understanding cultural obstacles

32 downloads 0 Views 711KB Size Report
both diseases are sexually transmitted and tend-though only initially, with HIV/. Jealous husband. •. •. •. •. Wife of jealous husband ~ ~ ~ ~ ~ ~. Male lover hexed.
AIDS Education and Prevention, © 1994, The Guilford Press

6(1) 81-89,

1994

UNDERSTANDING CULTURAL OBSTACLES TO HIY/AIDS PREVENTION IN AFRICA Sally

J. Scott, MA, and Mary Anne Mercer, DrPH

Global efforts to prevent further spread of the human immunodeficiency virus (HN) are faced with tremendous obstacles, and limited understanding of the cultural and social context of HN and AIDS is an important factor hindering prevention efforts to date. Examples from the authors' experience in providing technical support to 7 nongovernmental projects for HN/AIDS prevention in Africa illustrate the importance of including qualitative data in initial baseline studies for projects responding to the problem. Qualitative data are needed to provide a deeper understanding of the meanings of behavior and other phenomena that are identified through quantitative methods. Runyoka is discussed as an example of a traditionally-defined illness in Zimbabwe that appears to affect local understanding of HN/AIDS. The authors suggest approaches that would assist HIV/AIDS projects to better understand and respond to the social and cultural context of AIDS in local settings.

Well over a decade into the human immunodeficiency virus (HIV) epidemic, HN/AIDS education and prevention efforts in Mrica have thus far produced limited results, and face tremendous obstacles. While rates of seroconversion appear to have slowed in at least one city initially hard-hit by the epidemic-Kinshasa, Zaire (Torrey & Way, 1991)-the number of Mricans infected with HIV continues its steady climb. An estimated 7 million Africans have been infected with HN, most of them through unprotected heterosexual vaginal intercourse (WorJd Health Organization [WHO I, 1992). The epidemic has reached crisis levels in some parts of Mrica, where up to 25% of urban adults are HIV-positive, and growing numbers of children have been infected perinatally or left as orphans because their parents have succumbed to AIDS-related infections. While the epidemic first hit hardest in urban areas of central and east Africa, rates of infection are climbing in the rural areas, and southern and west African countries have reported dramatic increases in the number of people with AIDS or testing positive for HIV (Torrey & Way, 1991). Government agencies and nongovernmental organizations (NGOs) have mounted a range of AIDS education and prevention programs in Mrica at national and local levels. As staff of the HIV/AIDS Prevention in Africa CHAPA) Support The authocs ace with the Johns Hopkins University, School o( Hygiene and Public Health, Instltute (or International Programs, 103 East Mount Royal Avenue, Baltimore, MD 21202.

81

SCUIT AND MERCER

82

Program between late 1989 and early 1992, we provided technical support to seven nongovernmental community-level AIDSprevention projects in Africa. 1 As we worked with these projects, it became increasingly apparent that effective communitylevel AIDS prevention programs required not only adequate time, funding and technical expertise, but an awareness of the social and cultural factors affecting the interpretation, spread, and prevention of HIV and AIDS? In this paper we will discuss our observations of ways that the sociocultural context can influence responses to AIDS, and why it is critical for community-level projects to try to understand, and grapple with, this context.

