5
Primary HIV-Prevention Strategies

Over the past 10 years, African governments—through national AIDS control programs—international development agencies, private voluntary organizations, and other nongovernmental groups across Africa have devoted resources, time, and energy to developing low-cost interventions to arrest the spread of HIV and AIDS. Many different programs have distributed AIDS leaflets, badges, stickers, and other paraphernalia. Messages informing people about the danger of AIDS are regularly broadcast on radio and television, published in newspapers, displayed on billboards, and performed by local entertainers. Hundreds of peer educators across the continent visit local bars, beer gardens, hotels, STD clinics, and work sites to provide AIDS-prevention education and distribute free condoms. Millions of other condoms are being made available at very low cost through social marketing programs. 1 At the start of the epidemic, when many of these interventions were first conceived, the hope was that they would induce a sufficiently large behavior response to contain the epidemic.

How successful have these efforts been at preventing new cases of HIV infection? Despite the many limitations inherent in attempting to evaluate the effectiveness of interventions aimed at HIV prevention, clear evidence is emerging that such efforts can be successful, particularly among higher-risk groups

1  

Social marketing is the application of commercial marketing techniques to achieve a social goal. Condom social marketing programs make condoms more accessible and affordable. At the same time, condom social marketing programs promote the use of condoms in an attempt to make them more acceptable to target populations.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences 5 Primary HIV-Prevention Strategies Over the past 10 years, African governments—through national AIDS control programs—international development agencies, private voluntary organizations, and other nongovernmental groups across Africa have devoted resources, time, and energy to developing low-cost interventions to arrest the spread of HIV and AIDS. Many different programs have distributed AIDS leaflets, badges, stickers, and other paraphernalia. Messages informing people about the danger of AIDS are regularly broadcast on radio and television, published in newspapers, displayed on billboards, and performed by local entertainers. Hundreds of peer educators across the continent visit local bars, beer gardens, hotels, STD clinics, and work sites to provide AIDS-prevention education and distribute free condoms. Millions of other condoms are being made available at very low cost through social marketing programs. 1 At the start of the epidemic, when many of these interventions were first conceived, the hope was that they would induce a sufficiently large behavior response to contain the epidemic. How successful have these efforts been at preventing new cases of HIV infection? Despite the many limitations inherent in attempting to evaluate the effectiveness of interventions aimed at HIV prevention, clear evidence is emerging that such efforts can be successful, particularly among higher-risk groups 1   Social marketing is the application of commercial marketing techniques to achieve a social goal. Condom social marketing programs make condoms more accessible and affordable. At the same time, condom social marketing programs promote the use of condoms in an attempt to make them more acceptable to target populations.

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences (World Health Organization, 1992c; Coates, 1993; Lamptey et al., 1993; Choi and Coates, 1994; Stryker et al., 1995). At the same time, however, data from various surveillance systems indicate that current interventions are probably not yet having a significant impact on the epidemic at the continent or even the country level (Lamptey et al., 1993; see also Chapter 3). Despite the fact that levels of AIDS awareness are extremely high across the continent (see Chapter 4), getting people to change their behavior is difficult. Denial, fear, external pressures, other priorities, or simple economics can sometimes keep people from adopting healthier life-styles. There are many reasons why prevention efforts in Africa have not had as large an impact on the spread of the epidemic as desired. AIDS has struck the continent at a time when it is undergoing its worst financial crisis since independence. In some countries, other catastrophes—such as wars, droughts, or famines—have been more immediate and taken precedence over AIDS-prevention efforts. Throughout the continent, the overall magnitude of the response has been inadequate, and expectations about what could be achieved quickly have been unrealistic. A lack of indigenous management capacity and chronic weaknesses in the public health system have hindered the development and implementation of AIDS control programs. Individuals and organizations working against the spread of AIDS have had to face discrimination, complacency, and even persistent denial in the community. Many AIDS workers have become exhausted after struggling for so long against impossible odds; many others have died (Mann et al., 1992). Myths surrounding modes of transmission hinder the dissemination of correct knowledge and sustained behavior change (see, for example, Krynen, 1994; Nature, 1993; Ndyetabura and Paalman, 1994; Ankomah, 1994). But getting people to change their behavior is not impossible. Indeed, health educators in Africa have had a fair amount of success in the recent past. For example, broad-based education campaigns have persuaded large numbers of people to have their children immunized against various childhood diseases and educated mothers to give their children oral rehydration formula during episodes of diarrhea. Of course, attempting to modify more personal behavior, such as sexual practices, is more challenging. Yet, family planning programs have been successful even in some of the most disadvantaged countries of the world (see, for example, Cleland et al., 1994). Even the most cautious reviews of behavioral interventions aimed at slowing the spread of HIV conclude that although most have not been rigorously evaluated, some approaches do seem to work (e.g., Oakley et al., 1995). It is important to have realistic expectations about what can and cannot be achieved. Behavior change will never be 100 percent effective: some individuals will never choose to protect themselves, while others will relapse into old patterns of behavior after just a short period of time (Cates and Hinman, 1992; Lamptey et al., 1993). To increase the likelihood of success, interventions need to be culturally

