Assessing the Social and Behavioral Science Base for HIV/AIDS Prevention and Intervention: Workshop Summary (1995)

This section focuses on the contribution of behavioral and social sciences to the biomedical field, particularly intermediate technologies that address and utilize the interface of biology and behavior to reduce the probability of disease transmission. This is not a fundamentally different topic from Section IV of this report; rather, it combines the knowledge presented there with the medical field.

The work to develop a vaccine for HIV prevention is one of the few examples where the biomedical and social science communities have attempted a marriage, and the results have proved cautiously optimistic. There are many other approaches to HIV prevention that are medically based. These include trials of barrier methods, such as topical microbicides used in conjunction with condoms; intervention trials to control cofactors of HIV transmission, such as the detection and treatment of sexually transmitted diseases; and studies to assess the infectiousness of individuals who are already infected. One success story in the latter area is the AIDS Clinical Trial Group (ATG), known as the 076 trial, in which HIV-positive pregnant women treated with zidovadine (AZT) give birth to fewer babies who are seropositive. For any medically based prevention trial to work, however, a partnership is needed among three groups: biomedical scientists, social and behavioral scientists, and target communities. If the social and behavioral issues regarding trial design, recruitment, informed consent, adherence to protocols, and evaluation of outcomes are not addressed, the trials can be undermined so that they are not successful or valid.

Medically based prevention trials frequently require social and behavioral science data in the design phase. For example, investigators may need data regarding the sexual behaviors in a particular high-risk setting or community; the predominant means of HIV transmission in the target population (e.g., sexual compared to injection drug use); the sexual networks (including the risk profiles of partners, the partner exchange frequencies, and the geographic region from which the partners come); the population structure, mobility, and occupation; and the amount of disease recognition and use of health care in the community. Recently developed methodologies, such as spatial mapping, have greatly facilitated the combining of social and epidemiological research, but much more remains to be done.

To recruit people into a medically based clinical trial, the researcher needs to promote increased awareness in the community about the trial and gain the community's support through partnerships established in the planning process. Many high-risk communities are skeptical about the value of medical research and mistrust the government— attributes that are the legacy of unethical research, such as the study of the natural history of untreated syphilis in Tuskeegee, Alabama, sponsored by the U.S. Public Health Service. Some communities have experienced such social upheaval that networks and communication patterns are disrupted. Community assessment is necessary to detect both negative and positive attitudes and to determine what structures can facilitate communication and where productive alliances can be formed. Engaged opinion leaders in the community can help change misperceptions, gain community acceptance, and avoid errors in the field trial design. Even if community leaders become interested in participating, the people at risk for HIV still need to be informed and recruited. Social, behavioral, and biomedical scientists are beginning to team together to develop strategies for identifying and engaging high-risk individuals. For example, the design of brief assessment tools that can encourage people to answer fairly quickly questions they would ordinarily be reticent to discuss with a stranger is a high priority in these efforts. There also have been efforts to identify pregnant women who are not receiving adequate prenatal care and do not know they are HIV-positive.

The Centers for Disease Control and Prevention (CDC) and the National Institutes of Health (NIH) are currently involved in work on community assessment and engagement. They are developing strategies for enhancing community collaboration in the future implementation of large-scale HIV

vaccine trials. They will assess attitudes and knowledge about HIV, AIDS, and medical research within the target communities and describe potential communication barriers between researchers and prospective study volunteers. Social and behavioral scientists are working with members of the communities to identify social networks, informal structures, and health-seeking practices that facilitate more active community involvement and participation. Guidelines will be developed to help researchers and communities in applying effective, sensitive preventive interventions in various settings.

Social and behavioral scientists have begun to work with biomedical scientists and communities to develop strategies to ensure that subjects give their consent only after being fully informed and that they understand the research design of the studies for which they are enrolling. Participants need to understand accurately the limits of the studies' benefits. For example, if a woman is getting treatment for a sexually transmitted disease, she needs to know that the treatment does not prevent HIV per se.

CDC investigators have taken the concept of informed consent one step farther and is investigating the concept of “community consent. ” This derives from the premise that a successful HIV vaccine efficacy trial will require not only the consent of individual participants but also the willing collaboration of the host communities.

Participants need to understand the social as well as the medical risks of participating in trials. These may include HIV seroconversion that results from a vaccine trial or public identification as a high-risk person in nearly all prevention trials. Minimizing potential social harms when researchers go into communities and enroll people in HIV prevention trials will be important in the ethical contract established between scientist and volunteer.

Conducting research on “social harm reduction” before trials get underway can help to facilitate medically based prevention trials in communities. The concept of social harm parallels the concept of biomedical harm or adverse events within the context of clinical trials. By its very nature, the probability of social harm depends on the context of the individual. In order to inform potential participants adequately of the risk they face by enrolling in a trial, investigators must understand the participants' social, political, and economic circumstances, including the underlying conditions that promote or support HIV risk behavior. Investigators also must understand the attitudes that result in discrimination against those at risk for or infected with HIV.

After participants give their informed consent and enroll in a trial, they must adhere to the trial's protocols if the results are to be meaningful. Poor adherence and retention rates can limit both the external and internal validity, decrease the ability to detect the effects of intervention, and necessitate increased sample size and cost. Measures are needed to determine whether participants understand what they are supposed to do and whether they are doing it. It would be quite helpful to be able to identify people who are likely to drop out of trials or people who will miss appointments or not take medications. With such predictors, it may be possible to design “nested” interventions that will improve adherence and retention in the trials. Tracking mechanisms are needed so that participants' behaviors can be systematically, but unobtrusively, monitored over time. Finally, there is a need to determine appropriate non-coercive incentives tailored to meet participants' needs.

