Encouraging Discussion With Partners and Building Negotiation Skills: Hiv Prevention Strategies For Women in Relationships in Brazil, Tanzania and Indonesia

Susan E. Middlestadt, Academy for Educational Development
ABSTRACT - This presentation will discuss results from a research and intervention project being conducted in Brazil, Tanzania and Indonesia. The participants in each country are women in relationships. The goal is to increase their discussions with their partners about sex, health and ways to protect themselves from AIDS. Women are at risk of HIV disease, not only because of their own behavior but also, and often primarily, because of the behavior of their partners. For a woman, condom use is not a behavior. She must ask and then convince her partner to use a condom. A woman who has decided to limit herself to one sexual partner must also convince her partner to be monogamous. Thus, talking, discussing and negotiating represent important behaviors to analyse, understand and influence. Results show that many women recognize the benefits of such discussion in protecting themselves from AIDS, in bringing them closer to their partners and in helping him change his behavior. However, they also see negative consequences in making him angry, causing him to leave and making him suspicious. Women who talked with their partners differed significantly from those who don't in their perceptions of what significant others, such as closest friends, doctors, church members, thought they should do. Strategies for increasing dialog will be discussed. Special attention will be given to the challenges of applying consumer research tools in developing countries.
[ to cite ]:
Susan E. Middlestadt (1993) ,"Encouraging Discussion With Partners and Building Negotiation Skills: Hiv Prevention Strategies For Women in Relationships in Brazil, Tanzania and Indonesia", in NA - Advances in Consumer Research Volume 20, eds. Leigh McAlister and Michael L. Rothschild, Provo, UT : Association for Consumer Research, Pages: 297-301.

Advances in Consumer Research Volume 20, 1993      Pages 297-301

ENCOURAGING DISCUSSION WITH PARTNERS AND BUILDING NEGOTIATION SKILLS: HIV PREVENTION STRATEGIES FOR WOMEN IN RELATIONSHIPS IN BRAZIL, TANZANIA AND INDONESIA

Susan E. Middlestadt, Academy for Educational Development

ABSTRACT -

This presentation will discuss results from a research and intervention project being conducted in Brazil, Tanzania and Indonesia. The participants in each country are women in relationships. The goal is to increase their discussions with their partners about sex, health and ways to protect themselves from AIDS. Women are at risk of HIV disease, not only because of their own behavior but also, and often primarily, because of the behavior of their partners. For a woman, condom use is not a behavior. She must ask and then convince her partner to use a condom. A woman who has decided to limit herself to one sexual partner must also convince her partner to be monogamous. Thus, talking, discussing and negotiating represent important behaviors to analyse, understand and influence. Results show that many women recognize the benefits of such discussion in protecting themselves from AIDS, in bringing them closer to their partners and in helping him change his behavior. However, they also see negative consequences in making him angry, causing him to leave and making him suspicious. Women who talked with their partners differed significantly from those who don't in their perceptions of what significant others, such as closest friends, doctors, church members, thought they should do. Strategies for increasing dialog will be discussed. Special attention will be given to the challenges of applying consumer research tools in developing countries.

INTRODUCTION

Around the world, HIV infection is increasing and it is increasing most rapidly among women. The World Health Organization estimates that as of 1992 from 10 to 12 million adults have been infected by HIV and expects this number to reach 30 to 40 million by the year 2000. Further, the male-female ratio is soon expected to reach unity. There has not, however, been a corresponding increase in the number of successful HIV prevention programs for women. While there are many programs for women in the commercial sex industry, there are few for women in relationships. Furthermore, as yet, there are no effective methods of risk reduction that are completely under the woman's control. Abstinence and condom use require the cooperation if not the initiation of the partner. The "female" condom not only will be a long time in becoming widely available but will require the partner's knowledge and consent. Monogamy is successful for a woman only if her partner was, is and remains monogamous. In sum, women are at risk of HIV disease, not only because of their own behavior but also, and often primarily, because of the behavior of their partners. And, interventions are needed that help women influence the behavior of their partners. At the same time, it is clear that to be effective, programs for women must be developed in the context of her social, economic and/or emotional dependence on her partner. Briefly, the potential negative responses of the partner are often feared consequences of discussing AIDS, often more feared than AIDS itself.

