Partner Reduction

Introduction

  1. Definition of the Prevention Area

    Partner reduction is a prevention strategy focused on decreasing overall number of partners in order to lessen the risk of becoming infected with or transmitting HIV. Increasingly, partner reduction efforts have focused not only on reducing the number of partners, but also the number of concurrent partners.

  2. Epidemiological Justification for the Prevention Area

    Partner reduction is one of the most effective HIV prevention strategies to date, with a large evidence base to support this intervention. During the HIV epidemics of the late 1980s and early 1990s, campaigns advocating partner reduction and fidelity among couples and other prevention strategies were launched in the United States and Uganda. Among these was Uganda’s “Zero Grazing” campaign.

    Surveys showed that the campaigns were at least modestly effective with some survey participants reducing both their concurrent sexual relationships and their total number of sexual partners. These changes were followed by declines in the incidence of new HIV infections, first in the United States among men who have sex with men, and later in Uganda among heterosexuals.

    HIV incidence and/or prevalence has declined in other countries, such as Cambodia, Zambia, Zimbabwe, Thailand, Ethiopia, and Kenya, and most experts attribute these declines, at least in part, to partner reduction strategies.

  3. Core Programmatic Components

    Partner reduction strategies may be directed at specific risk groups, such as those who engage in commercial sex or men who have sex with men. In hyper-epidemics in Southern Africa, where HIV is prevalent among the general population, some experts have advocated more generalized strategies that address the entire population.

    Partner reduction efforts must take into account complex, interrelated social, class, cultural, and economic factors. For example, social and gender norms play an important role in perpetuating attitudes and beliefs that encourage, or at least tolerate, men and/or women having multiple or concurrent partnerships. Researchers have found that despite a high level of knowledge about the mechanisms of HIV transmission and prevention, many people continue to engage in high-risk partnerships for a multitude of reasons, including poverty, economic “wants,” and violence. Men who, often by necessity, take jobs at a distance from their families may engage in commercial sex. Impoverished women may be compelled to engage in transactional or commercial sex. Young girls may engage in intergenerational sex with older men—who are more likely to have HIV—to obtain gifts.

    Partner reduction messages are often coupled with other prevention messages, such as abstinence, delay of sexual debut, correct and consistent condom use, and HIV testing. Programs may need to deliver difficult messages that individuals are put at risk of HIV not only by their own behavior but by their partners’ behaviors. Another important message is that alcohol abuse is often associated with less cautious behavior that can lead to sexual activity with multiple partners, thereby increasing the risk of HIV infection.

  4. Current Status of Implementation Experience

    It can be difficult to prove that partner reduction programs are the cause of declines in HIV prevalence because multiple simultaneous factors may play a role. Most experts agree, however, that the body of evidence, especially studies of regions where partner reduction strategies preceded other interventions, suggests that partner reduction strategies play an important role in reducing the incidence of HIV.

    Surveys conducted in Uganda provide some evidence that the partner reduction campaigns of the 1980s and 1990s led to the decline in HIV prevalence that began in the early 1990s.

    Educational activities alone may not be sufficient to change behavior, however. Some regions showed stable or escalating rates of high-risk behaviors despite widespread knowledge about HIV transmission. In these instances, structural changes are needed to allow people the freedom to make choices that can reduce their risk of HIV infection, such as the choice to leave a violent spouse with concurrent partners.

UPDATED 5/2011

What We Know

Changes in Sexual Behavior and a Decline in HIV Infection Among Young Men in Thailand

Nelson, K.E., Celentano, D.D., Eiumtrakol, S., et al. New England Journal of Medicine (1996), 335(5), pp. 297–303.

The prevalence of HIV among commercial sex workers in Thailand is 50 percent or higher, leading Thai officials in the 1990s to promote condom use for sex workers. This study evaluated the effects of that and other HIV prevention programs in Thailand. Five cohorts of 21-year-old male army conscripts in Northern Thailand were evaluated. There were modest declines in HIV prevalence between 1991 and 1995. The percentage of men who had sex with a commercial sex worker declined while condom use increased. The authors concluded that active monitoring of brothels to "encourage and enforce" condom use are critical to the success of HIV prevention.