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences appropriate and locally relevant, reflecting cognizance of the social context within which they are embedded (see Chapter 2). They should be designed with a clear idea of the target population and the types of behaviors to be changed. In turn, impediments in the social environment to behavior change probably need to be removed or weakened (Turner et al., 1989). Therefore, behavior-change interventions should include promotion of lower-risk behavior, assistance in risk-reduction skills development, and promotion of changes in societal norms (Lamptey, 1994). In Africa, as elsewhere, HIV-prevention messages have included promotion of partner reduction, postponing of sexual debut, alternatives to risky sex, mutually faithful monogamy, consistent and proper use of condoms, better recognition of STD symptoms, and more effective health-seeking behavior. The purpose of the discussion in this chapter is to delineate opportunities for effecting beneficial behavior changes and to discuss how these opportunities might be realized. The discussion is based on an examination of interventions to achieve behavior change, an effort that has led to the development of a set of basic principles for successful strategies and programs. The remainder of this chapter is organized as follows. First we examine principles and issues in the design and evaluation of behavior-change intervention programs. We then examine the issues that challenge the design of effective interventions targeted to African men, women, and youth, respectively, and highlight strategies that have been implemented to address these issues. Each discussion is followed by an illustrative case study. Lessons learned from these programs are then highlighted. Next follows a discussion of strategies to prevent perinatal HIV transmission. The chapter ends with a set of recommendations for prevention research, which are made in light of the experiences of strategies and programs implemented in Africa to date. BEHAVIOR-CHANGE INTERVENTION PROGRAM DESIGN AND EVALUATION The rapid spread of HIV in sub-Saharan Africa since the early 1980s is a result of a multiplicity of factors, many of which have been discussed in previous chapters. Table 5-1 summarizes much of this information. As the table shows, it is analytically convenient to distinguish among four types of factors: (1) individual-level factors (i.e., ones the individual has some control over changing), (2) societal factors that may serve to encourage or discourage high-risk behavior, (3) health infrastructural factors that directly or indirectly facilitate the spread of HIV, and (4) structural factors related to development issues over which the individual has very little control. As the last column in Table 5-1 shows, each of these sets of factors requires a different length of time to bring about positive change. A basic comprehensive HIV-prevention program should aim to address the

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences TABLE 5-1 Factors Contributing to Sexual Transmission of HIV in sub-Saharan Africa Level Definition Examples Changes Required Comments Individual Factors that directly affect the individual and that the individual has some control in changing Biological: • History and/or presence of STDs; • Lack of male circumcision; • Anal intercourse; • Sex during menses; • Traumatic sex; • Cervical ectopy • Prevention/treatment of STDs; • Avoidance of sex during menses; • Prevention of traumatic sex Achievable in the short term     Behavioral: • Frequent change of sex partners; • Multiple sex partners; • Unprotected sexual intercourse; • Sex with a commercial sex worker; • Sex with an infected partner; • Lack of knowledge of STDs/HIV; • Low risk perception • Abstinence; • Mutual fidelity; • Consistent condom use; • Knowledge and skills of STD/HIV prevention Achievable in the short term

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences Societal Factors related to societal norms that encourage high-risk sexual behavior • High rates of prostitution; • Multiple partners by men; • Gender discrimination; • Poor attitudes toward condom use; • Social status of women; • Extended postpartum abstinence • Improvement in status of women; • Job opportunities for women; • Promotion of mutual fidelity; • Changes in societal attitudes toward condom use Achievable in the short to medium term Infrastructural Factors that directly or indirectly facilitate the spread of HIV, over which the individual has very little control • Poor availability of condoms; • Poor STD services; • High prevalence of STDs; • Poor communication services • Changes in health infrastructure; • Improvement in STD care, behavior-change communication, and condom provision Achievable in the short to medium term Structural Factors related to developmental issues, over which both the individual and the health system have very little control • Underdevelopment; • Poverty; • Rural/urban migration; • Civil unrest; • Low literacy rates for women; • Laws/policies, including lack of human rights; • Unemployment • General economic development programs; • Enactment of appropriate laws/policies; • Income-generating opportunities; • Improvement in education of women Feasible in the long term

