Sociological AIDS Research
Needs an Intervention
by Judith D. Auerbach*
In an April 2005 C-SPAN interview, Peter Piot, Director of the Joint United Nations Programme on AIDS (UNAIDS), proclaimed that AIDS has been redefined as "not just a medical curiosity but as an obstacle to social and economic development . . . and also as a threat to security and stability." While this statement may have been revelatory to those in the biomedical and clinical world who have never considered the social nature of AIDS, to the social science community it registered a big "duh." Since the beginning of the epidemic, sociologists have pointed out that HIV/AIDS is not solely an individual biological problem but is also a social phenomenon associated with social ties and other forms of social influences, be these cultural, political, economic, or environmental.
Sociologists have noted the imperative to understand and address social drivers that affect how, where, and why different groups of people are more vulnerable to or resilient against HIV infection. We have argued that, unless the social, political, and cultural contexts that affect individuals’ levels and experiences of "risk" are addressed, the overall impact of individual-level HIV prevention and treatment interventions will be limited. But, until recently, this perspective has had little currency in the collective discourse of the AIDS research community.
Historically, the impact of social science—and American sociology, in particular—on the response to AIDS domestically and globally has been notably minimal, despite the fact that sociologists have been engaged in AIDS research from the beginning of the epidemic in three areas: social determinants (or "drivers") of HIV/AIDS; social impacts of HIV/AIDS; and social and behavioral interventions to mitigate HIV/AIDS transmission. The social determinants literature constitutes the basic social science research designed to elucidate the fundamental social mechanisms underlying the spread of infectious diseases. Analyses typically document and monitor associations between sociodemographic characteristics (e.g., age, sex/gender, race/ethnicity) and HIV risk and infection, or investigate the social etiology of the disease (i.e., its roots in poverty, gender-based violence, discrimination, mobility, etc.).
Sociologists also have examined the multiple impacts, or consequences, of HIV/AIDS on individuals, families, communities, and societies. Social research has illuminated the effect of local epidemics on demographic trends, household structure and functioning, informal caregiving, gender relations and dynamics, education, economy (including food insecurity and poverty), and governance and security.
Sociologists have been much less engaged, however, in the domain of interventions research, which is where most of the action in HIV/AIDS has been for the past two decades. Early in the epidemic, the life-and-death urgency of the response led the National Institutes of Health (NIH)—then and now the largest funder of AIDS research globally—to focus on interventions in both the biomedical and the behavioral and social science fields. The imperative to develop effective therapies, a vaccine, and behavioral interventions became institutionalized in the prioritization of research activities and the allocation of budget at the NIH (through an annual strategic planning process involving NIH program staff and external scientists and advocates). Although the mid-1990s saw a shift back to emphasizing basic biomedical research, this refocusing did not apply to basic behavioral and social research, which continued to prioritize and fund intervention studies.
There are a number of possible explanations for why sociologists have not fared well in this scheme. Chief among them is the social organization of the biomedical research enterprise at NIH that has militated against the development of a robust repertoire of social strategies for HIV prevention and mitigation. One key feature of that organization is the institutionalized hierarchy of scientific disciplines, which places social science at the bottom, and manifests in the lack of social—as distinct from behavioral—scientists in NIH program staff positions and on peer review panels. Another, related feature is the movement toward "evidence-based public health," an outgrowth of "evidence-based medicine." This approach emphasizes the application and valuation of rigorous research methodologies in public health science that parallel those used in clinical research—chiefly, the adoption of experimental study design, and specifically, the randomized controlled trial (RCT)—for assessing intervention efficacy. As a result of these trends, the questions, models, and methods of biomedical and clinical research have dominated what is considered meritorious and fundable science at the NIH (and elsewhere).
The problem for sociologists, of course, is that it is exceedingly difficult—and often inappropriate—to fit social-level analysis and intervention design into this hegemonic mode of inquiry in HIV/AIDS research, which today is characterized by: (1) a focus on the individual as the unit of analysis, particularly with respect to behavior change and disease outcomes; (2) the valorization of the RCT as the gold standard—and thus, the only acceptable method—of intervention testing and evaluation; and (3) the imperative to demonstrate disease outcomes—specifically HIV incidence, not just behavior change or social change—from HIV prevention studies.
As sociologists well know, the complex social phenomena that fuel HIV transmission—such as gender relations, sexual violence, economic inequality and instability, and violations of human rights—cannot be reduced to a handful of variables that can easily be modified or controlled for testing in experimental designs. It is precisely the confounding of people’s individual, community, and social characteristics that explains the problem of HIV transmission in the first place.
Moreover, social interventions address outcomes that often are perceived as at best intermediate (behavior change) but more likely distal (poverty alleviation, gender equity, educational attainment), rather than the proximate outcome of ultimate interest in the pandemic (HIV incidence). So, these approaches typically are derided by the biomedical community as not really being HIV prevention interventions. Additionally, when social interventions are implemented, the time lag inherent in evaluating their impact on either intermediate, distal, or proximate HIV outcomes at the population level can be significant, and funders and policy makers want to see measurable results right away. All of these factors make it difficult to design and implement social and structural interventions that will pass muster with existing funders, scientific review groups, and policy makers.
Behavioral scientists—particularly health psychologists—on the other hand, have fared well under the NIH system, as they were able early on to pull off the shelf existing behavioral change intervention models from other areas of health promotion and disease prevention and test them in experimental designs for their efficacy in HIV/AIDS risk reduction at the individual and small-group levels. As a result, a robust portfolio of behavioral interventions has emerged over the past 20 years, such that numerous meta-analyses of the outcomes of these interventions have been published. In 2004, the Centers for Disease Control and Prevention identified 42 behavioral interventions demonstrated to be efficacious in "well-designed studies" (i.e., using experimental methods) and is disseminating a number of these interventions for adoption by community-based organizations throughout the United States. There is no comparable set of social interventions with demonstrated effectiveness in reducing HIV-associated risk and HIV infection rates from which program managers and policy makers can choose what to implement.
Engaging social scientists in the kind of HIV prevention research needed to fill this gap will require the following:
- Strong public statements by the likes of Peter Piot that we need social science, not just social awareness, to fight AIDS;
- Recognition in the scientific and public health fields that there are multiple ways of knowing (i.e., RCTs are not the only method for ascertaining evidence; see www.sfaf.org/evidence for a robust discussion of the notion of "evidence" in HIV prevention);
- Significant funding streams for social research—both basic and intervention studies—with peer review by true peers (i.e., those with expertise in social science, not epidemiology, psychology, or clinical research);
- Concerted initiatives (e.g., by social science professional associations and foundations) to advance basic and applied social research on AIDS (the Social Science Research Council attempted to do this but got very little interest from funders); and
- Cross-training in bio-psycho-social science at the graduate level, to facilitate the development of multi-dimensional theories, models, and methods to address the multi-dimensional nature of AIDS.
The field of HIV/AIDS is finally moving toward a more societal-level orientation for a host of reasons, including the recognition that even when efficacious biomedical or behavioral interventions are developed, their impact at a population level depends on people’s access, acceptability, and use under real-world conditions that are highly influenced (if not determined) by social forces and structures. Given this trend, it is imperative that more sociologists engage in AIDS research to help define and operationalize social-structural interventions to maximize uptake of effective therapeutic and preventive tools, and, even more importantly, to address the social forces that fuel the epidemic and create the need for such tools in the first place.
* Judith D. Auerbach, is Deputy Executive Director for Science and Public Policy at the San Francisco AIDS Foundation. From 1995 to 2003, she served as Director of the Behavioral and Social Science Program in the NIH Office of AIDS Research.