mE

SOCIOCULTURAL PERSPECTIVE

In the late 1980s, social scientists began to probe the social and cultural context of AIDS in Africa to explain why the epidemic has tom throughparts of the continent so quickly and why efforts to control the spread of AIDS have often met with much resistance. These qualitative findings address the issues from different and sometimes conflicting angles, but are nonetheless useful in understanding the problem from a sociocultural perspective. The research provides strong evidence that HIV transmission in Mrica is linked to the interaction of deeply-rooted sociocultural patterns and the disruptive course of colonial and post-colonial history. Caldwell and Caldwell (1989), for example, contend that many precolonial African cultures, particularly in West Africa, emphasized the production of children in order to continue the family lineage (usually the father's lineage) and did not stress strong marriage bonds. For women in many groups, premarital sex was not considered wrong, though adultery often was, and men were not expected to stay celibate before marriage or monogamous in marriage. Thus the Caldwells argue that precolonial cultures often allowed multiple sexual partners and even casual liaisons, particularly for men. Other researchers, working in central, eastern, and southern Africa, argue that disruptive political, economic, and social trends from the colonial era through the present day-such as rapid urbanization, separation of male migrant laborers from their families, lack of economic opportunities for urban women outside of prostitution, and general ill health and poverty-essentially weakened the marriage bond and greatly increased rates of prostitution (Ankrah, 1989; Bassett & Mhloyi, 1991; Larson, 1989; Schoepf, 1988). The evidence indicates that both precolonial norms and the dislocations of recent history have increased the transmission of HIV in Mrican communities. These large-scale, long-term trends may be difficult for a community-level project to influence, but such trends should not be ignored. Awareness of the complex factors that shape marital and sexual behavior can help staff design realistic and appropriate interventions that attempt to influence that behavior. Organizations undertaking IBased at the Institute for International Programs ofthe)ohns Hopkins University SChool of Hygiene and Public Health and funded by the U.S. Agency for International Development, Bureau for Africa. The HAPA grants projects were: World Vision Relief and Development in Kenya and Zimbabwe, Save the Children in Cameroon and Zimbabwe, Care in Rwanda, and Project HOPE in Malawi and Swaziland :lOur understanding of the social and cultural aspects of AIDS control also benefitted from several articles on AIDS outside of Africa, particularly the followlng, Feldman ( 1990), Farmer ( 1990), Parker ( 1990), and Worth (1990).

CULTIJRAL OBSTACLES

83

community-level AIDS control projects need to focus on local manifestations of these broader trends in order to tackle problems proportional to their resources. At the local level there are two approaches to community analysis, both strongly supported in the social sciences, that produce very different, though potentially complementary, kinds of data. The first approach looks at society as a collection of separate individuals, each maximizing his or her own self-interest. As measured by a standard knowledge, attitudes, and practices (KAP) survey, the sum of individual decisions constitutes the group tendency (Taylor, 1990). We will concentrate on the second approach, which views individual decisions as inextricably tied to social structures and cultural norms. To understand the AIDS-related behavior of individuals, it is essential to understand how individuals interpret the epidemic in light of the "whole system of people's thinking and doing" (Wallman, 1988). Here the quantitative data are useful, but not sufficient. Quantitative data alone, however important in identifying the prevalence of certain behaviors, do not provide us with an understanding of why they occur. Only qualitative data can clarify how an individual's beliefs and actions are interwoven with collective norms and structures. Concepts of critical importance to the spread and prevention of AIDS, such as health and sexuality, can differ tremendously between cultures, yet those differences remain hidden unless we are sensitive not only to what behavior is taking place, but to its meaning for each society.

11:IE HAPA PROJECTS

To illustrate the need to link qualitative insights to ongoing prevention efforts, we will draw from the experiences of the seven NGO HIV/AIDS prevention projects mentioned above. In addition to providing technical support, the HAPA Support Program assisted with overseeing the monitoring and evaluation of the projects. The HAPA grants were designed for groups with an established working relationship with the communities in the project target areas, and, in most cases, AIDS awareness and prevention activities were integrated with ongoing health or development projects. The projects focused on community-level training and education in AIDS prevention, through health workers, extension agents, and community leaders, such as political leaders, traditional healers, schoolmasters, and religious leaders. One of the first steps the new projects took in developing an action plan was to conduct a baseline assessment of the target populations. In gathering baseline data, five of the seven new projects followed what was for them the standard procedure of previous grants, such as the child survival, and undertook an initial knowledge, attitudes, and practices (KAP) survey. The projects found, however, that a KAP survey for gathering baseline information for their AIDSprojects was less useful than in the case of the child survival grants. AIDS was a new area for the NGOs. Well-tested and appropriate questions were not widely available, and field staff lacked training and experience in gathering quantitative data on HIV/AIDS. Governments were often cautious about the kinds of information that could be gathered about AIDS.As a result, it took a long time-usually several months, and in one case a full year-to collect and analyze the data. But more importantly,even after the results were analyzed and available, the surveys did not provide much intor-

84

SCOIT AND MERCER

Table 1. Comparison of Selected Responses to Baseline KAP Survey Questions for HAPA Grants Projects,

1989-90

Heard of AIDS % World Vision Kenya (n = 2630) CARE Rwanda (n = 360) HOPE Swaziland (n = 1930) Save the Children Zimbabwe (n = 1176) World Vision Zimbabwe (n = 2018)