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences four types of factors cited above as contributing to the spread of HIV, particularly those related to health. Steps in a comprehensive strategy include epidemiological and behavioral formative research to help design programs; implementation of behavioral interventions to facilitate individual, group, and societal behavior change; condom provision; structural/environmental strategies that support individual, group, and societal change; and program evaluation to assess success in program implementation, intermediate outcomes, and ultimate impact. In addition, evidence that STDs may facilitate HIV transmission and the interconnectedness of STD and HIV transmission modalities and prevention (see Chapter 3) strongly suggest an important role for STD treatment, control, and prevention in AIDS prevention (Laga et al., 1991; Grosskurth et al., 1995). Guiding Principles for Behavior-Change Interventions Interventions designed to modify people's behavior need to be based on sound principles of behavior change. Many behavioral theories are described in the literature and have been applied to understanding HIV risk behavior.2 However, no single theory sufficiently explains individual behavior changes or provides all the essential tools to change behavior (Coates, 1993). As a result, current thinking calls for a complementary combination of theoretical approaches that incorporates the key principles of behavior change into program design. Seven guiding principles for effective behavior change interventions targeted at HIV/AIDS prevention have evolved. These principles, which are consistent with the behavior science literature and with experience in program development in Africa (Family Health International/AIDSCAP Project, 1995), are as follows: Targeting—Interventions should focus on well-characterized, specific target audiences. Skills development—Interventions should include components that encourage individual acquisition of skills and tools that will help to prevent the transmission of HIV. Support—A supportive social environment needs to be created to foster HIV-prevention interventions and reinforce individual behavior-change efforts. Maintenance—HIV-prevention interventions need to include strategies that will foster the maintenance of behavior change over time. Collaboration—Every effort should be made in the development and delivery 2    These include the Health Belief Model, Social Learning/Social Cognitive Theory (including the self-efficacy construct), the Stages of Change Model, the Theory of Reasoned Action (including the behavioral intention construct), Motivation/Protection Theory, Social Inoculation Theory, Cognitive Behavior Modification Theory, the Harm Reduction Model, and the AIDS Risk Reduction Model.

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences of prevention interventions to work collaboratively with other sectors, ministries, and communities so that the potential for synergistic program effects is enhanced. Monitoring and evaluation—Programs must be monitored and evaluated in order to determine intervention implementation integrity, effectiveness, and cost-effectiveness. Sustainability—Because resources are limited and donor support is intermittent, HIV-prevention programs should be designed for sustainability by building capacity to pursue alternative resources. Evaluation is discussed in detail in the next section. Here we comment further only on the principles of sustainability and of targeting. For HIV/AIDS-prevention programs to be effective and sustainable, it is critical to build local capacity in both the public and the private sectors. The chronic shortages of human and fiscal resources in most of sub-Saharan Africa often make program design, implementation, and management difficult. Major deficiencies include poor infrastructure and paucity of management and technical skills in both research and intervention programs (see also Chapter 7). Special efforts should be made to improve the technical, organizational, management, and financial skills of individuals, as well as to strengthen institutional infrastructure. With regard to targeting and audience segmentation in HIV intervention programs, the primary objective is to obtain effective and rapid results by intervening with groups that are at the greatest risk of acquiring and spreading HIV infection (Lamptey and Potts, 1990). Although it would be highly desirable to design prevention programs that are based on a good understanding of the target population and the sociocultural, environmental, and structural context, few prevention programs are based on preliminary, or formative, research findings. As noted earlier, the groups most commonly targeted are those traditionally seen as at particularly high risk, such as truckers, uniformed service workers, commercial sex workers, or STD patients. Targeting can also extend to other at-risk populations, including out-of-school youth, school children, university students, male and female factory workers, women in the general population, and men away from home. The target audience may be defined by (1) epidemiological risk factors (e.g., HIV prevalence among commercial sex workers); (2) behavioral risk factors (e.g., clients of commercial sex workers); (3) occupation (e.g., factory workers); (4) geographic location (e.g., urban adults); (5) access points (e.g., truck stops); (6) demographics (e.g., gender, age); or (7) relevant sociocultural factors (e.g., widows, street children). In later sections of this chapter, interventions targeted principally to men, to women, and to youth are discussed; although restricting our consideration to these particular groups, we recognize that all interventions should include some attention to the sexual partners of the target group.

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences INTERVENTION EVALUATION Just as the stages of an intervention program can be described as design, implementation, and outcome, so can evaluation accompany or follow each of these stages to determine how successfully the program has met its stated objectives. The corresponding stages or types of evaluation are formative, process, and outcome (Coyle et al., 1991). Formative evaluation of the design of an intervention addresses the question of whether the proposed intervention has the potential to achieve its goals. It considers structural factors that may affect the program. For example, are policies in place to facilitate implementation of the intervention strategies (e.g., a policy requiring sex education in the schools that includes HIV/AIDS prevention)? Formative evaluation also assesses the content, nature, and design of the proposed intervention strategies in terms of success potential. For example, are they skills-based? Are they derived from preliminary, formative research? Do they match the target audience's needs? Are they built on the characteristics of other successful programs? Process evaluation techniques are used to monitor the program as it is being implemented. The goal is to determine how well short-term program objectives are being met. For example, are condoms being distributed? Are people tested for HIV returning to obtain their results? Is the number of patients attending STD clinics declining? Outcome evaluation is the last and ultimately the most important stage. It addresses questions about the success of the intervention. For example, has the program succeeded in changing behavior? Has that behavior change been sustained? And finally, has that behavior change succeeded in reducing the incidence of HIV? This assessment of behavioral outcomes should be performed only when there is reason to expect that such outcomes could have occurred and can be measured (i.e., earlier formative evaluation has confirmed that the intervention was well designed, process evaluation has corroborated that the intervention is being carried out well, the observation period has been substantial, behaviors have been measured, and the sample size has been sufficiently large to detect behavioral differences). Good outcome evaluation is rarely feasible unless it is planned as an integral part of the intervention and usually is possible only in the context of studies with adequate research funding and undertaken by research institutions such as universities. Interventions evaluated in this way can serve as models for others that will not undergo as rigorous an outcome evaluation. While the logical program evaluation sequence outlined above is time-consuming, expensive, and rigorous, following these steps can help avoid wasting scarce funds on evaluations that are limited in their ability to yield useful results and repeating or continuing intervention programs that are limited in their ability to achieve the desired goals (Oakley et al., 1995). For Africa, as elsewhere, we have limited information on what works well

