HIV Prevention Knowledge Base
Behavioral Interventions: Multiple and Concurrent Sexual Partnerships
Debate: Is Sexual Partner Concurrency Driving the Spread of HIV in Africa?
Mah and Halperin’s 2010 article in AIDS and Behavior, “Concurrent Sexual Partnerships and the HIV Epidemics in Africa: The Evidence to Move Forward,” sparked passionate debate around the impact of sexual concurrency on the HIV epidemic in Africa. Critics assert that the lack of consistent data on multiple and concurrent partnerships (MCP) calls into question whether MCP is a true driver of HIV in Africa. Proponents of the concurrency thesis acknowledge the inconsistencies in definition and measurement methods, but maintain the relevance of MCP to the spread of HIV in the region and argue that programming to address MCP remains an important element of a prevention response. Here, AIDSTAR-One summarizes a series of articles and commentaries responding to Mah and Halperin.
Concurrent Sexual Partnerships and the HIV Epidemics in Africa: The Evidence to Move Forward
Mah, T. & Halperin, D.T. AIDS and Behavior (2010), Vol. 14 No. 1, pp. 11-16.
Concurrent sexual partnerships are thought to contribute to the generalized HIV epidemics in sub-Saharan Africa. Such relationships may be particularly risky in part because they are likely to expose individuals to a partner with HIV during the highly infectious month immediately following infection. Concurrency is said to be more common in Southern Africa than elsewhere, fueled in part by migrant work that separates spouses. However, research about rates of concurrency can produce conflicting results depending on the survey method used. The authors discuss the pros and cons of the calendar method of questioning versus direct questioning.
Concurrent Partnerships as a Driver of the HIV Epidemic in Sub-Saharan Africa? The Evidence is Limited
Lurie, M.N. & Rosenthal, S. AIDS and Behavior (2010), Vol. 14 No. 1, pp. 17-24.
The authors assert that current measures and definitions for concurrency are not consistent, casting doubt on the assumption that concurrency is a key driver of the African HIV epidemic. Citing a 2001 study, they comment that concurrency should be “higher in countries with high HIV prevalence compared to those in low prevalence countries, but they were not,” and conclude that Mah and Halperin’s claim that concurrency can play a critical role in HIV transmission is not the same as empirical evidence showing that it does. They suggest that more targeted research with consistent methodology is needed before public health specialists can draw definite conclusions about the relationship between concurrency and the African epidemics.
The Evidence for the Role of Concurrent Partnerships in Africa’s HIV Epidemics: A Response to Lurie and Rosenthal
Mah, T. & Halperin, D.T., AIDS and Behavior (2010), Vol. 14 No. 1, pp. 25 -28.
The authors challenge the assertion by Lurie and Rosenthal that there is insufficient data to conclude that concurrency is any more common in Southern Africa than elsewhere. Mah and Halperin say that survey data they cited are from World Health Organization surveys that support the idea that African populations studied have higher rates of concurrency. They cite surveys showing that in Lesotho, where HIV prevalence is the third highest in the world, 55 percent of men and 39 percent of women report concurrency. They compare this to the 3 percent and 0.2 percent rates of concurrency among women in Thailand.
The Mathematics of Concurrent Partnerships and HIV: A Commentary on Lurie and Rosenthal, 2009
Epstein, H. AIDS and Behavior (2010), Vol. 14, No. 1, pp. 29-30.
Epstein responds to Lurie and Rosenthal, who assert there are insufficient empirical data to support the claim that concurrency is driving the HIV epidemic in Southern Africa. The author defends mathematical modeling exercises that support concurrency as a central factor in the spread of HIV, and criticizes one mathematical model that didn’t support the concurrency concept on the grounds that the researchers failed to distinguish between concurrency and serial monogamy.
Barking up the Wrong Evidence Tree. Comment on Lurie & Rosenthal, "Concurrent Partnerships as a Driver of the HIV Epidemic in Sub-Saharan Africa? The Evidence is Limited"
Morris, M. AIDS and Behavior (2010), Vol. 14 No. 1, pp. 31-33.
The author asserts that Lurie and Rosenthal make several mistakes with regard to their assertion that concurrency is not a proven driver of the HIV epidemic. First, she states that measuring concurrency in the index person who acquires HIV is a mistake since “concurrency increases your risk of transmitting infection, not acquiring it.” Second, the use of HIV prevalence is an improper message, according to the author, since prevalence is a cumulative measure over time, while concurrency is generally measured over a limited time period, such as 12 months. Only when both prevalence and concurrency are in equilibrium for some years can the effects of concurrency be measured, says the author. For this reason, HIV incidence should be measured instead of prevalence and it should be measured during time window matched to measures of concurrency.