97 96 87 91 88

Know sexual mode of transmission %

Never use condoms %

90 56

77

70

82 75 67

mation about social and cultural factors that might have a strong influence on AIDS-related ways of thinking and acnng.> The KAP surveys undertaken by the five projects were able to obtain some useful information, as seen in Table 1. The surveys indicated comparable levels of knowledge and practice related to HIV and AIDS, and in fact some of the apparent differences may have been artifactual, due to variations in the exact wording of the questions. Three other trends that echo findings from other studies in Africa were apparent: 1. Most people had heard about AIDS, but by and large thought that it was a disease for other people, particularly female prostitutes and "promiscuous" men. In the CARE Rwanda KAP, 65% of the respondents denied that there was any personal risk of their becoming infected, although most thought their friends and neighbors were at risk. Bassett and Mhloyi ( 1991 ) report that, in Zimbabwe, an unmarried or divorced urban woman is almost synonymous with prostitute, and this stereotype has made its way into the AIDS control program. "An early poster depicted a woman in a miniskirt and high-heeled boots, dragging on a cigarette, and the caption exhorted men to remain faithful to their families" (p. ISO). This popular perception of prostitutes as a guilty, high-risk group tends to discourage other at-risk people from considering the potential AIDS-related consequences of their own behavior.

2. Respondents tended to know the sexual mode of HN transmission, but also had a great many important misconceptions. In the World VisionlKenya project, for example, one in five respondents thought that AIDScould be contracted by shaking hands with someone infected, and nearly one in three thought that the virus could be transmitted by mosquito bites. In Uganda, Ankrah (1989) notes a similar pattern of simultaneous factual awareness of AIDS coupled with equally Widespread misconceptions about the disease, which she attributes to "a continued inclination toward old conceptions and beliefs" (p. 267). Unable to identify research that explains the persistence of these beliefs, she identifies this as an area in need of further investigation. 'It is important to note that several MAPA grants projects did obtain both quantitative and qualitati\'e baseline Information about their Impact areas. For example, one project initially worked with a local anthropologist to orient their efforts with an ethnic minority target group, and two projects used focus group discussions and/or in-depth intervic:ws to guide specific aspects of their interventions. However, In part because of their past experience with requirements of other programs, most of the MAPAgrantees paId proportionately greater attention to the KAP surveys than to other kinds of baseline information.

CULTIJRAL OBSTACLES

85

3. Only a minority of respondents had ever used a condom, and many also denied any need to use a condom The basis for their resistance to condom use was not explored in the surveys. Taylor, an anthropologist working in Rwanda, critiques an extensive 1988 survey that reported that 97% of the Rwandan respondents knew that AIDS is sexually transmitted, and 71 % were seriously concerned about getting AIDS, yet none mentioned the use of condoms during intercourse. The survey is incomplete and misleading, Taylor (1990) argues, because it did not include an investigation of why the Rwandans questioned are so resistant to condom use. Taylor's own anthropological research indicates that the social construction of a moral person in Rwanda involves the exchange of "gifts of self," which include the flow of fluids involved in sexuality and reproduction. Rwandans resist condoms, at least in part, because they block this all-important flow between two partners and raise the fear of a subsequent blockage of fertility. Because the 1988 survey failed to explore this cultural dimension of Rwandan sexuality, Taylor thinks that someone reading the survey results might reach the mistaken conclusion "that Rwandans are an obdurate, irrational lot, impervious to the lessons of Western science" (p. 1023).4 Our experience with the HAPA projects also indicates that, in responding to these kinds of paradoxical findings, there is a temptation for outsiders to attribute the target population's misconceptions and risky behavior to ignorance or superstition. The seemingly obvious solution is then to educate people about HN/AIDS, to give them the facts about transmission and prevention. However, as the social and cultural research we reviewed indicates, many of the beliefs and practices charted in the KAP studies clearly are not simply the result of ignorance of the facts. Instead, a critical factor is the way that facts fit into the existing frameworks that guide people's thinking and behavior. As Taylor argues in the Rwandan example, "resistance to condom use makes perfect sense once we understand something about the way the person is socially constructed in Rwanda" (1990, p. 1023). KAP surveys often point out how people react to the menace of AIDS, but do not uncover the underlying logic of their reactions. The challenge is to make that hidden logic explicit, then find ways to work with or around it.