The goal of this project is to understand the communication and negotiation behaviors of a woman with her partner and, based on this understanding, to design interventions that reduce constraints, communicate benefits and build skills around discussion with partners. In describing this project, this paper will illustrate a research process that is very different from the type of research process practiced in either academic institutions or commercial market research agencies. In essence, the research is conducted in an applied, social marketing context. Furthermore, it is collaborative one, not only between countries but between researchers, program designers and the people served by the project, its constituents.

RESEARCH CONTEXT

The Women in Development project is one activity of a larger project, the AIDS Technical Support: Public Health Communication Component (AIDSCOM). AIDSCOM is funded by the Offices of Education, Health and Population, Bureau for Research and Development, United States Agency for International Development (Project No. 936-5972, Contract No. DPE-5972-Z-00-7070-00). The AIDSCOM project is conducted by a team of five organizations, the Academy for Educational Development, Porter/Novelli, the Johns Hopkins University, the Annenberg School of Communication at the University of Pennsylvania and Prism/Day. This staff of this team provides technical assistance in the design, implementation, and evaluation of HIV/AIDS prevention programs in developing countries. The programmatic activity of the AIDSCOM project has a common set of objectives around using communications for behavior change to help prevent and control HIV disease. Similarly, the research is guided by a centralized research agenda. Within this common framework, specific projects and country programs are adjusted to the needs of the country and are conducted by local governmental and nongovernmental programmatic and research agencies.

The Women's project is being conducted in three sites: Rio de Janeiro, Brazil; Dar es Salaam, Tanzania; and Jakarta, Indonesia. Not only are these three sites from different regions of the world, but they represent different phases in the course of the epidemic. In Brazil and in Tanzania, HIV/AIDS has been recognized a problem for many years; there has been much attention in the media; and most everyone has heard of AIDS and knows the basics of sexual transmission. However, Brazil and Tanzania differ in the status of the epidemic among women. In Brazil, early cases were primarily male and it is only now that women are accounting for a significant proportion of the cases. In Tanzania, women have accounted for about half of the cases from the beginning. In contrast to both Brazil and Tanzania, HIV/AIDS is a relatively new issue in Indonesia. Infection seems to be entering through the international drug and commercial sex industries. As yet, there are few reported cases and the general population is not fully aware of how their sexual behavior places them at risk.

As a multi-site project, we have a particularly long list of collaborating agencies and individuals. Kathryn Carovano and Lorraine Lathan-Parker are coordinating the project with me in Washington. In Brazil, we are working with women attending family planning clinics through BEMFAM, a family planning organization. Our constituents are adult, sexually active women who are formally married, in consensual union with a companion or in steady relationships. Carmen Guimaraes, Elisabeth Verraz and Vera Vital Brasil are our in-country colleagues. In Tanzania, we are working with women insurance workers through OTTU, the women's department of a labor union and BIMA, an organization of insurance agencies. These women have jobs and are more educated than rural women in Tanzania; about half are married and half are not yet married. Again, all are sexually active. Lizbeth Loughran, Maudline Castico and Siham Ahmed are directing the project in Tanzania. In Indonesia, we are working with women attending a University of Indonesia clinic through LPT, an applied research unit of the Universty. The women attending the clinic are urban and moderately educated. Because of the culture, we are working with only married women in Indonesia. Bernadette Setaidi and Ninuk Widyantoro are our associates in Indonesia.

METHOD

The project consists of four phases: qualitative research; quantitative research; intervention; and evaluation. During Phase I, a qualitative study using face-to-face interviews is conducted with 40 women from each of the populations. The open-ended questions were designed to elicit salient outcomes, salient referents and effective strategies for two behaviors, "discussing with my partner the things we both need to do to protect ourselves from AIDS" and "asking my partner to use a condom everytime." Data from open-ended questions are then used to design the questionnaire for Phase II, a quantitative survey with about 200 women at each site. In addition, to demographics, knowledge about AIDS and practice of general health communication behaviors, the survey tapped constructs from three major theories of behavior, the theory of reasoned action, social cognitive theory and the health belief model.

An analysis of the Phase II survey is being used to provide input as to the content of Phase III, the intervention with about 100 women. In each country, the logic and form of the intervention is the same. It consists of weekly sessions with groups of women. The weekly meetings provide the women with a source of social support for behavior change. A facilitator using a structured guide will help the women who have successfully negotiated share their experiences and strategies with the women who have not. Through role playing and other interactive activities, the women can practice and improve their skills. In each case, the goal of the intervention is the same, to encourage communication and negotiation with partners. However, the degree of negotiation and the specific strategies to be used will evolve from the women, based on the data from the quantitative survey and the discussion groups. Phase IV is an evaluation with both qualitative and quantitative aspects.