The Concurrency Hypothesis in Sub-Saharan Africa: Convincing Empirical Evidence is Still Lacking. Response to Mah and Halperin, Epstein, and Morris
Lurie, M.N. & Rosenthal, S. AIDS and Behavior (2010), Vol. 14 No. 1, pp. 34-37.
Lurie and Rosenthal respond to authors who are critical of their contention that concurrency is not a proven driver of the HIV epidemic in Southern Africa. They note that their critics agree that the evidence is “limited” and that further research is needed. They assert that even if their critics can point to correlates between concurrency and high HIV prevalence, that association is not causation. The authors say that one study that assessed the temporal relationship between concurrency and the development of HIV did not find a causal link. The authors object to the assumption that in the face of uncertainty about concurrency that “nothing is lost” by discouraging concurrency since, they say, some forms of concurrency, such as polygyny, appear to be protective.
A Framework of Sexual Partnerships: Risks and Implications for HIV Prevention in Africa
The authors of this commentary state that “many standard HIV prevention strategies have not proven effective,” such as the medically based triad of condom promotion, HIV testing, and treatment of sexually transmitted diseases in the generalized epidemics of East and Southern Africa. They cite the successful reduction of HIV transmission in concentrated epidemics following a campaign promoting condom use among sex workers in Thailand. The authors cite concurrency as a significant contributor to the generalized epidemics of East and Southern Africa and say that campaigns to reduce multiple partners in several African nations have been effective since they have been followed by measurable reductions in multiple partners, which in turn have been followed by reduced HIV prevalence.
Relatively Low HIV Infection Rates in Rural Uganda, But with High Potential for a Rise: A Cohort Study in Kayunga District, Uganda
Studies have explored the determinants of HIV infection in Uganda. Using a community-based cohort of 2,025 volunteers from Kayunga, Uganda, researchers administered a questionnaire about participants’ sexual behavior and other HIV risk factors every six months. Blood samples were collected simultaneously. At baseline, approximately one of every 10 individuals tested positive for HIV-1. Within one year, 13 new cases of HIV were detected. HSV-2 (genital herpes) was strongly associated with HIV-1 infection. The most significant behavioral risk factor associated with new HIV infection was the number of times in the last six months that a participant had sex with partners whom they thought or knew were having sex with someone else.
Cultural Scripts for Multiple and Concurrent Partnerships in Southern Africa: Why HIV Prevention Needs Anthropology
From October 2007 to November 2008, 228 members of nongovernmental organizations in seven Southern African countries were interviewed to understand the role of culture and contextual factors in relation to MCP. Common patterns in cultural scripts regarding sexuality emerged. These scripts (such as “male sexuality is by nature un-restrainable” or “sexual violence sometimes demonstrates caring”) affirm and legitimatize MCP and need to be addressed to better orient messages to the population.
Why Multiple Sexual Partners?
In this commentary, individual, social, economic, and cultural factors underlying multiple concurrent partnerships are discussed. Understanding the causes of sexual concurrency in a given geography and with a given group is an important step toward developing effective HIV prevention programming. “High-quality multilevel (mass media, community, clinical setting, individual) approaches” are called for. One Love, Scrutinize, and O Icheke are cited as exemplary.
Young People's Sexual Partnerships in KwaZulu-Natal, South Africa: Patterns, Contextual Influences, and HIV Risk
The study authors conducted a household survey to examine the sexual behaviors of young people, aged 15-24 years in KwaZulu-Natal, South Africa. The region is one of that nation’s poorest provinces and has the nation’s highest HIV prevalence. The researchers found that one third of men reported MCPs and one quarter of women had partners who were five years older than themselves. Nonparticipation in school or civic organizations correlated with higher-risk partnerships for women. Relationships, on average, lasted more than a year and were defined as “serious” by the participants. Partnerships were both sequential and overlapping. Mobility and the distance between workplace and home greatly affected the kinds of relationships that were formed and sustained.
Age-disparate and Intergenerational Sex in Southern Africa: the Dynamics of Hypervulnerability
Age-disparate (age gap >5 years between partners) or intergenerational (>10 years) sexual partnerships are thought to disproportionately contribute to high HIV rates among young women aged 15-24 years. Most such partnerships are transactional in nature, rooted in cultural beliefs that men demonstrate affection by providing for women, and that women’s bodies are assets for transactions. Pairing of older men and younger women is further fueled by men’s preference for young, presumably disease-free, partners. Although these partnerships are often mutually advantageous rather than victimizing, women are usually not empowered to negotiate condom use. Along with partner reduction messages, interventions must address education and financial independence in order to empower women to protect themselves. Programs should also foster male norms that discourage exploitative relationships.