SHONA CULTURE AND THE EXAMPLE OF RUNYOKA Another resource for NGOs to use in gathering qualitative data is the work of local social scientists studying health, family life, or sexuality. An example of research appropriate to NGO AIDS-prevention efforts is the work by Zimbabwean anthropologistJane Mutambirwa, who describes how local patterns of cultural beliefs related to health, illness, sexuality, marriage, and morality influence peoples' perceptions of AIDS (1991). Her research focuses on the Shona ethnic group in rural Zimbabwe. Mutambirwa stresses that, although many from outside the culture tend to see local beliefs as haphazard and illogical, in fact those beliefs often represent a coherent and well-defined belief system from a local point of view. There are a great many ways in which those strongly-held beliefs will influence 4Bledsoe, in an article that draws from her research on women's reproductive strategies In liberia .and Sierra Leone as well from AIDS-related articles In local African newspapers and journals, helps clarify a range of potential social and cultural barriers to condom use (1990).

as

I 86

SCOIT AND MERCER

the way people understand a new phenomenon, such as HN and AIDS. For example, a prevalent belief in Zimbabwe links the spread of disease to bodily secretions, such as perspiration. This may be the primary basis for the Widespread fear of acquiring HIV infection through shaking hands. Religious and social norms in Zimbabwe powerfully and consistently emphasize that both men and women must have children before they are considered adults, capable of what is considered moral behavior. Those who die childless cannot be accepted into the spirit world of the ancestors, and must wander the earth as "evil, aggrieved or haunted spirits" (Mutambirwa, 1991). These beliefs, even if vaguely sensed and not explicitly articulated, may make it nearly impossible for a woman to insist on using condoms with her husband, or to choose to never have children if she is found to be HN-positive. One interesting example of a culturally-determined belief that has implications for our understanding of people's responses to information about AIDS is a traditional sexually transmitted disease known as runyoka. According to Shona-speaking informants in southeastern Zimbabwe, runyoka is a fatal disease believed to strike a man who has sex with another man's wife (Figure 1). The disease springs from a curse or hex, which a married man places on his wife to punish any other man who lures her into an illicit relationship. The problem seems to have different perceived characteristics in various parts of the country; other descriptions of runyoka indicate that it strikes men who break sexual taboos by having sex with a woman at certain times, such as when she is menstruating or has miscarried. Whatever the specific cause, the disease appears to be linked in people's minds to illegitimate sex, and strikes the guilty man, although not, it seems, the "innocent" husband or even his nonmonogamous wife. How does this cultural belief relate to the prevention of HN/AIDS? In a survey of the Save the Children HAPA project population in rural and periurban Zimbabwe in May 1991,22% of respondents overall (nearly 25% in the rural area) stated that they thought that AIDS was "the same" as runyoka. This widespread identification of HIV/AIDS with runyoka may reflect the similar physical aspects of both diseases, such as weight loss, weakness, and malaise. Yet these physical similarities are less important than the moral parallels that people draw between AIDS and runyoka: that both diseases are sexually transmitted and tend-though only initially, with HIV/

Jealous husband

• • • • Wife of jealous husband ~ ~ ~ ~ ~ ~

Male lover hexed by jealous husband

• • • • Wife

= legitimate sexuality--disease not transmitted ~ ~ = illegitimate sexuality--disease transmitted

KEY: • • •

Figure 1. Proposed Route of Sexual Transmission

of Runyoka in Zimbabwe

1,

) I

I

i

CULTIJRAL OBSTACLES

87

AIDS--to strike those people who engage more frequently in what arc considered illegitimate sexual relations. Both HIV/AIDS and runyoka arc often perceived as resulting more from a moral failing than from a physical interaction. 1111smoral understanding of HIV/AIDS makes it difficult to dismantle the destructive and pervasive stigma that heavily burdens those who are Hlv-Infected or 111with AIDS. The perception of HIV/AIDS as identical or related to runyoka could have important implications for an educational campaign that alms to reduce rls"",, behaviors contributing to the spread of HIV infection. A woman may assume that if she is faithful to her husband she can not be infected with HIV, even if her husband has other liaisons, because she is innocent of sexual wrongdoing. As a result, shc may not feel threatened and may have less reason to try to protect herself from the virus. In addition, she may think less about possibly passing the virus on to an unborn baby. Men also may believe that if they do not sleep with married women, or if they avoid other specific taboos, they may be safe from the disease.