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences because of the difficulties in carrying out rigorous evaluations, particularly outcome evaluations (Oakley et al., 1995). While the goal of rigorous evaluation must be maintained over the longer term, there is also demand for a faster, more pragmatic approach. Insistence on pure scientific rigor may be counterproductive for HIV/AIDS-prevention interventions, where innovative methods and program designs are urgently needed, and prevention efforts must rely almost exclusively on behavior change (Coyle et al., 1991). Examples of programs that have been implemented in sub-Saharan Africa and reported in studies are described briefly in Annex 5-1 and referred to in subsequent sections of this chapter. This is not a comprehensive listing; rather it illustrates recent interventions that have used a variety of strategies and programs in trying to effect positive changes in the behavior of targeted audiences. Many have attempted to conduct modest evaluations with the limited funds available to them. Some of these evaluations suffer from methodological weaknesses so that the results must be interpreted with caution; yet all contribute to the growing effort to evaluate programs, with the twin goals of improving continuing intervention programs and designing new programs that are more effective in achieving results and at lower cost. Even where the ''gold standard" of an intervention that consists of a randomized trial with high-caliber formative, process, and outcome evaluation cannot be met, there are practical guidelines for program developers. First, at the formative stage, it should be recognized that programs are most likely to achieve desired outcomes when their design is based on the characteristics and principles of existing effective programs, past program experience, formative research, and sound theoretical underpinnings. Even when existing programs have not undergone a gold standard evaluation, if there is consistency and convergence in results from multiple programs in different places and/or over time and/or from multiple measures of results, synthesis and interpretation of findings can guide the design of new programs. Planners must be cautious in applying such findings, maintaining awareness of the potential for biases that can influence the results. Second, while rigorous outcome evaluations are desperately needed, not all programs should attempt to implement them. However, all programs should be required to conduct process evaluation to assess program integrity. Finally, evaluation results should be translated and synthesized into new programmatic guidelines and materials and new training activities so that programs can be improved. A TYPOLOGY OF INTERVENTION PROGRAMS HIV-prevention strategies can be classified according to the foundation on which they are based: (1) formal institution-based programs (e.g., at the workplace, school, or clinic); (2) community-based programs (e.g., among informal youth groups or informal women's groups); and (3) population-based programs (e.g., national media campaigns or policy development). These categories are not

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences mutually exclusive. For example, the same individuals may be reached at work, in special community-based initiatives, or at home through the mass media. The following sections describe each of these program types in greater detail and illustrate them with studies included in Annex 5-1. Although, as already mentioned, most of these studies lacked the resources to carry out definitive evaluations, we summarize the investigators' reports of quantifiable impacts. Institution-based Programs Both institution-based and community-based programs are designed to reach individuals and small groups, with the aim of teaching and reinforcing protective behaviors. They are intended to give individuals the opportunity to acquire information, assess their own risk of HIV, interact with a provider, and obtain relevant behavioral and communication skills that can help in reducing high-risk behavior; they also generate notions of peer norms that are conducive to risk reduction (Lamptey and Coates, 1994). Institution-based programs include interventions in the workplace (factories, prisons, commercial farms, mining communities, military bases); in schools; and in health facilities, such as STD and family planning clinics, hospitals, and HIV counseling and testing clinics (see studies [4], [5], [7], [8], [10-12], [14-16], [18], and [20-24] in Annex 5-1). Targeted groups within these institutions may not be homogeneous in terms of individual behavior or social norms, and may or may not exhibit more high-risk sexual behavior than the general population. The relative ease of access to these institutional populations renders such programs attractive and potentially cost-effective. Several institution-based programs in Africa have demonstrated changes in risk behavior (Ngugi et al., 1988 [4]3; Loodts and Van de Perre, 1989; Kamenga et al., 1991; Allen et al., 1992b [7]; Mwizarubi et al., 1992 [12]; Williams and Ray, 1993 [20-24]; Wynendaele et al., 1995). In Rwanda, for example, the initiation of an education, STD treatment, and condom distribution program for all military recruits led to a reduction in the incidence of urethritis (Loodts and Van de Perre, 1989; World Health Organization, 1992c [11]). In Rwanda, 1,458 women attending antenatal and pediatric clinics at the Centre Hospitalier de Kigali received pre-and post-test counseling, were shown an educational video, and were given free condoms. One year later, HIV seroconversion rates had decreased significantly (from 4.1 to 1.8 per 100 person-years among women whose partners were tested and counseled) (World Health Organization, 1992c [10]). Preliminary results from other ongoing programs in Tanzania and Zimbabwe also are encouraging (Mwizarubi et al., 1992 [12]; Williams and Ray, 1993 [20-24]; World Health Organization, 1992c [14]). In Tanzania, a workplace 3    Numbers in brackets immediately following certain references are the numbers assigned to the studies in Annex 5-1.