Interest in the trials also needs to be maintained within the targeted communities. Typically, willingness to participate in trials declines over time, particularly in communities experiencing socioeconomic stress. In a parallel to the adherence of individual participants, the issue of “community adherence” also needs more investigation—that is, whether sustained community support can be maintained even with near optimal study volunteer participation and adherence.

The assessment of behavioral and social variables in trials is critical to the investigators' ability to analyze results and interpret findings. For example, if participants in a vaccine trial do not engage in risky behavior, they will not have the potential for getting infected with HIV, which means researchers may not know whether the vaccine worked. If participants in a vaccine trial incorrectly assume that they are protected from infection, they may increase their risky behavior, compromising their welfare and the researchers' ability to interpret the results. Another example points out the need to study the mobility of people as they move in and out of their communities: a community-wide program to control sexually transmitted diseases can be undermined if people outside the community who have sexually transmitted diseases come into the community and reinfect it.

The researchers who are involved in the development of intermediate technologies in medically based prevention trials are often nontraditional in their thinking and are able to work comfortably across disciplines; they appreciate that social, behavioral, and medical sciences together form the basic

research substrate for successful prevention trials. For example, one researcher studying volunteer adherence in clinical trials may be a social scientist whose primary appointment is in a Department of Internal Medicine, while another maybe a physician whose primary appointment is in a health behavior unit in a School of Public Health. Such interdisciplinary collaborations will be needed to address the social science challenges within biomedical HIV research endeavors.

Intermediate technologies that draw heavily on basic social and behavioral science research include needle exchange, vaccines, vaginal microbicides, and the diagnosis and treatment of sexually transmitted diseases. For example, needle exchange evaluation draws on molecular biology, mathematical models, and social science. Viral infection rates among exchanged needles assessed by polymerase chain reaction molecular hybridization can be interpreted biomathematically within the social context of the program for evaluative insights (Heimen and Kaplan, 1993). From the social sciences, knowledge regarding external political and legal constraints on individual action, networks, diffusion, behavior change, and education is incorporated into the intervention and data interpretation. This type of intervention asks society to make a major policy change by allowing the legal exchange of needles, and asks individuals to make themselves available publicly to the needle “exchanger” and to make behavior changes through regular needle exchanges. It is this mix of social responsibility and policy change, coupled with personal responsibility and acceptance of help, that is novel and promising. Much has been learned from cross-cultural, especially cross-national, comparisons on needle exchange. A number of cities (e.g., Sydney, Australia, and Tacoma, Washington) have prevented HIV epidemics among drug users by initiating needle exchange programs (National Research Council and Institute of Medicine, 1995). The social context of needle exchange require full examination if such programs are to be implemented throughout the United States. Social science research offers opportunities to examine the cultural and political viability of proposed interventions. The NRC/IOM report (1995) recommended more systematic research on community issues, including the concerns of ethnic groups who are known to be at high risk of HIV infection, and on the beliefs, attitudes, values and motivational factors that influence the reservations expressed by some professionals who potentially could have a role in needle exchange programs.

Vaccine trial preparedness requires not only the development of the vaccine but also the development of community readiness so the vaccine will

be accepted and utilized. In the United States, the highest HIV incidence rates are occurring in populations that often face potent barriers to participation in vaccine trials. What if a vaccine trial were held and no one came? Just as scientific consensus is expected in the wise selection of a vaccine product, so too must consensus be sought—broadly in society and specifically in the targeted communities —for such trials. Just as scientists are working on a systematic way to answer the questions related to vaccine interventions, so too must scientists overcome the social barriers to success in HIV vaccine trials, recognizing that the paradigms that hold true for one population may not hold true for another (Hoth, 1994; Grady, 1995).

Barriers that women can use themselves to prevent HIV infection will be important for personal protection. Condoms will not be worn by all men all of the time, a lesson learned from experience in the family planning field. Thus, the development and testing of vaginal microbicides represents an example of a critical prevention research issue. However, introducing the concept of self-administering of a vaginal suppository, cream, or film may be quite foreign to women in certain cultures or circumstances. Behavioral scientists, then, may help guide parallel research to determine how to maximize community participation.

The diagnosis and treatment of sexually transmitted diseases has become rather sophisticated. Recent work reported from Tanzania suggests that control of sexually transmitted diseases will decrease the spread of HIV (Grosskurth, 1995). Analogous work on this topic is currently being conducted in Uganda by Maria Wawer and colleagues. Such large-scale randomized controlled trials are likely to have important lessons not only for the developing world but also for the United States.

In all of these examples of intermediate technologies, there are questions of design, recruitment, informed consent, adherence, and evaluation that are critical to the trials' success. Of paramount importance are the issues of experimental design in community-based research and the ethical considerations in such research. Some government agencies have expressed a degree of commitment to address these issues by funding social and behavioral science research. For example, the National Institute for Allergies and Infectious Diseases (NIAID) has recognized the critical importance of community preparedness and social harm reduction in vaccine research and is sponsoring suitable behavioral research in collaboration with other NIH institutes and with the CDC. In addition, NIAID has initiated support for a broad range of prevention research initiatives aimed at answering immediate research questions while building vital experience in anticipation of HIV vaccine efficacy trials. Workshop participants strongly endorsed efforts among agencies to develop collaborative and interdisciplinary research programs.