Reassessing HIV Prevention
This commentary challenges several assumptions about HIV epidemiology, including the idea that poverty is an important factor in the spread of HIV. The authors cite data from recent Demographic and Health Surveys (DHS) to suggest that within Africa, high HIV prevalence is not associated with high levels of poverty or conflict, but instead correlates with high rates of MCP, and low levels of male circumcision. The authors also cite evidence that microbicides, vaccines, and HIV testing and treatment of sexually transmitted diseases have little effect on HIV transmission. They conclude that male circumcision (which can reduce a man’s risk of contracting HIV by more than one half) and reduction in sex partners are the most effective interventions for generalized epidemics.
Small House, Hure, Sugar Daddies, and Garden Boys: A Qualitative Study of Heterosexual Concurrent Partnerships Among Men and Women in Zimbabwe
2006 study about the nature of sexual partnerships in Zimbabwe found that nearly one third of men and over one quarter of women had more than one regular sexual partner. These concurrent partnerships ranged from short-term casual relationships to longer-term emotional relationships. Short-term relationships included intergenerational sex and transactional sex in which money or goods were exchanged for sex. Sex workers engaged in short- and long-term relationships as some sex workers had long-term boyfriends or regular clients. Long-term relationships included those with spouses, girlfriends, and “small house” relationships in which a man supports his partner and possible children but is not formally married.
Why Is HIV Prevalence So Severe in Southern Africa? The Role of Multiple Concurrent Partnerships and Lack of Male Circumcision: Implications for AIDS Prevention
The combination of high rates of concurrent sexual partnerships with low rates of male circumcision seems to distinguish Southern Africa from other regions affected by HIV, and to fuel the world’s largest generalized HIV epidemics. Although African men and women do not have more sex partners than people do elsewhere, their partnerships are more likely to overlap for months or years, creating stable overlapping networks of sexual relationships through which HIV can spread rapidly. In contrast to serial monogamy, HIV can spread more rapidly through concurrent partnerships, in part due to the greater likelihood of contact during the highly infectious month immediately following infection. Consistent condom use can be achieved in casual partnerships, but consistent use is much more difficult to attain in longer-term “trusting” relationships, due to low risk perception.
Sexual Network Structure and the Spread of HIV in Africa: Evidence from Likoma Island, Malawi
In generalized HIV epidemics, such as those of Southern and East Africa, infections occurring among low-risk individuals account for the majority of new HIV cases. In a study in Likoma Island, Malawi, the prevalence of HIV infection was higher among women than men. Researchers found that half of the island’s sexually active, young adult population was linked in a giant network, rendering them highly effective in spreading HIV among lower-risk groups. The structure of the networks observed in Likoma appears compatible with a broad diffusion of HIV among lower-risk groups.
Measuring Sex Partner Concurrency: It's What's Missing That Counts
Measures for concurrency are not consistent, leading to conflicting study results. In order to assess the different results obtained by two common survey methods, the researchers conducted a study in which both methods (the calendar method and direct questioning) were used to query 680 young adults in three U.S. cities. Although the total rates of concurrency (over one half of participants) were similar using both methods, nearly one third of individuals reporting concurrency in one measure did not do so in the other. The researchers conclude that despite the greater detail provided by the calendar measure, direct questioning might better identify those at greatest risk for HIV infection.
Ten Myths and One Truth about Generalised HIV Epidemics
The author of this commentary states that common misperceptions impede HIV prevention efforts. For example, he says, commercial sex workers are not the problem since formal sex work is uncommon in generalized epidemics. Nor are men the problem since there is a high proportion of discordant couples in which women, not men, are HIV positive. The author comments on other perceived myths related to HIV, such as those surrounding condoms, HIV testing, and youth. He concludes that concurrency is central to generalized epidemics and that behavioral interventions are effective.