Changes in Sexual Behaviour Leading to the Decline in the Prevalence of HIV in Uganda: Confirmation from Multiple Sources of Evidence

Kirby, D. Sexually Transmitted Infections (2008), 84(S2), pp. ii35–ii41.

Seven diverse sources of information— including historical documents, newspaper articles, and multiple small- and large-scale surveys from 1988 through 1995—provided evidence about changes in sexual behavior following a campaign launched in 1986 to reduce sexual partners in Uganda. According to these varied sources, the dramatic decline in HIV prevalence in Uganda during the 1990s was preceded by a reduction in sexual partners. Furthermore, they indicate that increases in condom use did not take place until the early 1990s, after HIV prevalence was already declining. The author concludes that concurrently emphasizing partner reduction and condom use is “much more effective than primarily promoting condom use (or abstinence).”

Evidence for Population Level Declines in Adult HIV Prevalence in Kenya

Cheluget, B., Baltazar, G., Orege, P., et al. Sexually Transmitted Infections (2006), 82(S1), pp. i21–i26.

National survey data show that HIV prevalence in Kenya peaked at about 10 percent in the late 1990s and declined to 7 percent by 2003. Age at first sex and use of condoms increased while the percentage of adults with multiple partners fell. Gonorrhea, chlamydia, and syphilis all declined in young women attending an antenatal clinic in Nairobi. The researchers concluded that HIV prevalence decreased in some but not all areas of Kenya and that some of the decline was due in part to high levels of AIDS deaths. Prior to 2000, there were more new infections than deaths—a situation that reversed after 2000. Prevention interventions increased after 2000, too late, say the authors, to explain the earlier decline in HIV prevalence.

Gender Differences in the Risk of HIV Infection Among Persons Reporting Abstinence, Monogamy, and Multiple Sexual Partners in Northern Tanzania

Landman, K. Z., Ostermann, J., Crump, J. A., et al. PLoS ONE (2008), 3(8), p. e3075.

Data from over 6,500 clients attending a HIV testing and counseling center in northern Tanzania between November 2003 and December 2007 found that one-fourth of females and one-tenth of males seeking testing were living with HIV. As expected, an increased number of sexual partners was associated with increased risk for seropositivity for men and women alike. Monogamous women who reported that their partner had other sex partners (or did not know) were 36 percent more likely to be infected with HIV than a monogamous woman whose partner was also monogamous. Furthermore, monogamous women were more likely than monogamous men to become infected with HIV. Despite study limitations, the authors concluded that their research demonstrates the limited protection of monogamy for women and highlights the role of male partner concurrency in infecting women. The authors suggest that the “overly simple formulation” of the ABC prevention strategy (if not abstinence, then be faithful, and if not faithful, then use condoms) is misguided. They argue that HIV prevention efforts that promote abstinence, partner reduction, and mutual monogamy take place concurrently with efforts to empower women to “better control their exposure risk.”

Role of Widows in the Heterosexual Transmission of HIV in Manicaland, Zimbabwe, 1998–2003

Lopman, B. A., Nyamukapa, C., Hallett, T. B., et al. Sexually Transmitted Infections (2009), 8(S1), pp. i41–i48.

AIDS is the most common cause of young widowhood in southern Africa and was found among 11 percent of women in this rural population. This study of over 900 widowers found HIV prevalence rates at follow up to be 54 percent among ever-widowed men and 63 percent among ever-widowed women. HIV infection was largely acquired prior to widowhood, however. While 50 percent of 48 sexually active widows reported engaging in transactional sex, only 3 percent of married women reported doing the same. Widows were significantly more likely to use condoms than married women, however (53 percent versus 8 percent, respectively). The authors used their findings in a mathematical model, estimating that between 8 percent and 17 percent of HIV infections over 20 years could be attributed to widow/widower sexual activity. Although widowers were more likely to transmit HIV to their partners, the greater prevalence of widows contributes to their higher overall transmission of HIV. The authors conclude that increasing the financial independence of widows through employment opportunities may reduce their reliance on new sexual partners for financial support.

Sexual Risk Factors for HIV Infection in Early and Advanced HIV Epidemics in Sub-Saharan Africa: Systematic Overview of 68 Epidemiological Studies

Chen, L., Jha, P., Stirling, B., et al. PLoS ONE (2007), 2(10), pp. e1001.

