or among married individuals with a partner of unknown HIV status (29.3% pre-ART and 17.6% post-ART, P = 0.3). Additionally, 23.2% of infections pre-ART and 26.0% post-ART (P = 0.5) occurred among initially concordant HIV-negative couples in whom the new index infection must have been introduced from an extramarital relationship.

Table 1:
The distribution of the population by marital and HIV status at the beginning of two intersurvey intervals prior to and after the availability of ART, and the distribution of incident HIV infections observed during the intersurvey intervals.


Our data indicate that only a minority of total HIV infections occur to a person in an identifiable stable HIV-discordant couple relationship in this generalized HIV epidemic setting. In much of Southern Africa where age of first marriage is delayed, an even smaller proportion of people are in stable partnerships than in Eastern Africa [6,7]. Moreover, viral sequencing showed that 29% of new infections in the previously negative partner in a discordant relationship came from a partner outside the primary relationship [8]. Thus, in addition to the question of the efficacy of a CVCT intervention with couples, universal VCT targeted on HIV-discordant couples is unlikely to have a major impact on the epidemic [1,2]. Although it might be argued that our observed infection rates were lowered because of counseling and disclosure among the discordant couples, disclosure and acceptance of couples counseling has been quite low in this cohort, despite efforts to promote CVCT. Part of the explanation for this minority contribution of new infections among identifiable HIV-discordant couples is that infectivity is quite low during the chronic phase of infection [9,10]. The data post-ART reflect an even lower proportion of transmissions from within identifiable discordant couples, perhaps suggesting some collateral benefit of ART in reducing infectivity [11]. Thus, although there is definite merit to counseling and testing, including counseling of couples, it is likely to have limited impact, particularly if the focus remains on those testing positive and on discordant couples. Rather, prevention efforts need also to embrace specific interventions, such as male circumcision, condoms and behavior change to limit outside partners, aimed at the broader population in which most infections are occurring.


1. 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.
2. Bunnell R, Cherutich P. Universal HIV testing and counselling in Africa. Lancet 2008; 371:2148–2150.
3. Shelton JD. A tale of two-component generalised HIV epidemics. Lancet 2010; 375:964–966.
4. Wawer MJ, Gray RH, Sewankambo NK, Serwadda D, Paxton L, Berkley S, et al. A randomized, community trial of intensive sexually transmitted disease control for AIDS prevention, Rakai, Uganda. AIDS 1998; 12:1211–1225.
5. Porter L, Hao L, Bishai D, Serwadda D, Wawer M, Lutalo T, et al. HIV status and union dissolution in sub-Saharan Africa: the case of Rakai, Uganda. Demography 2004; 41:465–482.
6. Bongaarts J. Late marriage and the HIV epidemic in sub-Saharan Africa. Populat Stud 2007; 61:73–83.
7. Marston M, Slaymaker E, Cremin I, Floyd S, McGrath N, Kasamba I, et al. Trends in marriage and time spent single in sub-Saharan Africa: a comparative analysis of six population-based cohort studies and nine Demographic and Health Surveys. Sex Transm Infect 2009; 85(Suppl 1):i64–71.
8. Celum C, Wald A, Hughes J, Sanchez J, Reid S, Delany-Moretlwe S, et al. Effect of aciclovir on HIV-1 acquisition in herpes simplex virus 2 seropositive women and men who have sex with men: a randomised, double-blind, placebo-controlled trial. Lancet 2008; 371:2109–2119.
9. Wawer MJ, Gray RH, Sewenkambo NK, Serwadda D, Li X, Laeyendecker O, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis 2005; 191:1403–1409.
10. Boily MC, Gaggaley RF, Wang L, Masse B, White RG, Hayes RJ, et al. Heterosexual risk of HIV-1 infection per sexual act: systematic review and meta-analysis of observational studies. Lancet Infect Dis 2009; 9:118–129.
11. Donnell D, Baeten JM, Kiarie J, Thomas KK, Stevens W, Cohen CR, et al. Heterosexual HIV-1 transmission after initiation of antiretroviral therapy: a prospective cohort analysis. Lancet 2010; 375:2092–2098.
© 2011 Lippincott Williams & Wilkins, Inc.