Concurrent Sexual Partnerships amongst Young Adults in South Africa: Challenges for HIV Prevention Communication
This 51-page report examines concurrent sexual partnerships (defined in this survey as a person who has “two or more sexual partners in the past month”) among young people aged 20-30 years old in South Africa. The researchers found that for these young people, the concepts of sex and love are often separated; sex with love is reserved for a ‘main’ partner, while sex without love is common with ‘other’ or concurrent partners. This duality is widely accepted as normative and results in a definition of faithfulness whereby keeping infidelity secret is a sufficient criterion for considering oneself to be faithful. Among people with concurrent partners, condom use declined rapidly with a main partner, and was inconsistent with other partners. The authors conclude that in the South African context, high overall HIV prevalence in conjunction with concentrated sexual networks suggest the need for programs that focus on reducing concurrency. Condom promotion remains an important cornerstone to HIV prevention, and programs need to increase their efforts to promote correct and consistent condom use. HIV testing is a useful complementary strategy for people in established relationships, or for those considering establishing them.
Brief but Efficient: Acute HIV Infection and the Sexual Transmission of HIV
Empirical biological data suggests that individuals who are acutely infected with HIV (during the first several months) are more likely to infect others than they are during a later phase of their disease, when their viral load is likely to be lower. Using blood and semen samples from 30 patients with acute and long-term HIV-1 infection, the researchers modeled the effect of changes in viral concentration in semen on the probability of transmission per coital act and found that the probability of heterosexual transmission of HIV increases by 8-10-fold during the acute phase of infection compared to infection during the chronic phase of HIV.
Concurrent Partnerships and the Spread of HIV
Mathematical models comparing the spread of HIV in two populations—one in which serial monogamy was the norm and one in which long-term concurrency was common—indicate that concurrent partnerships amplify the rate of HIV spread. Although the total number of sexual relationships was similar in both populations, HIV transmission was much more rapid with long-term concurrency and the resulting epidemic was about 10 times greater.
Exploring the Potential Impact of a Reduction in Partnership Concurrency on HIV Incidence in Rural Uganda: A Modeling Study
The study modeled the effects of reducing concurrency in a rural Ugandan population as a means to reduce HIV incidence while keeping the amount of sex constant. The behavioral, biological, and demographic data used in the model was collected on an annual basis in the rural South-West area in Uganda. There were three measures of concurrency in the model—long-term, short-term and total concurrency—and it was found that men’s rates were higher than women’s. About 10 percent of men reported a total concurrency level, 4 percent were long-term, and 2 percent were short-term. Less than one percent of women reported any type of concurrent behaviors. The model explored the effects of a 20 percent and 50 percent reduction in concurrency on HIV incidence. It was found that a 20 percent reduction in concurrency would reduce HIV incidence in 2020 by 4.1 percent in men, 9.2 percent in women, and 7.1 percent overall. A reduction of 6.0 percent in men, 16.2 percent in women, and 11.9 percent overall was found in HIV incidence in 2020 if concurrency was reduced by 50 percent. It was also found that reducing concurrency in one gender caused reductions in HIV incidence in the other gender. The authors conclude that countries with high levels of concurrency could benefit from a behavioral campaign targeting sexual concurrency, and that women’s rates of HIV incidence would be reduced if their male partners reduced their levels of concurrency.
Symposia Meeting Report: Addressing Multiple and Concurrent Partnerships in Southern Africa: Developing Guidance for Bold Action
This 19-page report provides an overview of a regional meeting held in Gabarone, Botswana, that was organized by Harvard, the World Bank, and UNAIDS. Approximately 40 representatives from various regional and international organizations participated. Topics include the theory, implementation, and evaluation of MCP programs in the region. The report provides links to the AIDS Prevention Research Project at Harvard, which in turn links to meeting symposia documents, prior meeting reports, such as the XVII International AIDS Conference in Mexico City, PowerPoint presentations, videos, and poster session reports.
Strategic Considerations for Communications on Multiple and Concurrent Partnerships within Broader HIV Prevention in Southern Africa (draft)
This eight-page guidance is the result of the joint UNAIDS/Harvard AIDS Prevention Research Project and the World Bank meeting, with input from civil society in Southern Africa. Intended as an aid to national programs and implementers, the draft document offers a rationale for its focus on concurrent relationships, strategic goals for communications around multiple and concurrent partnerships (MCP), and guidance regarding appropriate types of messages regarding MCP.
UNAIDS Consultation on Deﬁnition and Measurement of Concurrent Sexual Partnerships (April 20-21, 2009)
This brief, preliminary report on the meeting of the UNAIDS Reference Group on Estimates, Modelling, and Projections, in Nairobi, Kenya, provides an overview of the meeting goals and outcomes. Thirty-five experts addressed the problem of inconsistent definitions of concurrent sexual relationships and how information about concurrency can be obtained during surveys. The experts reached consensus on the following definition of concurrency: “Overlapping sexual partnerships where sexual intercourse with one partner occurs between two acts of intercourse with another partner.” A final set of specific recommendations for survey questions to capture information about concurrency is here.