Thus, the methodology of the project, both in its research and in the intervention aspects, is oriented toward the constituents or in the language of social marketing, the consumers. This consumer approach helps ensure that the research and the intervention are effective, appropriate and sensitive to the particular social and cultural context of each group of women. In addition to research findings, a final product will be a revised facilitator's guide that is appropriate for the constituents who helped in its development.

RESULTS

The project is still underway. But I would like to take the opportunity to illustrate this particular type of consumer research process by discussing some early findings. This paper will discuss qualitative findings on the specific communication and negotiation behavior chosen for study and for intervention, the consequences of communication and negotiation elicited from the open-ended questions of advantages and disadvantages, the referents given by the women as approving or disapproving, and the strategies recommended by the women as potentially useful.

Target behavior

A major decision in designing behavior change interventions is the selection and definition of the target behavior. The behavioral researcher plays a major role in this decision process. All too often, program design focuses on providing information and educating people. Attention is paid to selecting the target consumer, deciding what they need to know and choosing an appropriate channel to reach that consumer. In many social marketing domains, particularly health domains, the assumption is made that making people aware of the disease or condition and knowledgeable that they are at risk and that they can take certain steps to reduce that risk is sufficient to motivate behavior change. Thus, many HIV control programs begin by making people aware of the epidemic, teaching them about modes of transmission and recommending safer sex practices. In our work on HIV disease, we have found that this approach does not work for very many, if any, people. Instead, it is necessary to take a behavioral approach, that is, to define the target behavior and its determinants and to work back from that behavior. An important job of the behavioral researcher is to help the team define "what we want to encourage the women to do."

A key conclusion of the AIDSCOM Women in Development project is that the primary negotiation behavior to be addressed by the intervention needs to be different for the three different groups of women. While, after the fact, this is not surprising given the vast differences in the cultural context of the three groups of women we are working with, a discussion of the definition issues usefully illustrates the consumer research process necessary for AIDS prevention work. We began the project planning to study two behaviors, "talking with my husband or steady partner about the sex, health and AIDS prevention strategies" and "always using condom with my husband or steady partner" and hoping to intervene on one, the talking behavior. Based on pilot research with each group of women, the behavior eventually studied in detail and targeted by the intervention differed in terms of the topic of the discussion, the description and definition of the partner with whom the discussion was to occur and the specific communication or negotiation action recommended.

In terms of the topic, we knew we didn't want to encourage the discussion of "AIDS" or even "AIDS prevention". One can discuss AIDS without discussing changes in one's own sexual behavior. Discussing AIDS can mean discussing someone else's problem (e.g., foreigners, promiscuous people, gay men, prostitutes) and AIDS prevention can mean avoiding people with AIDS or taking steps to restrict the activity of people with AIDS or groups perceived to be at higher risk. The risk reduction strategy to be discussed would need to differ by country, by population and by individual. Thus, we didn't just want the couple to talk about condoms, either inside or outside the relationship. Condom use might be appropriate for some couples, but other changes in sexual behavior might also be effective strategies to discuss. In the pilot work to Phase I, we tried "sex, health and AIDS prevention strategies." We felt these three terms combined things women could easily talk about (health), with those that were more difficult (sex) and with a reason to combine them (AIDS prevention) without designating a particular risk reduction strategy. In Indonesia, using this topic, we found that the women felt we were talking about three completely different topics that had no relationship to one another. The term "strategies" was too abstract. More important, it seemed that the women in Indonesia, not being highly exposed to communications on AIDS and AIDS prevention, were not yet fully aware that their sexual behavior was connected to AIDS prevention. In Tanzania and in Brazil, we opened with a paragraph listing a variety of risk reduction strategies (all of which were familiar to the women) and noted that they all required talking with their husband or steady partner. In Tanzania, the paragraph ended with the need to talk about the "things we both need to do to protect ourselves from AIDS." In Brazil, the talk was to be about "AIDS, the risk of infection, and about how both of you can prevent and take care of yourselves." Notice the emphasis on both members of the couple. This approach was more successful, perhaps because of the increased familiarity of the urban women in Brazil and Tanzania with HIV and AIDS.