Multiple partners, paid sex, and sexually transmitted infections (STIs) are considered risk factors for HIV infection at the early stages of the HIV epidemic when HIV is concentrated among high-risk groups. These risk groups are assumed to be less important once the epidemic has spread into the general “low risk” population. The study authors used 68 epidemiological studies from 18 countries in sub-Saharan Africa from 1986 to 2006 to test this common assumption. The general trends over time indicate that these risk factors are also strong predictors of HIV infection in high-prevalence settings, where epidemics were considered generalized. Contrary to common wisdom, multiple partners, paid sex, and STI infection remain important risk factors in countries with high HIV prevalence. The authors recommend reexamination the UNAIDS definition of a “generalized” epidemic and the recommendation that STI treatment in countries with advanced epidemics is likely to have a low impact. This analysis indicates that prevention efforts should focus on vulnerable groups, regardless of where the trajectory of the HIV epidemic lies.

Trends in HIV-Related Behaviors and Knowledge in Uganda, 1989-2005: Evidence of a Shift Toward More Risk-Taking Behaviors

Opio, A., Mishra, V., Hong, R., et al. Journal of Acquired Immune Deficiency Syndromes (2008), 49(3), pp. 320–326.

According to several large-scale surveys of adult Ugandans, knowledge about HIV peaked in 2001 and then remained stable while self-reported behaviors both improved and worsened. At the same time that the proportion of men who were abstinent increased, the number of men who had sex with multiple partners also increased, rising from one in four men in 2001 to one in three in 2005. The authors suggest that two factors may underlie increases in sexual risk-taking behaviors: complacency as antiretroviral drugs have become widely available between 2000 and 2005, and a shift in funding away from behavior change programs to medical access programs.

Masculinities and Multiple-Sexual-Partners in Historical Perspective in KwaZulu-Natal: The Making and Unmaking of Isoka

Hunter, M. Culture, Health, & Sexuality (2005).

Perceptions of masculinity change over time and within various social and political contexts, according to this doctoral dissertation of a student describing his assessment of male behaviors in KwaZulu-Natal. South Africa, says the author, has been greatly influenced by capitalism, migrant labor, and Christianity. In the 1970s, high rates of unemployment affected men’s ability to marry, establish an independent household, and become the head of a household. Multiple partners were a means by which men could express their masculinity since traditional customs were more difficult to achieve. The author closes with critical assessments of groups that are currently challenging old concepts of masculinity.

HIV Decline Associated with Behavior Change in Eastern Zimbabwe

Gregson, S., Garnett, G. P., Nyamukapa, C. A., et al. Science (2006), 311(5761), pp. 664–666.

Researchers followed a cohort of nearly 10,000 adults in Manicaland over three years to assess the trajectory of the HIV epidemic. Overall HIV prevalence declined among men and women, with steepest declines taking place among men ages 17-29 (23 percent) and women ages 15-24 (49 percent). At baseline, nearly half of 17- to 19-year-old males reported being sexually active; three years later, only one-fourth of such males reported the same. Women ages 15-17 also reported declines in sexual debut (21 percent to 9 percent over three years). Furthermore, there was a significant decrease in the number of casual partners that men and women reported in the previous year. Similar declines in HIV prevalence took place in areas with and without partner reduction interventions. Because these findings mirror national HIV and local antenatal clinic HIV prevalence rates, the researchers concluded that there is a trend toward a declining prevalence of HIV in Zimbabwe. This decline may follow a pattern similar to Uganda’s—driven by delay of onset of sexual debut, partner reduction, with some contribution of consistent condom use with casual partners. Because prevalence estimates "reflect accumulation of infections over a period of more than 10 years," however, the researchers state that HIV prevalence is "insensitive to behavior change."

The Role of Partner Reduction and Faithfulness in HIV Prevention in Sub-Saharan Africa: Evidence from Cameroon, Rwanda, Uganda, and Zimbabwe

Mishra, V., Hong, R., Bignami-Van Assche, S., & Barrere, B. USAID, Demographic and Health Surveys Working Papers (2009), No. 61.

