Partner Reduction

Multiple sexual partnerships have long been a concern in HIV prevention programming, which has focused on partner reduction. Multiple partnerships bring increased risk of HIV acquisition: A meta-analysis of 68 epidemiological studies from 1986 to 2006 with 17,000 HIV-positive people and 73,000 HIV-negative people found that women who reported three or more sex partners had three times as much likelihood of HIV acquisition versus women with up to two partners (Chen et al., 2007 cited in Vergidis et al., 2009). An analysis of DHS data in Zimbabwe found that 64% of men reported more than three lifetime sexual partners compared to 13% of women (Sambisa et al., 2010).

When designing prevention interventions it is important to understand women's various partnership patterns. A woman may be married with only one sexual partner. Or she may be married with multiple sexual partners. A young woman might be sexually active with uninfected boys her own age. Or she may be in a much riskier cross-generational sexual relationship, with an older male partner upon whom she relies for school fees. [See Prevention and Services for Adolescents and Young People] A woman may also have multiple partners to enable her to survive financially. Women may work as sex workers, an occupation that requires multiple sexual partners. [See Prevention for Key Affected Populations] Men may also have multiple partnerships, which may place women at risk for HIV acquisition. Concurrent, or overlapping, sexual partners presents additional risks for HIV acquisition.

Concurrency is an Important Aspect of Multiple Partnerships

More recently, the concept of concurrent sexual partnership has dominated discussions of multiple partnerships. UNAIDS defines concurrency as "overlapping sexual partnerships in which sexual intercourse with one partner occurs between two acts of intercourse with another partner" (UNAIDS, 2010d). While this definition helps distinguish between multiple partnerships that are serial and those that occur at the same time, "this limited definition is unlikely to capture the rich and variable characteristics of sexual partnerships [and] detailed understanding of the sexual behaviors in a given context [that] is necessary for the optimal design of prevention interventions" (Powers et al., 2011a: 665). A recent study analyzing data from Thailand and Uganda found that prevalence of concurrent partnerships is aligned with HIV prevalence (Morris et al., 2010). However, a more recent study found that increases in lifetime numbers of partners for men, not concurrency, raised the individual risk of seroconversion in women from the same area (Tanser et al., 2011 cited in Padian and Manian, 2011). In addition, other studies found that regions with large numbers of partners and low concurrency were associated with high incidence of infection of women (Tanser et al., 2011 cited in Padian and Manian, 2011).

New methodologies are being developed to measure changes in concurrency and HIV incidence (Maher et al., 2011). However, countries will need to improve their surveillance in order to obtain information on concurrency. For example, prior to 2007 in Jamaica, surveillance forms collected information on the number of lifetime partners rather than partners during a 12-month period (Duncan et al., 2010b).

Concurrent Partnerships During the Acute Infection Stage Can Pose a Higher Risk for HIV Acquisition for the Seronegative Partner

Viral load and infectivity are higher in the early, acute stage of infection, so recently infected individuals with concurrent partners are more likely to transmit HIV to others than recently infected individuals that have one or no partners (Pilcher et al., 2004; Morris and Kretzchmar, 1997 cited in Carter et al., 2007). Acute infection can result in an estimated 26-fold increased risk of transmission during the first three months following HIV acquisition (Hollingsworth et al., 2008). However, detection of acute infection requires specific testing for p24 antigen and is not available in many resource-limited settings (Cohen and Gay, 2010). [See also HIV Testing and Counseling for Women and Treatment] "Concurrent partnerships increase the overall probability that uninfected partners will have sexual intercourse and be exposed to a partner during acute infection" (Mah and Halperin, 2010: 14). Other modeling has also found that acute infection amplifies the importance of concurrent partnerships and suggests "intervention programs that are effective in reducing concurrency may play a crucial role in stemming the incidence of new HIV infections" (Eaton et al., 2011: 691). Others have also noted the importance of acute HIV infection and concurrency to explain high rates of HIV (Goodreau, 2011).

Multiple Partnerships Are Common and Place both Women and Men At Risk of HIV Acquisition

While both men and women may have multiple sexual partners, in some environments "men's extramarital sexual activities are not only socially condoned but are a defining element of masculine identity..." (Stephenson, 2010: 179; Hirsch et al., 2009). [See Transforming Gender Norms] However, studies have found a very strong relationship between people having had more than one sexual partner and living with HIV but found no association between concurrence in men and HIV incidence in women (Tanser et al., 2010 cited in WHO et al., 2011b) or between concurrency and HIV prevalence in men (Maher et al., 2011 cited in WHO et al., 2011b).