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences intervention targets truck drivers, their assistants, and their sexual partners (Mwizarubi et al., 1992 [12]). In Zimbabwe, a workplace-based intervention program targets 4,500 factory employees and their families (Williams and Ray, 1993 [20-24]). This program provides STD treatment, behavior-change intervention, and condoms. The intervention employs a combination of drama, printed materials, group talks, and interpersonal counseling by 64 peer educators. Although evidence about the impact of counseling and testing on risk behavior is mixed (Higgins et al., 1991), a few evaluations conducted to date in developing countries suggest that such efforts may be effective, at least in the short term, in changing sexual practices under certain circumstances or among certain groups (such as couples) (Higgins et al., 1991; Kamenga et al., 1991; Allen et al., 1992a [7], 1992b [8]; Wynendaele et al., 1995 [5]). In Rwanda, for example, results of an HIV counseling and testing program demonstrated an increase in condom use (Allen et al., 1992b [8]). HIV seroconversion rates were also reported to have decreased among seronegative women whose partners were tested, but not among women whose partners were not tested. In Uganda, an evaluation of a counseling and testing intervention found significant reported decreases in risk behaviors among both seronegative and seropositive individuals (Moore et al., 1993).4 Community-based Programs Community-based programs use group interventions to reach communities. These interventions include the use of peer educators to reach sex workers or school-aged youth, use of traditional health providers to reach rural communities, or programs targeted to other community groups (see studies [1], [6], [9], [13], [25], and [26] in Annex 5-1). Community participation can sometimes be a critical factor for program success and sustainability (Lamptey and Coates, 1994; Population Council, 1995). According to the guiding principles described earlier, HIV/AIDS-prevention programs should be designed at the outset with attention to their acceptability within the community and target groups, the external or institutional support required to develop and sustain the skills and talent needed to make them work, and the infrastructure support and the individual and collective commitments needed to maintain them over time (Lamptey and Coates, 1994). Community involvement is often only an empty slogan in programs without any real involvement of the community in decision making. 4    In 1994, Family Health International/AIDSCAP and WHO/GPA initiated a multisite clinical trial in which those seeking HIV counseling and testing were randomly assigned to receive either counseling and testing or health information (De Zoysa et al., 1995). It is hoped that rigorous evaluation of this project will resolve outstanding questions about the impact of HIV counseling and testing on the frequency of risk behaviors,

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences and mortality among young children in developing countries concluded that "vitamin A supplementation can effectively reduce mortality rates in young children, and probably also reduce the risk of severe morbidity" (Beaton et al., 1993:66). Although the exact role of vitamin A remains unclear, further research into its effectiveness in inhibiting perinatal transmission and reducing the mortality and morbidity of HIV-infected children is clearly warranted. RECOMMENDATIONS Numerous interventions are being implemented throughout Africa, but most are still information-based health education campaigns. Many of the messages communicated are generic or vague and do not address specific risk behaviors. Innovative approaches are typically small scale and lack rigorous evaluation. Furthermore, it is not easy to demonstrate the success of a particular intervention because it is difficult to define and measure such outcome variables as "better health status" and to determine whether the intervention in question was the reason for a desired change. Consequently, the need for solid evaluation research is still urgent. KEY RECOMMENDATION 2. An increase in research funding for the development of social and behavioral interventions aimed at protecting women and adolescents, especially girls, from infection deserves highest priority. An important step in arresting the spread of AIDS in sub-Saharan Africa is to recognize that, although African women have relatively high autonomy by the standards of developing countries, their low and separate status remains a major obstacle to HIV prevention. In many societies, the presence of unmarried, postpubertal girls is a new phenomenon. Guidelines for their sexual behavior and that of others toward them are not well established; their low social status makes them particularly vulnerable. Moreover, in many areas of sub-Saharan Africa, high HIV incidence has been detected among adolescents and young adults, especially girls. Research on which to design culturally relevant programs targeted to adolescents and to adults who might be their sexual partners is an important priority. KEY RECOMMENDATION 3. More evaluation research is needed to correlate process and outcome indicators—such as reported condom sales and behavior change—with reductions in HIV incidence or prevalence.