Naturally given that partnership patterns are a major source of differences between and within cultures, the partner with whom the discussion was to occur differed for the different groups of women. We wanted to designate a sexual partner with whom the women had a longer-term and committed relationship. Difficulties with communication and negotiation are the greatest in longer term relationships. In Indonesia, all the women were married, so the husband was the appropriate choice. In Brazil, the term husband or steady partner seemed to communicate about one partner to women in formal marriages as well as to those in consensual unions and to those in steady relationships. In Tanzania, we began with the term "husband or steady partner". For some marrried women, this phrase implied two different men. We therefore explicitly asked the women in Tanzania to think of one man: for married women, their husband and for unmarried women, their fiance or special boyfriend.

The specific action to be studied and targeted for change by the intervention also differed. It became clear in our pilot work, that there were a variety of possible communication and negotiation behaviors and that the appropriate one to encourage would depend on the particular context and culture of each group of women. For this project, we felt that an appropriate action to choose would be one that some but not all women in our constituent group were already practicing. If no women were yet practicing it, we would run the risk of encouraging a communication behavior that would put our constituents at risk of harm. If all the women were practicing it, there would be no room for change. In Brazil, in consultation with our colleagues, we began with "conversa", talking. On the basis of responses to behavior, intention and attitudinal questions, we verified that about half of the women were positively predisposed toward talking and about half were not. Thus, we seemed to have identified an action that was possible in that some women were already practicing the behavior, yet around which there was still room for change.

In contrast, we found in our qualitative research in urban Tanzania that the women were already talking to their partners. They found it easy to talk, almost 100% had ever talked and many had asked their partners to use a condom. This is not surprising given the length of time AIDS has been an issue, particularly for our constituents in Tanzania, relatively highly educated, urban and employed women. Thus, it was necessary and possible to go beyond the behavior of talking to more difficult and potentially more effective negotiation behaviors.

TABLE 1

COMMUNICATION AND NEGOTIATION TERMS IN TANZANIA

Table 1 presents a list of negotiation and communication terms that we considered and discussed with our colleagues and constituents in Tanzania. Notice they range from easy ones like talking, to more difficult ones like refusing to have sex if he won't use a condom. We decided that on the basis of the qualitative data alone, we could not select a single behavior to study in detail in Tanzania. Thus, the Phase II questionnaire used in Tanzania contained four sections, one each on four behaviors representing different points of difficulty: discussing with him, forcing him to use a condom, refusing to have sex if he wouldn't use and using a condom. And, analyses of the Phase II data would help decide the behavior to target as well as to identify potential determinants.

More generally, this list of negotiation behaviors illustrates another key conclusion of the project, that communication or negotiation represents not one behavior and not even a set of behaviors. Communication and negotiation represent a sequence, series or continuum of connected behaviors. A woman in deciding whether or how to try to negotiate with her partner, a behavioral scientist trying to understand and a program designer deciding where to intervene all must recognize that the behaviors are linked. And further, the goal of the behaviors is an outcome, to produce a change in the behavior of another.

Consequences of negotiating

Most theories of behavior recognize the role of the positive and negative consequences of a behavior in influencing the practice of that behavior. Thus, in the qualitative research phase of this project, the women were asked to indicate the advantages and disadvantages of talking with their partner. Before examining these responses, it is useful to consider their complexity. Table 2 presents a particularly informative quotation from a 28 year old, single woman with no children who participated in the qualitative phase in Tanzania.

TABLE 2

This quotation describes a chain of consequences, similar to the sequence of behaviors discussed above. The initiation of a discussion may or may not lead to asking him to use a condom outside the relationship, which may or may not lead to him agreeing to use a condom, which may or may not lead to him using a condom, which may or may not lead to infection and death for the man, which may or may not lead to infection and death for the woman, which may or may not lead to social and economic consequences for the nation. As described above, in deciding whether or not to engage in a particular negotiation behavior, one must consider as one of the consequences of that behavior that another behavior might occur carrying its own set of consequences. Clearly, it is difficult both to decide and to study these behaviors as single behaviors in isolation.

Table 3 gives the consequences elicited from the qualitative studies conducted in the three sites. Notice that this list includes multiple versions of a number of consequences. The list represents more of a universe of consequences than the salient or top of mind consequences.

