Table of Contents
Making Prevention Work
1. Behavior Change Communication: From Individual to Societal Change
2. Improving STD Prevention and Treatment
3. Prevention Marketing: Condoms and Beyond
4. Policy Development and HIV/AIDS Prevention: Creating a Supportive Environment for Behavior Change
5. Behavioral Research: Using Results to Design Behavior Change Interventions (See Below)
6. Evaluating HIV/AIDS Prevention Programs: Developing New Tools for Meaningful Measurement
7. Women, Men and HIV/AIDS: Building Gender-Sensitive Programs
8. Managing HIV/AIDS Programs and Building Capacity to Sustain Prevention Efforts
9. Prevention and Care: Mutually Reinforcing Approaches
10. Crossing Borders: Reaching Mobile Populations at Risk
Partners and Acronyms
5. Behavioral Research: Using Results to Design Behavior Change Interventions
Understanding the behaviors that put people at risk of HIV infection is the fundamental challenge of HIV/AIDS prevention. Since the beginning of the epidemic in the early 1980s, behavioral research projects have investigated such sensitive issues as sexual behavior and drug use and have illuminated many of the cultural and social factors that influence risk behavior.
In response to rapid shifts in the epidemiology of the HIV/AIDS pandemic, strategies for conducting behavioral research have changed over the past six years. Well-designed rapid studies increasingly received priority over more long-term thematic research, and scientists moved away from repetitive studies of knowledge, attitudes, beliefs and practices toward research that provided practical information to guide interventions.
As the epidemic expanded, behavioral research also moved beyond studying the behavior of traditional "high-risk groups" to research with populations previously considered at low risk of HIV infection, such as adolescents and women. And as the number of people living with HIV increased dramatically, understanding the risk behavior of those who are HIV-positive and testing interventions to support their behavior change began to receive more attention.
Understanding Risk Behavior
Behavioral research conducted under AIDSCAP examined issues central to both global and local prevention efforts. Research projects ranged from small, rapid, program-related studies to large-scale, multisite efficacy trials.
AIDSCAP's behavioral research was designed to expand the scientific knowledge of HIV risk behavior and to have a direct impact on the development of prevention programs. Program-related research addressed critical issues such as barriers to condom use among young women in the Dominican Republic, strategies for heterosexual couples to use in renegotiating sexual relationships to change high-risk behavior, and whether knowledge of HIV status leads to behavior change.
Behavioral research results were used to design effective interventions in many countries. In São Paulo, Brazil, for example, the success of a controlled intervention trial targeting young adults completing their primary and secondary education in night schools led to an HIV/AIDS prevention program for adolescents in 2,800 public secondary schools. A total of 300,000 manuals based on the curriculum developed for the trial were distributed to students throughout São Paulo. And in Thailand, a pilot intervention study using the Royal Thai Army's command structure and informal networks among conscripts was adapted by the military to provide HIV/AIDS prevention services for all recruits.1
AIDSCAP's largest intervention trial, cosponsored by the Joint United Nations Programme on HIV/AIDS (UNAIDS) and implemented with research institutions from Kenya, Tanzania, Trinidad and the United States, was a multisite study of the efficacy of voluntary HIV counseling and testing (Box 5.1).2 The results of this study will offer guidance to policymakers on the cost effectiveness of such services.
|
5.1 Study Examines Counseling and Testing for HIV/AIDS Prevention
When 22-year-old Anne Kanjiri of Nairobi found out that she had tested negative for HIV, the news was more than just an opportunity to celebrate. It was a chance to make some changes in her life.
"I used to have many sexual partners before I was counseled and tested for HIV," she said. "When I got my results, I decided to get married instead."
To researchers from the Kenya Association of Professional Counselors (KAPC), her response is as significant as the results of her test. Kanjiri* was a participant in the first randomized study of the impact of counseling and testing (C&T) on behavior change for HIV/AIDS prevention among people voluntarily seeking such services in developing countries.
Through randomized controlled trials conducted in Nairobi by the KAPC and at other centers in Tanzania, Trinidad and Indonesia, the co-sponsors of the study hope to discover whether HIV testing accompanied by personalized, one-on-one counseling can influence individuals to adopt preventive behaviors and lower their risk of HIV infection. In the past, a limited number of nonrandomized studies of C&T's impact among specific populations had yielded mixed results.