This study used national-level data to explore the association between risk of HIV infection, multiple sexual partnerships, and partner faithfulness among men, women, and cohabiting couples in the HIV epidemics in Cameroon, Rwanda, Uganda, and Zimbabwe. M ultiple sexual partnerships varied among countries, but tended to be common in these countries. In general, men reported having more partners than did women, and were less faithful (lifetime faithfulness and recent faithfulness) than were women. As one would expect, the higher the number of lifetime sexual partners, the greater the likelihood of HIV infection. Similarly, those who reported being faithful to their partners were less likely to be infected with HIV than those who were unfaithful.

A Typology of Groups at Risk of HIV/STI in a Gold Mining Town in Northwestern Tanzania

Desmond, N., Allen, C. F., Clift, S., et al. Social Science & Medicine (2005), 60(8), pp. 1739–1749.

HIV prevention interventions have generally targeted risk groups such as migrant laborers and sex workers in mining communities in sub-Saharan Africa. In 2002, the authors conducted several hundred informal interviews of women who openly engaged in commercial sex, and of women who engaged in less visible forms of transactional sex, such as shopkeepers or barmaids, who made most or part of their income by selling sex. They concluded that traditional epidemiological categories of "general," "bridging," and "core" populations did not adequately describe the range of HIV risk. The lure of economic opportunity in mining towns led to dislocation and travel that facilitated spread of HIV among seemingly low-risk individuals. Interventions, say the authors, should focus on decreasing income disparities and providing housing for mineworkers’ families.

Alcohol Abuse, Sexual Risk Behaviors, and Sexually Transmitted Infections in Women in Moshi Urban District, Northern Tanzania

Ghebremichael, M., Paintsil, E., & Larsen, U. Sexually Transmitted Diseases (2009), 36(2), pp. 102–107.

This study found that among nearly 1,000 women living in an urban district of northern Tanzania, those who abused alcohol were more likely to report having been a victim of physical abuse or sexual violence. They were also more likely to report having more than one sexual partner in the previous three years. There was no significant association between alcohol abuse and sexually transmitted infections (STIs). The authors concluded that alcohol abuse was not directly associated with prevalence of STIs but was indirectly associated through its association with multiple partners. The authors suggest that the subordinate status of women may lead some women to stay with men who are unfaithful and violent and that such women may use alcohol to “self-medicate” underlying problems. They conclude that alcohol use screening should be a part of comprehensive STI and HIV prevention programs.

Barriers to Preventing Human Immunodeficiency Virus in Women: Experiences from KwaZulu-Natal, South Africa

Abdool-Karim, Q. Journal of the American Medical Women’s Association (2001), 56(4), pp. 193–196.

The authors of this study surveyed women in KwaZulu-Natal, South Africa, from 1991 to 1993, and found that nearly 9 out of 10 women were sexually active. Virtually all of the women understood the role of condoms in HIV prevention, but only a little over 1 in 10 had ever used a condom. Many women said they did not insist on using a condom because their partners would get angry if they did. Unlike other researchers, these authors found that high levels of knowledge about HIV transmission did not influence safer-sex behavior changes for several reasons, including threats of violence against women, alcohol abuse, financial dependence on men, and women’s status as child bearers.

Putting it Into Practice

Proven HIV Prevention Strategies

The Global HIV Prevention Working Group. (2006).

This fact sheet reviews effective strategies to prevent all three routes of HIV transmission: sexual, blood-borne, and mother-to-child. The interventions are listed by route of transmission. Although effective interventions have been identified over the past two decades, access to these interventions is low. Researchers with UNAIDS and the World Health Organization estimate that increased program access could avert half of the 62 million new HIV infections that are expected to occur between 2005 and 2015.

Men as Partners (MAP) Project

EngenderHealth and The Planned Parenthood Association of South Africa. (2011).

Men as Partners (MAP) focuses on gender equity which includes gender based violence and behaviors that make both men and women vulnerable to HIV infection. The program is implemented through workshops with men, and through local and national public education campaigns that use a variety of channels to reach men. MAP challenges stereotypes that equate manliness with risky behaviors, such as violence, alcohol use, pursuit of multiple sexual partners, and domination over women. The program takes a positive approach toward male involvement and the sessions are respectful, open-minded, and address the needs of the men in the community. The project is active in over 15 countries.