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences Rigorous designs, such as controlled intervention studies to assess the effectiveness of different prevention approaches, are needed. To date, few rigorous evaluations of intervention programs in sub-Saharan Africa have been conducted (see reviews in Choi and Coates, 1994, and Crump, 1995). Evaluations that have been reported often lack precision in their measurement of risk behaviors and are therefore not very informative. As a result, few strategies can demonstrate whether they are effective. Barriers to rigorous evaluation research include lack of human resources, expertise, financial resources, and equipment. Overcoming these barriers requires major changes in research infrastructure. Nevertheless, it is a priority to begin now a few large-scale behavioral interventions, including adequate baseline surveys, multiround surveys, and longitudinal studies with comparison cohorts, even if these interventions are relatively expensive. It is only with these types of studies that more definitive information on the effectiveness of various interventions, which is so desperately lacking for most studies in sub-Saharan Africa, can be obtained. The longer such studies are delayed, the longer will exist the uncertainty about which HIV-prevention strategies work best, for whom, and under what circumstances. In the interim, basic program evaluation and some formative and operational research can be completed, and such work should be required by donors as part of program implementation awards. Recommendation 5-1. Interventions that promote gender equality deserve high priority as AIDS-prevention strategies in every country. Women's primary source of risk is their society-wide subordination, not their lack of knowledge (Heise and Elias, 1995). Governments can effect change in many ways to empower women: reducing the financial necessity for multiple partnerships by changing laws to give women equal access to training and jobs, equal rights of inheritance and property ownership, equal access to education, and equal wage scales; enacting and enforcing laws against rape; building the capacity of women for collective action; and educating everyone about women's rights. Enhancing the status of women is a long-term strategy that would have many beneficial effects for development, in addition to the likely effect of reducing the transmission of HIV and other STDs. Recommendation 5-2. In the short term, a female-controlled vaginal microbicide that would allow women to protect themselves without their partner's participation is an urgent research and development priority for international donors. A microbicide is not a quick-fix substitute for the fundamental structural reforms necessary to achieve gender equality, but rather a temporary and partial response to this problem as it influences HIV transmission (Elias and Heise,

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences 1994; Heise and Elias, 1995). Yet in the same way that the use of spermicides by women can reduce fertility, the use of a microbicide could, in and of itself, help arrest the spread of HIV. Recommendation 5-3. Research is needed to address the HIV-prevention needs of several other populations with marked vulnerability, particularly the mobile and the disenfranchised. There is a need to reach mobile individuals and groups with comprehensible and acceptable programs, particularly where linguistic and cultural barriers exist between migrants and the local population. Ways of effectively providing preventive services to the disenfranchised populations in the ever-growing urban slums and in refugee camps need to be developed; a major challenge to such programs is the lack of resources and social support for individuals in such settings. Recommendation 5-4. Additional research should be conducted to determine the impact of specific STD interventions on the incidence of HIV infection within defined populations. Research is needed to determine the extent to which STDs help cause HIV infection, to examine the importance of the behavioral synergy of STD and HIV transmission, and to design more effective intervention programs. There is a need for assessment of the relative efficacy and feasibility of various interventions for STD treatment and sexual behavior change in reducing HIV transmission. This research includes assessing the effects of programs that target individuals at high risk of acquiring and transmitting STDs, as well as the effects of community-based STD programs. The interventions themselves could comprise STD education, condom distribution, increased STD screening, and mass antibiotic therapy. Data on the effectiveness of these interventions, particularly those focused on decreasing STD prevalence, are essential for evaluating the impact of STD reduction on the spread of HIV. Behavioral research on ways of ensuring acceptance of various STD control strategies should be directly integrated into the epidemiological research. Recommendation 5-5. Research is needed to assess the effectiveness and cost-effectiveness of the syndromic approach to STD diagnosis and treatment. Clinical testing for STDs is expensive and not widely accessible. Therefore, research is needed on better ways to identify STDs more accurately through symptoms. In addition, new screening methods, including urine-based assays for chlamydia and gonorrhea and self-administered vaginal swabs for trichomonas