TIlE QUALITATIVE RESPONSE The example of runyoka illustrates why community-level AIDS prevention programs need to identify and respond to the AIDS-related sociocultural context in the project area. Yet this process poses substantial challenges. While some AIDS-related organizations are starting to take the social and cultural context more explicitly into account, many stiU do not perceive the importance of bringing qualitative Information to bear on ongoing prevention programs. Some types of social science research are lengthy and expensive, and thus impractical for resource-strapped NGOs. 1110se NGOs concerned with gathering qualitative data in timely and less expensive ways may tum to Rapid Assessment Procedures (RAP), a loosely-grouped set of methodologies drawn from anthropology and marketing research, such as focus group discussions, participant observation, and interviews with key informants (Scrimshaw, 1987). The RAP techniques require intensive but short-term initial training of project staff, and then can be used to assess social and cultural factors ora specific group or area prior to project design and implementation, or to refine, or even redefine, a project already off the ground. The short duration of most RAP methodologies-usually only a few weeks-is well suited to the tlmcframes and pocketbooks of many organizations, but does put strict llrnlts on the scope of the data gathered. To return to the Zimbabwean case: Given that a significant percentage of the project population thinks AIDS is similar to runyoka, how might project staff respond? To their credit, the Zimbabwe Save the Children project had already recognized and targeted belief in runyoka for their prevention efforl'l. Selecting the best educational approaches to use, however, would clearly benefit from further investigation of a qualitative nature. The specitlc actions that might be taken as a result would depend on the findings of that investigation. We can not a...surnc that local NGO staff, often heavily influenced by advanced educations, urban surroundings, and expatriate counterparts, have a deep understanding of traditional (particularly rural) belief systems, or that they feci comfortable taking traditional beliefs seriously in a formal project setting. However, local staff probably do have a basic awareness of possible cultural obstacles to AIDS prevention, and outside consUltants or staff can help legitimize discussion of these issues. For example, project staff, drawing on the research of social scientists or using RAP techniques,

scorr

88

I

AND MERCER

might find that married partners were not using condoms because they considered sexual relations within marriage legitimate and therefore immune from contagion (as with runyoka). In response, the project might team up with local family planning efforts to: (a) teach clients how AIDS is different from runyoka, emphasizing that everyone, not just those perceived to have illegitimate sexual relations, is vulnerable to HIV infection and AIDS; and (b) promote the use of condoms within marriage for birth spacing and HIV/AIDS prevention. Essentially NGOs need access to staff who are trained in the timely collection of qualitative data--either as employees or as consultants. The drawback is, clearly, that training staff in qualitative data-gathering skills, or utilizing the services of those who already have the skills, can take up significant amounts of project resources, even if a PVO utilizes time and money-saving techniques such as RAP. At this time, the need for qualitative data to assist in the development and evaluation of AIDS projects is not yet broadly enough understood to make it automatically a part of start-up activities for most projects. On the other hand, we have seen in Zimbabwe and elsewhere that project staff are eager to improve the effectiveness of their prevention strategies, and many find qualitative methods to be exciting and useful tools. At the 1990 workshop for HAPA grants field staff in Zimbabwe, a local communications expert introduced the participants to focus groups, and took them to an NGO impact area to practice moderating and recording focus group discussions. In over 50% of the evaluation forms received after the workshop, project staff specifically cited the focus group sessions as having an impact on the way they carry out their project after the workshop, as well as their way of thinking about issues related to their projects. CONCLUSION This paper has focused on the experience of seven community-level nongovernmental projects in sub-Saharan Africa, but projects of all sizes, governmental and nongovernmental, in the "developed" and the "developing" worlds, may benefit from greater awareness of the sociocultural context influencing the AIDS-related beliefs and behaviors of their target populations. In many projects, an informal understanding of this context already influences intervention strategies. For example, the sex-positive approach to AIDS prevention found in comic books, videos, and workshops by and for gay American men reflects their culture's tendency to approach sex with greater frankness and humor than do other segments of the American population. A project staffed by members of the target community will tend to have less trouble perceiving and incorporating that community's essential norms and values. When project staff are not from the target community, it is essential to generate both the motivation and the mechanisms to understand the implicit assumptions shaping how the target population responds to AIDS. Donor organizations tend to strongly encourage the development and monitoring of quantitative indicators and objectives; now it is essential that donors alsO begin consistently to support the gathering and use of qualitative data. The provision of training in rapid data collection would be particularly useful for organizations that would like to incorporate these methodologies into their project work, but cannot shoulder the costs of a workshop or a consultant. Establishing a regional network of trainers in qUalitative data gathering would make the acquisition of these skills more affordable and more accessible. Building up the skills of community-level institutions is essential to slowing the spread of HIV/AIDS in Africa and elsewhere, and