"Two kinds of women run the greatest risk: the one who stays home and trusts her husband and the one who turns tricks." Brazilian woman (Hebling and Guimaraes, 2004: 1213)A recent analysis of DHS data in 14 sub-Saharan countries found that women (47%) were as likely to be the HIV-positive partner in a serodiscordant couple as men (Eyawo et al., 2010). The number of those having multiple partnerships has been decreasing: analysis shows that the percentage of young men with multiple partners in the 12 months before the most recent survey decreased significantly in 11 countries, including four countries with national adult HIV prevalence exceeding 10% in 2009. Among women, most recent surveys showed a decrease in multiple partners in six countries (WHO et al., 2011b). However, in Rwanda and Zimbabwe, the number of young women with multiple partners has increased (WHO et al., 2011b).

Married Women Are Still At Risk for HIV Acquisition

Marriage is often portrayed to women and girls as a haven from the risk of HIV infection. In fact, the risk of HIV transmission between sexual partners is nonexistent when both partners are uninfected at the time of marriage and subsequently engage in sexual activity exclusively with each other. However, these conditions are often not met. In some countries, married women are at high risk of acquiring HIV (UNAIDS, 2006: 22; Hirsch et al., 2007; Hageman et al., 2009; Ugonnet et al., 2002 cited in Matovu et al., 2007), particularly in generalized epidemics. An estimated 55% to 92% of new heterosexually acquired HIV infections among sexually active adults in urban Zambia and Rwanda occur within serodiscordant martial/cohabitating relationships (Dunkle et al., 2008). In some countries, HIV prevention messaging has focused on "love faithfully" and "zero grazing" which may have inadvertently increased risk for married women who were seronegative with HIV-positive husbands (Grabbe and Bunnell, 2010). Unless attention is given to gender norms, married women may not understand that they are at risk of HIV acqustion. A study of 50 low-income Chilean women found that "women who are vulnerable to HIV do not perceive themselves at risk. They believe that HIV is something that happens to homosexually active men or to [sex workers], not something that happens to women in a stable relationship" (Cianelli et al., 2008: 298). Couples interventions to reduce transmission in serodiscordant relationships could have a large impact on the epidemic (Dunkle et al., 2008). However, transmission may result from partners outside marriage (Spino et al., 2010). A study using a national survey in China found that men who paid for sex were ten times more likely to have a STI, yet less than 4% used a condom consistently with their spouse (Huang et al., 2011).

Additional risks may be posed by polygyny (i.e. legal or customary marriage with multiple wives), which may place women at risk of HIV acquisition with low rates of condom use and unequal power relations (Bove and Valeggia, 2009). A study of 1,137 women in a village in Kenya found that women in polygamous marriages were more likely to be HIV-positive than those in monogamous marriages (Negin et al., 2009). Polygynous women in focus group discussions in Nigeria agreed that any kind of sexual negotiation within marriage was difficult (Saddiq et al., 2010). Other studies, however, have not found an elevated risk among polygynous marriages in Western Africa (Reniers et al., 2010; Reniers and Watkins, 2010). Yet, a study based on the 2005 to 2006 Zimbabwe Demographic and Health Survey with 4,023 married women found that currently married women who were in polygynous marriages were at higher risk of spousal violence. Currently married women who experienced physical violence only, or both physical and sexual violence, were significantly more likely to be HIV-positive than those who had not experienced any physical or sexual violence (Nyamayemombe et al., 2010).

It is critical that partners know their serostatus and practice safe sex. Both married and unmarried women need basic knowledge of HIV and how to prevent transmission. However, married women are often not reached by prevention messages because married women "were not considered part of the so-called risk groups. Prevention efforts have been focused on pregnant women, sex workers, and people who inject drugs. Therefore, the majority of women received a message of false security that women who are married and monogamous have no risk for acquiring HIV" (Ross Quiroga, 2006: 1-2). Despite the fact that HIV transmission occurs within stable partnerships or marriage, a review of the literature on couples' HIV prevention found that "couples-focused approaches to HIV prevention are still in an early phase of development" (Burton et al., 2010: 9). Many couple-focused approaches are ready to be scaled up. [See Male and Female Condom Use, HIV Testing and Counseling for Women, and Safe Motherhood and Prevention of Vertical Transmission ]