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences culture and bacterial vaginosis gram stain, should be incorporated into research. Efforts are needed to make these techniques available and affordable in developing-country settings for surveillance, diagnosis, and validation. Recommendation 5-6. For long-term program planning and resource allocation, cost-effectiveness studies should be incorporated in donor research work and the cost-effectiveness of HIV prevention compared with that of other health interventions. Few intervention evaluations have adequately assessed effectiveness in terms of behavior change or seroincidence declines, much less cost-effectiveness. Results of evaluation studies currently in progress in several countries in sub-Saharan Africa (Family Health International/AIDSCAP, 1994) are expected to provide data on the cost-effectiveness of various HIV-prevention strategies. However, determining the effectiveness of HIV-prevention strategies is methodologically complex and will take several more years to complete. In the meantime, since resources are insufficient and may well decline further, efficient resource utilization is paramount. Thus, basic analysis of overall program costs and specific intervention costs is critical. Simple cost analyses and cost-effectiveness estimates could provide data that would be helpful for public health decision making and program design. Recommendation 5-7. Operations research should be a high priority. The growth of the HIV/AIDS pandemic in the past 20 years in sub-Saharan Africa has led to the development of institutional and community-based responses and a corresponding need for operations research to improve the effectiveness, cost-effectiveness, and quality of these responses. Primary research needs include scaling up successful experimental interventions, improving the effectiveness and reducing the cost of existing programs, examining the cost-effectiveness of linking HIV prevention with HIV/AIDS care, and improving the sensitivity and specificity of criteria for targeting interventions. Recommendation 5-8. Research should be undertaken to measure the impact of female-controlled barrier contraceptive use on HIV transmission. Studies should be undertaken to determine the effectiveness against STDs and HIV of female-controlled barrier contraceptives such as female condoms and spermicides. This research should encompass field-based studies of the acceptability of these methods. Moreover, greater efforts need to be made to integrate appropriate HIV/AIDS-prevention messages and programs for STD diagnosis, referral, and treatment into family planning programs.

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences Recommendation 5-9. Behavioral research is needed to develop effective pregnancy-related HIV counseling programs. Given the rapid spread of HIV among women in sub-Saharan Africa, perinatal transmission continues to have a major impact on infant and child morbidity and mortality among populations with a high HIV seroprevalence. Studies using modified treatment regimens with Zidovudine (AZT), hyperimmune gamma globulin, vitamin A, vaginal washes, and other means of intervention should be undertaken to determine their overall effectiveness and cost-effectiveness in decreasing HIV perinatal transmission.

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences ANNEX 5-1 SELECTED AIDS INTERVENTION PROGRAMS IN AFRICA Country (area) Intervention Type Target Population and Size Program Components Key Results Reference 1. Ghana Community-based Commercial sex workers • Behavior-change interventions; • Condom provision • Consistent condom use among sex workers rose from 6 to 71% in 6 months.; • After the intervention, 64% of those followed up reported always using condoms with clients. Asamoah-Adu et al., 1994 2. Ghana Population-based Ghanaian youth • Behavior-change interventions • The proportion of sexually active 15-year-olds fell from 44 to 27%.; • Men reported an 18% decrease in sexual partners over 3 months. Family Health International/AIDSCAP, 1992 3. Guinea Population-based 500,000 couples in Guinea • Behavior-change interventions; • Condom provision; • Family planning • In the first 6 months, 2.3 million condoms were distributed. Hess, 1993 4. Kenya Institution-based Female prostitutes and women attending clinics • Behavior-change interventions; • HIV testing and counseling; • Free condom provision • The proportion of counseled women who use condoms increased from 10% at pre-test to 81% at post-test. Ngugi et al., 1988 5. Malawi Institution-based (hospital) General population (mostly married males) • HIV/STD counseling • STD treatment and control significantly improved (88% of those counseled were cured; 77% of those in the control still presented with STDs). Wynendaele et al., 1995 6. Nigeria Community-based Commercial sex workers • Behavior-change interventions; • Condom provision; • STD services; • Vocational and literacy programs • Post-counseling, the reported ever use of condoms increased from 41 to 71%. Williams et al., 1992 7. Rwanda (Kigali) Institution-based (clinics) 53 HIV-discordant couples • HIV testing and counseling; • Free condom provision • Within the first year, the percentage of women never using condoms fell from 25 to 3%.; • Post-test, >60% reported condom use in their most recent sexual intercourse act.; • Post-test, "always" condom use doubled, from 12 to 24%.; • Condom use was associated with a lower rate of new HIV infections.; • Discordant couples using condoms increased from 4 to 57% after one year. Allen et al., 1992b 8. Rwanda (Kigali) Institution-based (clinics) 1,458 childbearing women in Kigali • HIV testing and counseling; • Free condom provision; • Free spermicide provision; • Behavior-change communications (videotape) • Condom use increased from 7 to 22% of women.; • HIV seroconversion rates decreased significantly among women whose partners were tested and counseled. Allen et al., 1992a 9. Rwanda Community-based 5 rural communes, 100 km from Kigali • Behavior-change interventions; • Condom provision • Condom use increased from 4 to 44%.; • Reported ever use of condoms increased from 7 to 50%. Mercer et al., 1993

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences This page in the original is blank.