Kanjiri was just one of 4,298 volunteers who participated in the study at the AIDSCAP-sponsored sites in Kenya and Tanzania and at the site in Trinidad funded by UNAIDS. Because of the breadth of the study and the size of the populations, AIDSCAP and UNAIDS expect the result to be significant for prevention programs worldwide.
An international effort, the research was conducted by the KAPC in Kenya, the Muhimbili University College of Health Sciences in Tanzania and the Queens Park Counseling Center in Trinidad, with The Center for AIDS Prevention Studies at the University of California at San Francisco serving as the coordinating center.
Study participants at the three sites were assigned randomly to one of two interventions, counseling and testing (C&T) or health information. Subjects in the C&T arm received pretest counseling and had blood taken for HIV antibody testing. After test results were available, each C&T participant received counseling. Those in the health information group -- the study's control arm -- were shown an informational video about HIV and STD prevention and given condoms, along with training in how to use them.
All participants were invited to return to the study center twice. At six months, participants in both groups were tested for STDs and offered the option of counseling and an HIV test. Researchers also administered a follow-up questionnaire to those in both groups at six months and again at 12 months.
During the interviews, data were collected on sexual behavior, psychological status, knowledge and attitudes about HIV/AIDS and other STDs, and care-seeking for STDs. The STD test results will supplement the self-reported behavioral data, providing a more objective measure of behavior change.
The results of the study will be available by the end of 1997. But before the data analysis had even begun, researchers had already learned a great deal from their experiences in Kenya and Tanzania. Early fears that they would not be able to recruit enough participants soon disappeared as the enthusiastic response to the study revealed a great demand for HIV counseling and testing in both Dar es Salaam and Nairobi. Even after recruitment had ended, four to ten people came to each center every day to seek counseling and testing.
People's willingness to return for the their test results was another sign of how much they valued the service. An unusually high proportion of those assigned to the C&T arm -- more than 85 percent -- returned for their results one week after the test without any prompting.
Researchers at both sites did have trouble recruiting enough people to participate in the study as couples. Most people preferred to enroll alone, often bringing a partner to the center months later for counseling and testing. Counselors reported that although couple counseling was extremely challenging at the beginning of the study, they found it more effective because it gave them opportunities to assist couples in negotiating behavior change.
Economic data gathered during the study will help determine whether voluntary HIV counseling and testing is a cost-effective prevention intervention in developing countries, and whether such services can be sustained. Researchers already know that more than half the study participants would be willing to pay up to the equivalent of U.S.$2 for C&T services. Concerns about the cost of such services will undoubtedly continue, but further analysis of the data will give policymakers clear guidance on whether C&T is effective -- and ultimately worth the expense.
*Confidentiality was strictly observed in this study. Anne Kanjiri later agreed to be interviewed by a journalist and quoted by name. |
AIDSCAP also advanced the science of HIV/AIDS prevention by addressing emerging global issues such as the role of structural and environmental interventions in reducing HIV incidence, prevention interventions to help protect women in stable relationships, and the linkages between HIV prevention and care. A series of concept papers reviewed the research on each of these issues and defined research questions for the future.3-5 These questions led to research studies and pilot interventions that offer recommendations and models for the next generation of HIV/AIDS programs.
In Kenya, for example, an in-depth study of communication between men and women in stable relationships identified the best ways to encourage dialogue between partners about HIV risk and condom use. This research study showed that through counseling, couples could learn to discuss sex and HIV risk reduction within their relationships. In the Dominican Republic, formative research among sex workers and their clients, employers and steady partners was used to design a demonstration project adapting a highly successful structural and environmental intervention from Thailand (Box 5.2). And in Tanzania, one of the first studies to examine the links between care programs for HIV-positive individuals and their motivation to adopt safer sexual behavior to protect others will be used by program staff and policy-makers to design HIV/AIDS support programs that include prevention activities.
Findings from these and many other behavioral research projects were disseminated through publication of peer-reviewed articles and presentations at national and international scientific conferences. AIDSCAP's contribution to the scientific literature on HIV risk behavior and prevention includes more than 50 journal articles and more than 75 conference presentations.