Married adolescent girls are particularly vulnerable and are often more at risk of HIV infection than unmarried sexually active girls. For example, a study analyzing Kenyan and Zambian data from 1997 and 1998 found that married adolescent girls living in urban areas had higher incidence of HIV than unmarried sexually active girls in the same age group. "Although married girls are less likely than single girls to have multiple partners, this protective behavior may be outweighed by their greater exposure via unprotected sex with partners who have higher rates of infection" (Clark, 2004: 149). Young women who engage in transactional sex have an incentive to change partners and to end relationships to find more lucrative male partners (Wamoyi et al., 2010). Women may also oppose condom use to keep partners in a competitive world of multiple concurrencies. For some women, concurrent sexual partners may be culturally acceptable as long as they are conducted covertly (Jewkes and Morrell, 2010).

Reducing Concurrent Partnerships Can Reduce HIV Transmission

"There are... few demonstrated replicable approaches to reducing multiple sexual partnerships on a large scale" (Potts et al., 2008: 750). There is currently programmatic focus on partner reduction, yet this review (Ross, 2010) identified few evaluations of interventions of partner reduction, particularly among adult men. Interventions targeting concurrency "are still largely in their infancy and their introduction should be linked to careful evaluation" (Ross, 2010: S12). A 2011 review found that "most interventions to raise awareness of the risks of concurrency are less than two years old: few evaluations and no randomized controlled trials of these programmes have been conducted" (Epstein and Morris, 2011: 1). Some agree that more specific knowledge concerning the role of concurrency is needed (Shelton, 2009), but have found that the evidence that concurrency is a key issue for HIV acquisition and transmission is compelling (Mah and Shelton, 2011). Others have also argued for more research on this topic (Lurie and Rosenthal, 2010) and still others have critiqued modeling of concurrency as a key driver of the epidemic or have conducted surveys, for example in Uganda, that did not find that concurrency was correlated with increased risk of HIV acquisition (Sawers and Stillwaggon, 2010; Sawers et al., 2011; Maher et al., 2011).

But a rigorous study with participants in Malawi who had physical exams, HIV tests and responded to questionnaires with detailed questions about sexual partners during the prior two months found that among those reporting multiple recent partners, both long-term concurrency and narrowly spaced consecutive partnerships could present substantial risk for efficient transmission of HIV (Powers et al., 2011a). Others have found vast variation from 1% in Ethiopia to 28% of men reporting two or more sex partners in the 12 months prior to being interviewed, with no correlation between the prevalence of multiple partnerships and the severity of the HIV pandemic (Bingenheimer, 2010). Still others have found that it is multiple partnerships by men, rather than concurrent partnerships, that increase the risk of HIV acquisition for women, at least in one area of South Africa (Tanser et al., 2011). Ultimately, "the needs of the married and cohabitating population have been neglected... despite the fact that more than half of HIV infections in the severe epidemics of Southern and East Africa are occurring in this group" (Delvaux and Nostlinger, 2007: 56).

Interventions to reduce concurrent partnerships that are gender transformative are urgently needed. "Although there is no disagreement that multiple concurrent partnerships contribute to risk for HIV transmission, and thus should be subject to HIV prevention programming responses," the fact that concurrent partnerships are the norm in many places "makes such partnerships difficult to address directly" (Padian et al., 2011b: 274). Multiple partnerships is closely tied to gender norms of masculinity, where men are required to have multiple sexual partnerships simultaneously, be unfaithful to their regular sexual partner and buy sex as proof of their masculinity (Peacock et al., 2008). Many women are unaware that their husbands or sexual partners may have other sexual partners. Surveys in Africa have found that women are less likely to have concurrent partners than men (Sawers and Stillwaggon, 2010).

However, "if education focuses merely on abstinence and fidelity as methods of prevention, those who become infected and those already living with HIV may feel that their needs are being overlooked" (Ansari and Gaestel, 2010: 634). Programs need to work with communities to address gender norms that put women at risk through expectations of fidelity, while failing to address gender norms that are expect multiple partnerships among men as a sign of masculinity. [See Prevention and Services for Adolescents and Young People and Transforming Gender Norms] Increasing couple communication is a promising strategy to begin addressing these risks and raise awareness that married women are indeed at risk for acquiring HIV. A few areas regarding partner reduction still have major gaps that need to be filled, including interventions that address the risks of polygamous marriage (Sandoy et al., 2008), and the role of homophobia in leading men who have sex with men to feel they must hide their sexuality through concurrent partnerships with women.