Current Thinking About Behavior Change and Prevention of Sexual Transmission of HIV

Wilson, D. World Bank (2006).

This PowerPoint presentation outlines how HIV epidemics are more heterogeneous than initially thought. Concurrent sexual relationships in southern Africa are cited as an important factor with half of all transmissions occurring during the first five months of infection. The “lethal cocktail” of concurrent sexual relationships and low circumcision rates are said to increase transmission up to 30-fold, and may be a factor explaining the hyper-epidemic in southern Africa. The effects of protective strategies are reviewed for concentrated and generalized epidemics, and male responsibility for reducing transmission is emphasized.

Trusted Partner Campaign

Population Services International (PSI). (2005).

PSI’s “Trusted Partner” mass media project was launched after a research team found that youth often use ineffective criteria to evaluate their personal risk of HIV within trusted relationships. Trusted Partner uses a variety of media messages to counter myths about HIV and to promote fidelity, awareness of the risks of mixing alcohol and sex, and HIV testing. One poster produced by the project provides a visual reminder that even healthy-looking people can have HIV. Trained field officers encourage community leadership and peer education in drinking establishments, while creating forums to address partner reduction and risky drinking. The program has been reproduced at low cost in 11 countries in east and southern Africa.

Transforming Male Gender Roles to Combat Cross-Generational Sex, ‘Be a Man’ Campaign

Clifton, D, Population Reference Group, Interagency Gender Working Group (2009)

The ‘Be a Man’ campaign is one component of the Young Empowered and Healthy (Y.E.A.H) initiative that is being implemented in Uganda to address cross-generational sex. The campaign was initiated in 2006 to challenge traditional male norms, attitudes, and behaviors. Objectives of the campaign included improving gender equitable beliefs and attitudes, and redefining what a “respectable” African male is in modern day society.
The campaign used a combination of television, radio, small media, and community channels. A qualitative evaluation was conducted in 2007. It found that the target group favored responsible men who did not have multiple partners, drink, or use violence against women. However, respondents also identified certain circumstances when a man was justified in not performing these behaviors.

Couple-to-Couple Influence in HIV Prevention: The UPHOLD-TUKO Network Experience in Uganda

Kintu, P., Mugume, A., Buzaalirwa, L., et al. HIV Implementers Meeting, Kigali, Rwanda (2007).

This presentation describes the experiences and lessons learned of the UPHOLD-TUKO Network in Uganda. The Tuko Club is a civil society organization of married couples that promotes mutual faithfulness among couples. The 2004 Uganda National HIV/AIDS Sero-Behavioral Survey found that the overwhelming majority of high-risk behavior occurs with spouses and that almost half of new infections arise from sex with one’s spouse. Married couples are trained as trainers. The definition of “marriage” is broad-based (e.g., religiously married, cohabiting, traditional) to ensure inclusivity of all couples. Standardized messages are supported through the use of training materials with consistent content that are translated into the local languages.

Listen to the Bell: HIV/AIDS Prevention Through Zero Grazing

Kivamwo, S. Association of Journalists Against AIDS in Tanzania (2007)

The Sikia Kengele (“Listen to the Bell”) campaign uses community opinion leaders known as “Bell Ringers” to initiate talks on the risks of multiple partners and the benefits of knowing one’s HIV status. The campaign is conducted in high-risk areas of Tanzania such as major transportation corridors, mining communities, and plantations. The Bell Ringers initiate discussions in bars, stadiums, churches, mosques, farms, and bus stations, and they prepare “Giant Bell” road shows in which a high profile team of influential individuals and educators (and large bell) travel to communities, using music, dance, dramas, and cinema to encourage people to reduce the number of sexual partners.

Evaluation of the African Youth Alliance Program in Ghana, Tanzania, and Uganda: Impact on Sexual and Reproductive Health Behavior among Young People (Summary Report)

Williams, T., Mullen, S., Karim, A., & Posner, J. JSI Research & Training Institute (2007).