) )

j

89

CULTIJRAL OBSTACLES

requires a high degree of collaboration al, and local levels.

and coordination at the international

nation'

REFERENCES Ankrah, E. M. (1989). Methodological problems in studying its prevention and spread. Social Science and Medicine, 29, 265-276. Bassett, M. T., & Mhloyi, M. (1991). Women and AIDS in Zimbabwe: The making of an epidemic.

International journal of Health Services, 21, 143-156. Bledsoe, C. (1990). The politics of AIDS and condoms for stable heterosexual relations in Africa: Recent evidence from the local print media. W. P. Handwerker (Ed.), Births and Power. The Politics of Reproduction Boulder, Colorado: Westview Press. Caldwell, J., Caldwell, P., & Quiggin, P. (1989). Disaster in an alternative civilization No.2. Health Transition Centre, the Australian National University. CARE International. (1991, August 1). Southeast

integratedAIDS education and training pilot project, july 1, 1989-]une 30, 1991: Midterm progress report. CARE Rwanda. Byumba

Farmer, P. (1990). Sending sickness: Sorcery, polio tics, and changing concepts of AIDS in rural Haiti. Medical Anthropology Quarterly, 4. Feldman, D. (1990). Culture and AIDS [introductory essay]. New York: Praeger. Larson, A. (1989). Social context of human immunodeficiency virus transmission in Africa: Historical and cultural bases of East and Central African sexual relations. Reviews of Infectious Diseases, 2 (5), 716-731. Mutambirwa, J. (1991). Aspects of sexual behavior in local cultures and the transmission and prevention ofHIV/AIDS. In M. A. Mercer & S. J. Scott (Eds.), Tradition and transition: NGOs respond to AIDS. Baltimore: The Johns Hopkins University School of Hygiene and Publie Health. Parker, R G. (1990). AIDS education and health promotion in Brazil: Lessons from the past and prospects for the future. }. Mann, H. Fineberg, & J. Sepulveda (Eds.), AIDS education and communication London: Oxford University Press. Project HOPE. (1991). Quarterly program report

PVO HIV/AIDSprevention in Africa, FY 1989: Swaziland, Africa january 1, 1990 to March 15, 199/. Save the Children Federation. (1990, October).

HAPAgrant midterm progress report:Training of trainers for AIDS education, September I, 1989-August 31, 1990. Zimbabwe Field Office. Schmidt, N. J. (1990, April). Resources to assess the social impact of AIDS in Mrica. AIDS & Society, pp. 14-18. Schoepf, B. G. ( 1988). AIDSand Society in Central Africa: A View from Zaire. N. N. Miller (Ed.), AIDS in Africa. Lewiston, NY: The Edwin Mellen Press. Scrimshaw, S. C. M., & Hurtado, E. (1987). Rapid

assessment procedures for nutrition and primary health care. Los Angeles: The United Nations University and UCLA Latin America Center. Taylor, C. C. (1990). Condoms and cosmology: The 'fractal' person and sexual risk in Rwanda. Social Science and Medicine, 31, 1023-1028. Torrey, B. B., & Way, P. O. (1991). SeroprevaIence of HIV in Mrica: Winter 1990 (CIR Staff Paper No. 55). Washington, D.C.: Center for International Research, U.S. Bureau of the Census. Wallman, S. (1988). Sex and death: The AIDS crisis in cultural context.journal OfAcquired Immune Deficiency Syndrome, 1, 571-578. World Health Organization, Global Program on AIDS. (1992). AIDS Update, June 30. World Vision Relief and Development (1990, March). Report on AIDS project knowledge, attitude and practice survey. World Vision Kenya. World Vision Relief and Development. (1990, April). Survey report on AIDS knowledge, atti-

tudes and practices in commercial farming areas:Marondera district. Preparcd by Prances Chinemama for World Vision Zimbabwe. Worth, D. (1990). Minority women and AIDS: Culture, race, and gender. In D. Feldman (Ed.), Culture and AIDS. New York: Praeger.