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences Country (area) Intervention Type Target Population and Size Program Components Key Results Reference 10. Rwanda Institution-based (clinics) 1,500 attending clinics in Kigali and their partners • Behavior-change interventions; • Condom provision; • Free HIV testing and counseling • Consistent condom use rose from 7 to 22% in 1 year.; • Prevalence of gonorrhea decreased from 13 to 6% among HIV-positive women.; • HIV seroconversion rates decreased from 4.1 to 1.8 per 100 person-years among women whose partners were tested and counseled. World Health Organization, 1992c 11. Rwanda Institution-based (workplace) Military recruits • Behavior-change interventions; • Condom provision • Incidence of urethritis fell from 12 to 5% in 16 months. World Health Organization, 1992c 12. Tanzania Institution-based (workplace) Truck drivers and their partners • Behavior-change interventions; • Condom provision • Condoms distributed increased from 60,000 to 700,000 in 6 months.; • Condom use among women rose from 50 to 91%.; • Condom use among men rose from 54 to 74%. Mwizarubi et al., 1992 13. Uganda (Rakai) Community-based Young adults aged 13-39 in rural communities • Condom promotion and provision by trained peers and community health workers • Reported ever use of condoms increased from 7.6% in 1990 to 12.5% in 1992. Konde-Lule et al., 1994 14. Uganda Institution-based (workplace) The general population; 400,000 workers • Behavior-change interventions; • Condom provision • Condom use rose from 3.5% in 1990 to 14.5% in 1991. World Health Organization, 1992c; McCombie and Hornik, 1992 15. Zaire (Kinshasa) Institution-based (textile factory and commercial bank) 149 HIV-discordant couples • HIV testing and counseling; • Free condom provision • Condom use increased from <5 % to 71%, 60%, and 77% at 1, 6, and 18 months, respectively. Kamenga et al., 1991 16. Zaire Institution-based (workplace) Commercial sex workers in Kinshasa • Behavior-change interventions; • STD treatment; • Condom provisions; • Counseling and testing • Condom use among women rose from 8 to 60%.; • Annual HIV incidence fell from 18 to 3% in 2 years. Laga et al., 1994 17. Zaire Population-based 13 million urban youth and their parents • Behavior-change interventions; • Condom provision • Condom sales rose from 100,000 in 1987 to 18 million in 4 years.; • Fidelity rose from 29 to 46% in 6 months. Population Services International, 1994d 18. Zambia (Copperbelt) Institution-based (workplace, school) 2 million people • Behavior-change interventions; • STD treatment; • Policy component; • Survival skills workshop • Three STDs treated at four clinics declined by 54, 53, and 42%, respectively. Mouli, 1992 19. Zambia Population-based General population • Behavior-change interventions; • Condom provision • In the first year of the program, 4.6 million condoms were sold.; • Among condom users, 50% indicated they first used a condom within the months following PSI's condom launch. Population Services International, 1994d

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences This page in the original is blank.

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences Country (area) Intervention Type Target Population and Size Program Components Key Results Reference 20. Zimbabwe Institution-based (workplace) 4,500 factory, workers, families, and local community • Behavior-change interventions; • STD treatment; • Condom provision • There was a decrease in the number of STD patients in 3 years.; • There were 37,000 condoms a year distributed. Williams and Ray, 1993 21. Zimbabwe (Marondera) Institution-based (workplace) Men, women, and youth • Behavior-change interventions; • Condom provision; • Income generation for commercial sex workers • Condoms distributed rose from 385,000 in 1991 to 962,400 in 1993.; • There was an 80% decrease in STD treatment. Williams and Ray, 1993 22. Zimbabwe (Eastern Highlands) Institution-based (workplace) 15,000 male and female plantation workers and their dependents • Behavior-change interventions; • Condom provision; • STD treatment; • Home care & management • There were 3,600 condoms/month distributed.; • There was a 59% decline in the number of patients treated for STDs. Williams and Ray, 1993 23. Zimbabwe (Rio Tinto) Institution-based (workplace) 3,500 miners and their families • Behavior-change interventions; • Condom provision; • STD treatment • Condoms distributed rose from 500/year to 56,000/year after 2 years.; • STD patients at four clinics declined between 47 and 78%. Williams and Ray, 1993 24. Zimbabwe (Mutare) Institution-based (workplace) 150,000 men, women and youth • Behavior-change interventions; • Condom provision • The incidence of STDs decreased by 48% in 2 years.; • There were 2.5 million condoms distributed in 1992.; • The program was expanded to 10 additional workplaces. Williams and Ray, 1993 25. Zimbabwe (Bulawayo) Community-based General population and high-risk groups • Behavior-change interventions; • Condom provision; • Income generation • There were 2.5 million condoms distributed in 2 years; • Consistent condom use in commercial sex workers rose from 5% in 1989 to nearly 50% in 1992.; • Positive attitudes toward condoms and expressed confidence in correct condom use increased from 35 to 76%.; • Those reporting "always" use of condoms increased from 28 to 50%. Wilson et al., 1992 26. Zimbabwe (Gabarone) Community-based Women in the community • Behavior-change interventions; • Communication and negotiation skills workshop   World Health Organization, 1992c

OCR for page 155
Preventing and Mitigating AIDS in Sub-Saharan Africa: Research and Data Priorities for the Social and Behavioral Sciences This page in the original is blank.