In addition to guiding program interventions and addressing global research issues, AIDSCAP's behavioral research strategy emphasized capacity building. Through collaborative partnerships, training and technical assistance, the project enhanced the capacity of more than 150 social scientists and 100 institutions to conduct behavioral research for HIV/AIDS prevention. AIDSCAP also supported the Visiting Scholars Program for developing country researchers of the Center for AIDS Prevention Studies (CAPS) at the University of California at San Francisco, an FHI partner in the project's behavioral research activities (Box 5. 3).
Each of AIDSCAP's ten competitive research grants was awarded to a research team of scientists from developing and developed countries. Most of the research the project commissioned to answer important questions in HIV/AIDS prevention and all program-related studies were also conducted with host-country partners. This collaborative approach encouraged the transfer of knowledge and skills, enhanced the acceptability of research results and gave local researchers a stake in ensuring that their findings were used to improve prevention programs.
|
5.2 100 Percent Condom Use: Adapting Thailand's Policy in the Dominican Republic
Since 1989, Thailand's government has required condom use in all the country's brothels. Enforcement of this policy through legal sanctions against brothel owners, combined with a mass media campaign, has led to dramatic increases in condom use and decreases in STD rates.
The Thai "100 percent condom policy" has been hailed as an example of the kind of structural and environmental interventions needed to reduce barriers to individual HIV risk reduction. But would the policy work in other countries?
Results of a study conducted by AIDSCAP and the NGO COIN (Centro de Orientación e Investigación Integral) suggest that in the Dominican Republic, the answer to that question may be yes. Formative research identified strong support and practical recommendations for adapting the Thai policy to the Dominican context.
Rapid ethnographic research conducted in the summer of 1996 included participant observation in five brothels and other commercial sex establishment and repeated in-depth interviews with brothel owners, clients, and sex workers and their steady partners. In just six weeks, researchers conducted more than 200 interviews.
The most surprising finding was the positive response from sex workers, brothel owners and clients to proposals to promote and monitor condom use in commercial sex establishments. Instead of considering medical check-ups of sex workers and legal sanctions oppressive, most respondents saw such policies as supportive.
Sex workers said policies requiring condom use in commercial sex would not only protect their health, but also make it easier to negotiate condom use with clients. The women reported that they spent a great deal of time and effort trying to convince clients to wear condoms. Owners and managers believed that having an STD-free establishment would increase prestige, clients and profits. Steady partners of sex workers and -- for the most part -- clients were also supportive, citing fears about their own health and the health of their partners and children.
The study results suggested that a program promoting condom use through both policy change and solidarity among sex workers and brothel owners was likely to be most effective in the Dominican Republic. The results also revealed the importance of addressing sex workers' risk of STD and HIV infection from steady, non-paying partners.
These findings were used to design a pilot 100 percent condom intervention that included training for brothel staff to promote group solidarity as well as promotion of a government policy on mandatory condom use. Such a policy would be enforced by sanctions (fines and brothel closings) against Dominican commercial sex establishments that did not enforce consistent condom use and certificates for those that did. Compliance would be monitored primarily through regular STD screening of sex workers.
AIDSCAP began testing some components of the pilot project in May 1997 in ten sex establishments. Brothel owners and managers, sex workers and other brothel employees all received training in the principles, rights and responsibilities of a "collective agreement" to promote and support condom use in commercial sex. COIN held the first of a series of special workshops on communication and sexual negotiation for sex workers and their steady partners. It also organized and supervised the effort to institutionalize structural strategies such as STD screening, cards certifying that sex workers are STD-free, improved condom access, and spot checks for used condoms in brothel rooms.
Full implementation of the pilot project and eventual expansion nationwide, however, will require a formal government policy mandating condom use during all commercial sex acts and imposing legal sanctions for noncompliance. The dialogue that AIDSCAP and COIN initiated with the government continues in order to encourage the policy changes needed for effective structural interventions to support 100 percent condom use among sex workers and their clients in the Dominican Republic. |
|
5.3 Building Local Capacity in HIV/AIDS Research
Dr. Maria Eugênia Lemos Fernandes arrived at the University of California at San Francisco in the summer of 1989 with an idea. A physician and director of the HIV/AIDS prevention program in the Department of Health in the state of São Paulo, she wanted to find out more about the epidemiology of HIV-1 and HIV-2 among one of the groups most as risk of infection -- women involved in commercial sex -- and the factors that influenced that risk.