TABLE 3

CONSEQUENCES OF NEGOTIATING SAFER SEX WITH PARTNER

Notice that these consequences are consequences of each of the negotiation behaviors, of asking, of persuading, of refusing to have sex and of using a condom and that many refer directly or indirectly to the outcomes resulting from the performance of these behaviors. This table illustrates a key point about consequences evident in this and many other studies on health behaviors, including HIV disease: the importance of NONhealth consequences as behavioral determinants. Clearly, some of the consequences are health consequences. The women are concerned that negotiation may or may not effectively protect them from AIDS, from other sexually transmitted disease and from pregnancy. And, indirectly, they perceive the discussion may or may not help him change his sexual behavior. However, most of the consequences on this list are NONhealth ones. The women are concerned about the positive and negative effects of negotiation on their relationship. These consequences must be addressed to help encourage women to initiate discussions with their partners.

The list of consequences was remarkably similar for the three different sites and is in fact similar for a variety of other target audiences. There are differences, however, between the groups in the association that these consequences have with behavior, both within and between the three sites.

Social referents for negotiating

There is substantial evidence that normative factors are important determinants of a variety of behaviors in the domain of HIV disease. On the one hand, this is obvious in that many of the behaviors are sexual behaviors that occur in interaction with another person. Less obviously, it is becoming increasingly clear that sustained changes in sexual behavior are often based on changes in social norms. Open-ended questions on people who would approve and disapprove were asked to identify the potential reference groups.

TABLE 4

SOCIAL REFERENTS FOR NEGOTIATING SAFER SEX WITH PARTNER

Table 4 gives the referents elicited from the three studies. There are few surprises in this list. As you can see, many of the referents are the same as for all behaviors, spouse, close friends, parents and other relatives. The three sites differed as to which specific other relatives were mentioned. Brothers were more frequently mentioned in Brazil; aunties, being traditionally responsible for education in sexual matters, were elicited in Tanzania; and in Indonesia, saudara or relatives of the same generation, were mentioned. Health professionals are frequently mentioned when considering a variety of health domains and religious leaders come up given the sexual behaviors that are at issue. A unique reference group was elicited in these studies, the friends and relatives of the husband. The woman's friends and relatives were most often mentioned as approving relatives; whereas, the man's relatives and friends were often listed as disapproving. Clearly, the concern for the reactions of the male partner extended to a concern for the social pressure being exerted on him by the others in his social environment.

Strategies for negotiating

It is also important to identify ways to facilitate effective negotiation. Open-ended questions were asked about the best time to talk, the best place to talk, topics that were easy to bring up, topics that were difficult to bring up and factors that would help discussion. Table 5 presents some of these results.

TABLE 5

POTENTIAL STRATEGIES FOR SUCCESSFUL NEGOTIATION

Notice that some of the women felt that discussion should occur indirectly, e.g., a discussion to the children with the husband listening or placing a pamphlet or a condom in the partner's suitcase when he is going away. Others felt that it was possible and more effective to approach the issue directly, in a straight way. Notice that many of the women felt it would be useful to have an external stimulus to initiate the discussion. A news article or a visit to the clinic could be a technique of giving a immediate reason for the discussion, outside of one's own behavior or suspicion about the other's behavior. Finally, a universal and immediate request at all three sites was to have a session with the men. In fact, this session was worked into the project as an additional optional session in Tanzania.

Perceived and actual skills in negotiating

Along with strategies for communication and negotiation comes skills at performing these behaviors. The women clearly felt a need to develop their skills in initiating discussions and in negotiating. Table 6 presents a quotation that effectively describes this need to know how to discuss.

TABLE 6

24 YEAR OLD HOUSEWIFE FROM BRAZIL

CONCLUSIONS

One specific purpose of this presentation was to describe a particular research and intervention project, a project designed to develop strategies to help women in relationships protect themselves from HIV disease. The preliminary results from the research on this project have shown that one way to help women in relationships protect themselves from HIV disease is to help them communicate and negotiate with their partners. The research has identified potential intervention points in the form of consequences, social referents, strategies and skills. Upon completion, the project should result in specific, culturally sensitive and appropriate recommendations for interventions for women that address these points in the three sites of the project.

A second, more general purpose was to describe a particular research context, an applied collaborative social marketing research process. I hope from this presentation, I have provided some insight into how collaborative, applied research occurs and illustrated a number of advantages of this research process. Basically, in addition to the rewards of working with bright, motivated colleagues around the world and of making a social contribution, a collaborative research process highlights our methodological assumptions, encourages the use and integration of qualitative and quantitative methods, forces a focus on the consumer and challenges our theoretical definitions and conceptualizations. I hope I have inspired some consumer researchers to join me and my colleagues here and around the world in this collaborative work.

----------------------------------------