UNAIDS Meeting Report: Multiple Concurrent Partnerships Campaigns and Communications, Towards a Coordinated Regional Response (September 17-18)
This 24-page meeting report builds on the earlier October 2006, SADC Regional Consultation. The document provides an executive summary, statement of goals, and includes sections on Research on Emerging Evidence; Updates on Experiences and Lessons learned from MCP; Effective Social Change Communication; and Building a Community of Practice.
M&E for MCP Programmes and Consultation on Concurrent Sexual Partnerships
The first of this two-part document is a 54-page guidance on monitoring and evaluation (M&E) for MCP programs, which is intended for use by National AIDS Commissions, HIV M&E officials and organizations in Eastern and Southern Africa. Definitions, epidemiologic evidence, and goals are provided along with guidance for measurement, measurement tools, and ways to incorporate M&E into existing efforts. Four appendices include guidelines and tools for qualitative research and options for measuring concurrency. The second part of this document is an 18-page Consultation on Concurrent Sexual Partnerships, which recommends the following definition of concurrency: “Overlapping sexual partnerships where sexual intercourse with one partner occurs between two acts of intercourse with another partner.” Due to confusion about the acronym, MCP, and its implications, experts recommend that when referring to concurrency, terms such as “concurrent partnerships,” or simply “concurrency,” should be used. If an acronym is desired, they suggest using "CP."
Policy Brief: Sexual Risk Behaviour among Men with Multiple, Concurrent Female Sexual Partners in an Informal Settlement on the Outskirts of Cape Town
This two-page brief highlights a respondent-driven survey of 421 hard-to-reach men living in an informal, urban settlement in South Africa. Sponsored by the U.S. Centers for Disease Control and Prevention, the study authors developed a surveillance system to measure key risk behaviors and HIV prevalence in a population of men who have multiple and concurrent sexual partners. Key findings were that the men used condoms inconsistently, engaged in “high levels of transactional sex and intimate partner violence,” and frequented establishments where alcohol consumption is high in order to find sexual partners. The researchers recommend interventions based on those findings.
Multiple and Concurrent Partnerships: A Support Document for Advocacy Actions and Decision Making Process in Mozambique
The brief web document provides an overview of study results of MCP in Mozambique and the reasons individuals engage in MCP. The study found that MCP is common between certain populations such as young girls and older men and men who have sex with women of the same age and/or younger. The report recommends that MCP interventions focus on certain target populations. Contact information is provided to obtain the full 18-page report, which is published by the Johns Hopkins Bloomberg School of Public Health Centre for Communication Programs.
Report on a PEPFAR technical consultation on “Addressing Multiple and Concurrent Sexual Partnerships in Generalized HIV Epidemics,” Washington, DC (October 29-30, 2008)
The 54-page report summarizes a technical consultation on MCPs held in Washington, DC. The meeting convened researchers, government representatives and program implementers from several nations, including the U.S. and Southern Africa, to address four themes: 1) the relationship between MCP and HIV transmission; 2) core components of MCP programs; 3) engendering community support for MCP activities; and 4) measuring program outcomes. The consultation concluded with group work and discussion on next steps. The report includes appendices with additional resources and a list of participants.
The Invisible Cure: Africa, the West, and the Fight Against AIDS
The author of this book, Helen Epstein, relates the story of her discovery of a long-forgotten study of Ugandan sexual behavior in the late 1980s and early 1990s. The study, conducted by Maxine Ankrah, an African American researcher, explored the myth that condoms were central to the decline in HIV prevalence in Uganda. Ankrah’s research demonstrated that the decline was preceded by a successful campaign to reduce MCP in Uganda. Epstein examines the risks of concurrent sexual relationships and the social and economic upheavals that gave rise to an "earthquake" in gender relations in Africa, contributing to the spread of HIV.
SADC Expert Think Tank Meeting on HIV Prevention in High-Prevalence Countries in Southern Africa Report. Maseru, Lesotho, (May 10-12, 2006)
This 18-page report provides an overview of the 2006 meeting’s stated goals to “reflect on the key drivers of the epidemic in the region and to provide suggestions for accelerating HIV prevention.” The 35 expert participants set priorities, including promotion of interventions to: reduce the number of multiple and concurrent partnerships; increase male circumcision; encourage male involvement and responsibility for sexual and reproductive health; promote HIV prevention and support; increase consistent and correct condom use; and continue programs to encourage delay of sexual debut along with programs for condom use and reduced partnerships.