During the next ten weeks, she spent much of her time working with colleagues at the Center for AIDS Prevention Studies to design a study that would provide the background information needed to develop prevention interventions to reach sex workers in three cities with high HIV prevalence: São Paulo, Santos and Campinas.
Like other graduates of the CAPS Visiting Scholars Program, an intensive program designed to build the HIV/AIDS prevention research skills of scientists from developing countries, Dr. Fernandes was able to take the study protocol she developed in San Francisco, conduct the research in her own country and publish the results.1 This research provided the baseline data for interventions implemented as part of the AIDSCAP program in Brazil and the rationale for important financial support from the World Bank for HIV/AIDS prevention in the state of São Paulo.
Eight years later, Dr. Fernandes is still working to help women and men in São Paulo and other parts of Brazil protect themselves from HIV/AIDS. As head of the NGO Associação Saúde da Família and the former resident advisor of the AIDSCAP program in Brazil, she manages an HIV/AIDS program that builds the capacity of local organizations to develop effective and sustainable prevention projects in the states of Fortaleza, Salvador, São Paulo and Rio de Janeiro.
Each year up to ten scientists participate in the CAPS program. During their time in San Francisco, they attend seminars on epidemiology, research design, data management, biostatistics, and the behavioral and psychosocial aspect of the HIV/AIDS epidemic. But the program's main emphasis is helping the scientists design HIV/AIDS prevention research projects to carry out in their own countries.
CAPS provides pilot project funding and offers continuing technical assistance to enable the scientists to implement these research protocols. A number of participants have returned to San Francisco to participate in a writing sabbatical program that gives them time to work with CAPS faculty analyzing their data and preparing manuscripts.
This collaboration between CAPS and visiting scientists has resulted in quality research projects in more than 23 countries and forged a strong international network of HIV/AIDS researchers. Alumni from the program have published a number of articles in peer-reviewed journals, including nine articles featured in a special issue of the journal AIDS in 1995, and many -- like Dr. Fernandes -- have assumed leadership positions in national and international HIV/AIDS prevention efforts.
AIDSCAP supported scientists' participation in the Visiting Scholars Program for six years. Recent behavioral research projects conducted by AIDSCAP-funded scholars include:
- a study on the impact of a peer-led educational intervention that reached Balinese youth through a network of traditional youth organizations in Bali.
- an assessment of risk behavior among male clients at a Bombay STD clinic and formative research to design an HIV prevention intervention for clients.
- a pilot study on the use of social networks to empower young, unmarried women in Senegal to negotiate for safer sex.
As these examples illustrate, one of the strengths of the CAPS program is its focus on practical research in developing countries. For a relatively modest investment of $10,000 or less for each study, the program produces quality collaborative research and develops a strong local research capacity. The visiting scientists serve as principal investigators for the studies they design, ensuring their commitment to the research and involvement in all aspects of the research process.
For Dr. Fernandes, the experience has had an enduring impact. "My participation gave me a new vision of public health that certainly influenced my professional life and consequently the quality of my work in designing and implementing programs in my country," she said.
- Lurie P, Lemos Fernandes ME, Hughes K, et al. (1995). Socioeconomic status and the risk of HIV-1, syphilis and hepatitis B infection among sex workers in São Paulo, Brazil. AIDS 9(suppl 1):S31-S37.
|
Lessons Learned
- HIV/AIDS prevention requires a multidisciplinary approach to research.
Answering many of the most important research questions requires perspectives from such disparate fields as STD management and prevention, social marketing, medicine, counseling, psychology, epidemiology, communications and family planning.
AIDSCAP's HIV Counseling and Testing Study, implemented in Tanzania and Kenya by AIDSCAP and in Trinidad by UNAIDS, is a good example of a multidisciplinary research project.2 A randomized, controlled trial of the impact of voluntary HIV counseling and testing on risk behavior, the study examined psychological, behavioral, epidemiologic, operations and cost-effectiveness issues (Box 5.1).