The African Youth Alliance (AYA) Program, implemented in Ghana, Tanzania, and Uganda, uses peer educators to encourage counseling and testing for HIV and sexually transmitted infections. A 2005 evaluation of the program found that it had a significant impact on condom use, partner reduction, and several knowledge measures. The impact of the AYA program was greater for females than males. For example, partner reduction was demonstrated in both males and females in all three countries but it was statistically significant only in females in Ghana and Tanzania.

Tools and Curricula

Men As Partners: A Program for Supplementing the Training of Life Skills Educators

EngenderHealth and the Planned Parenthood Association of South Africa. (1999).

Developed for use by Men As Partners educators, trainers can use this manual to design workshops that best fit their needs: A one-day HIV and AIDS workshop, for example, or a five-day life skills workshop. It can also be used for different groups, such as men only, adolescents, or men and women together. Interactive exercises, case studies, and icebreakers are used to highlight key points about life skills. Training activities include topics on gender and sexuality (such as gender stereotypes, sexual orientation), male and female sexual health (anatomy and physiology, myths and facts about family planning), HIV and other sexually transmitted infections, relationships (fatherhood, unhealthy relationships), and violence, including examining and defining sexual and domestic violence.

Partner Reduction Session Flip Chart for the General Population

Population Services International. (undated).

The flipchart has 11 question sheets with accompanying pictures to stimulate discussion among people participating in HIV prevention group discussions. The topics cover multiple partners, assessing risk, condom use, interspousal communication, sexually transmitted infections, voluntary counseling and testing, gender and HIV, and drug and alcohol abuse.

Partner Reduction and Protection: Outreach Guide 10

Pact/Botswana. (undated).

This practical 48-page outreach guide with picture codes details 18 exercises aimed at educating participants about partner reduction and condom use. Some of the exercise titles include “Personal Risk Assessment,” “Double Standards,” “Excuses not to Use Condoms,” “Condom Relay Games,” and “Encouraging Reluctant Partners.” Case scenarios (such as a young woman with a poor boyfriend and a married sugar daddy) along with a series of related questions provide structured guides for discussion and education.

Additional Resources

“Know Your Epidemic, Know Your Response”: A Useful Approach, If We Get It Right

Wilson, D. & Halperin, D. T. The Lancet (2008), 372(9637), pp. 423-426.

This commentary provides advice to the HIV prevention community for responding effectively to vastly differing HIV epidemics throughout the world without becoming overly complicated. Despite differences in concentrated, generalized, and mixed epidemic settings, partner reduction can help reduce HIV transmission in generalized epidemics. The authors argue that the response to AIDS has “given insufficient emphasis to aligning prevention priorities with epidemic transmission dynamics, compromising effective prevention with mismatched or unfocused responses.” They conclude that much can be done to help curb HIV transmission around the world if the right approaches are taken, such as partner reduction and male circumcision.

Field Assessment of Emergency Plan Centrally-Funded HIV Prevention Programs for Youth

Speizer, I. & Lopez, C. (2007). MEASURE Evaluation Project, University of North Carolina at Chapel Hill, NC, USA.

This report evaluates “abstinence and be faithful for youth” (ABY) programs funded through The President’s Emergency Plan for AIDS Relief. The researchers visited 20 sites in Ethiopia, Haiti, Kenya, Mozambique, and Tanzania to assess how program efforts had fared. The authors recommend how these programs can strengthen their work, and also provide an assessment tool that can be used by funders, program planners, and program managers to identify characteristics of strong ABY programs.

Outreach Guides: HIV/AIDS Prevention Programmes

Pact/Botswana. (2007).

This series of outreach guides developed in Botswana address many topics that HIV prevention programs can use for their efforts. The guides include participatory methods, and were developed to help promote changes in behavior. Ten guides include information and activities on abstinence promotion, alcohol abuse reduction, improving couple communication and parent-child communication, as well as partner reduction.

UNAIDS Combination Prevention Briefs

Joint United Nations Programme on HIV/AIDS (UNAIDS). (2008).

The Joint United Nations Program on HIV/AIDS (UNAIDS) published a series of briefs to provide an overview of combination prevention approaches in high-prevalence countries in Eastern and Southern Africa. The briefs contain background information, programmatic challenges, and recommendations for action with a focus on modes of transmission, multiple concurrent partnerships, vulnerabilities of women and girls, and male circumcision