Collaboration and Capacity Building
- Matching local research institutions with NGOs that implement interventions is a particularly effective way to organize research.
Such partnerships can offer NGOs a sustainable source of technical assistance and help strengthen local research capacity. For example, a Tanzanian NGO collaborated with a researcher from Muhimbili University on an AIDSCAP-sponsored study of how care and support services for people living with HIV/AIDS affect risk behavior. This collaboration enabled the Tanga AIDS Working Group to continue focusing on providing HIV/AIDS services, including counseling and home-based care, while the researcher provided technical assistance in research methodology and data analysis.
- Collaboration with international institutions generates support and a high profile for projects.
Bringing together highly skilled and experienced people from international institutions to collaborate on research is time consuming, but it facilitates acceptance of the findings because key institutions have been involved in the research process. AIDSCAP's counseling and testing study linked host-country and international scientists through an executive committee that included the local principal investigators, AIDSCAP and CAPS scientists, and collaborators from the World Health Organization's Global Programme on AIDS and UNAIDS. The committee made decisions about the study through regular meetings and conference calls.
Appropriate Scale
- Large, multisite intervention trials such as AIDSCAP's HIV counseling and testing study have an important but specific role to play in HIV/AIDS prevention.
Such projects can create opportunities to share resources, make cross-site comparisons, and sometimes even pool data to enhance statistical power. But they should only be used to test technologies and approaches that have global significance and have not already been the subject of a rigorous efficacy trial.
- Rapid, relatively inexpensive studies are useful for projects that are: (1) linked to interventions under development, (2) of local or regional interest, (3) associated with interventions that are highly culturally specific or that vary significantly by population type, or (4) adaptations of successful interventions from other regions or target populations.
For example, results from a nine-month qualitative study of the contributing factors and motivations for risk behavior among Nicaraguan sex workers, their clients, and men who have sex with men provided information critical to the development of a national HIV/AIDS communication strategy. And in the Dominican Republic, rapid research techniques were used to assess the feasibility of adapting Thailand's "100 percent condom policy" (Box 5.2).
Ethical Issues
- Interventions tested in efficacy trials can be both feasible and of high quality.
The tradeoffs involved in testing interventions in resource-poor countries are a source of continuing debate. Some argue that it is unethical to test "gold-standard" technologies or approaches in countries that cannot afford to implement them. Others contend that the only way to encourage adaptation of innovations and to attract financial support for their implementation in developing countries is to first establish their efficacy.
AIDSCAP's experience shows that ensuring the feasibility of interventions to be tested does not necessarily mean that quality must be compromised. For example, research among military recruits in Thailand assessed the impact of an intervention designed to be used at other military bases. Instead of testing an intensive, one-on-one intervention that would have been too costly to replicate, the study used the military's leadership structure and the natural social networks among the soldiers to reach recruits with education, counseling, HIV testing and peer support. When study results showed that the intervention had led to changes in behavior, the Thai military adopted it on a wider scale.
- Ethical review of behavioral research protocols by local review boards ensures that the research is culturally sensitive and responsive to the needs of the communities where it is carried out.
A local committee typically offers the best perspective on local sexual and social mores and practices and often has greater credibility with local scientists than committees overseas.
Research Tools
- Theoretical behavior change models provide useful frameworks for examining HIV risk-taking behavior in a variety of cultures and societies.
Behavior change interventions appear to be influencing risk behavior, but without a theoretical framework, the reasons why some individuals have adopted prevention measures and others have not remain elusive. AIDSCAP's use of behavioral theories developed in the United States to study sexual behavior in a number of developing countries confirms that theory can be a flexible and valuable tool for HIV prevention worldwide.
In northern Thailand, for example, an application of the Health Belief Model and the Theory of Reasoned Action revealed that both models were useful for analyzing consistent condom use in commercial sex. The Theory of Reasoned Action, however, was more useful because of its more accurate incorporation of peer group effects on risky sexual behavior.6
- Research that helps target audiences identify solutions to their own problems can lead to extremely effective program development.
Such research is particularly useful for designing programs and policies to remove or overcome structural and environmental barriers to behavior change. One example is a pilot effort to adapt Thailand's "100 percent condom policy" in the Dominican Republic, which was designed based on formative research results from intensive discussions and in-depth interviews with sex workers and their employers and clients (Box 5.2). In Haiti, the local NGO Groupe de Lutte Anti-Sida (GLAS) used participatory action research with factory workers to continually adapt and improve its workplace prevention programs.7
Recommendations
- Research for HIV/AIDS prevention should not focus exclusively on behavioral issues. Understanding sexual behavior and identifying ways to influence that behavior requires a multidisciplinary perspective with contributions from diverse fields such as STD management and prevention, social marketing, medicine, counseling, psychology, epidemiology, communications and family planning.
- International organizations and donor-funded programs should foster mutually beneficial partnerships between local research institutions and the NGOs that implement HIV/AIDS interventions. Such collaboration builds important connections between research and interventions and offers a sustainable source of technical assistance to the NGOs.
- Large-scale, multisite efficacy trials should be reserved for tests of interventions that have not been rigorously evaluated and that have (1) global significance, (2) important policy implications and (3) complex intervention components.
- HIV/AIDS programs should put more emphasis on rapid research that provides the information needed to improve interventions or to adapt successful interventions for application in different geographical areas or with new populations.
- Proposals for behavioral research should be reviewed and approved by an ethical review committee whose members are thoroughly familiar with the customs and traditions of the community in which the research is being conducted. A local review board is usually best placed to provide this perspective.
- HIV/AIDS researchers should expand the use of formal behavioral theories of HIV risk behavior. Even though these theories were developed in the United States, they have also proved useful for understanding risk behavior in developing countries.
- HIV/AIDS programs should support research that allows target audiences to propose solutions to their own problems. Such research is particularly useful for identifying ways to remove or overcome structural and environmental barriers to behavior change.
Future Challenges
Assessing Biological Data
There is growing interest in the use of biological data, such as sexually transmitted infection and HIV serostatus data, as proxy measures of risk behavior. Data on self-reported behavior may be biased as a result of poor recall and the social stigma associated with risk behaviors. But collection of biological data brings its own unique set of problems, including the social and psychological impact of receiving positive HIV and STD results, the need to provide STD treatment, misclassification bias due to the limits of STD diagnostic capacities, and the high cost of biological testing and associated counseling and treatment. Research is needed on the combined use of biologic and behavioral data to assess HIV/AIDS interventions.
Examining Care and Support
People living with HIV/AIDS experience severe social and psychological stress, particularly soon after learning that they are infected. The results of some studies suggest that people who are infected with the virus are more likely to infect others when there are few social and psychological services available to assist them in coping with their HIV status. But as the numbers of people with HIV and AIDS increase, more research is needed to identify support services that encourage preventive behavior.
Understanding Social Change
Most HIV/AIDS research examines sexual behavior and behavior change among individuals. While this research and the interventions it generates have demonstrated some success in changing individual behavior, there is an urgent need to focus on interventions that influence social norms and other factors beyond the individual. Developing tools to measure such change at different levels of social organization and testing those tools should be a research priority for HIV/AIDS programs.
References
- Celentano D, Nelson K, Suprasert S, et al. (1996). Risk factors for HIV-1 seroconversion among young men in Thailand. Journal of the American Medical Association 275(2):122-127.
- De Zoysa I, Phillips KA, Kamenga MC, et al. (1995). Role of HIV counseling and testing in changing risk behavior in developing countries. AIDS 9(suppl A): S95-S101.
- Sweat MD, Denison JA (1995). Reducing HIV incidence in developing countries with structural and environmental interventions. AIDS 9(suppl A):S251-S257.
- DeZoysa I, Sweat M, Denison J (1996). Faithful but fearful: reducing HIV transmission in stable relationships. AIDS 10(suppl A):S197-S203.
- MacNeil J, White RA (1995). HIV care: moving from rationale to research. AIDSCAP/Family Health International, Arlington, VA.
- VanLandingham M, Supraset S, Grandjean N and Sittritai W (1995). Two views of risky sexual practices among northern Thai males: The Health Belief Model and the Theory of Reasoned Action. Journal of Health and Social Behavior 36(March):195-212.
- Dadian MJ (1996). Turning obstacles into opportunities: Haitian workplace project breaks new ground in HIV/AIDS prevention. AIDScaptions 3(3):4-7.
