Last updated: October 01, 2012
Studying Interactions Among Social, Behavioral and Genetic Factors in Health
A Workshop Presented By
The Office of Behavioral and Social Sciences Research
The National Human Genome Research Institute
October 31, 2001
Summary of Major Recommendations
Establish a long-term research-planning group
Given multiple needs such as the development of interdisciplinary concepts and research teams, the participants recommended the formation of a planning group following the model of the John D. and Catherine T. MacArthur Foundation, such as its Network on Mind-Body Interactions. Such a planning committee would consist of a core group of experts who would meet multiple times per year over a period of several years. As needed, additional ad hoc experts would be invited to participate in the planning group's activities. The planning committee would identify areas ready for research and recommend means for overcoming methodological or resource barriers to stimulating and conducting research. The planning committee might also recommend convening workshops or conferences aimed at a wider audience. (Alternatively, such conferences might be organized independently from the research-planning group.) The underlying premise for organizing such a group is that sustained interchanges among experts in diverse fields is necessary to stimulate research across the genetic, behavioral and social sciences.
Explore "piggy-backing" on existing research projects
Several pre-existing (longitudinal or cohort) studies of either socio-behavioral or genetic factors in health might be useful vehicles for studying interactions among these classes of variables through the addition of appropriate measures. The Add Health study is an example of a large, nationally representative research endeavor where genetic information could be added to planned data collection waves and then analyzed in light of already collected behavioral, social and biological information. Other datasets have other advantages with different challenges (e.g., National Longitudinal Survey of Youth has extensive longitudinal structure showing aging from adolescence into adulthood). Before embracing such supplementations, criteria need to be developed for the conditions under which they would be beneficial and worthwhile as well as conditions where new, specially designed studies would be more appropriate.
Identify "high-payoff" approaches to research
It would be useful to have some good success stories, as these will attract others to the field. What strategies are most likely to produce "breakthroughs" most quickly? For example, should initial efforts concentrate on health conditions that are known already to be primarily genetically determined (e.g., Huntington's disease), but without a known behavioral-social component? Or vice versa? Or should initial efforts be focused on health conditions with already specified behavioral/social and genetic influences (e.g., substance abuse, HIV)? Consideration should also be given to the value of cross-national or cross-cultural studies that offer opportunities for quasi-experiments (e.g., providing greater variance in social environments than might occur within a given country or replication of findings across differing genetic or environmental backgrounds).
Explore the utility of animal vs. human research subjects
For which kinds of research questions would animal studies be useful? Are there good animal models for genetic and/or behavioral-social processes to use? Are new animal models needed? To what extent can results be generalized from animal models to humans?
Specify the meaning of behavioral and social environments
Conceptual and methodological work is needed to define and standardize measures of social environment in various contexts to make it easier to compare studies. Which aspects or dimensions of the social environment and at which level of analysis (e.g., society, social institution, small group, dyad) should be included and measured? We need to consider both immediate and. cumulative (over long periods of time; life course) effects of interactions among behaviors, environments and genes. Can the environment as culture be meaningfully disaggregated into measurable components? Perhaps the configuration or Gestalt needs to be taken into account. A further complexity may be that the meaning of social variables may depend upon the social context in which they operate (the "same" variable, e.g., education, may exert influence through different pathways or mechanisms). To what extent is environment an independent variable and to what extent do people select their environment to match their genetic traits?
Provide interdisciplinary training
Attention needs to be given to identifying particular skills and knowledge necessary for engaging in research at the interface between behavioral/social and genetic sciences. What kinds of researchers are needed and how can they best be trained? Are there particular social institutional arrangements (e.g., multidisciplinary research centers) that are required to provide training and to sustain research in these areas? When should such training begin? At the undergraduate, graduate or post-doctoral/professional levels? At medical schools and/or graduate schools? How effective would short-term training institutes be? At whom should they be aimed? New investigators or more senior scholars? The National Insitutes of Health (NIH) should consider the use of interdisciplinary K awards. Ultimately, consideration should be given to how to integrate the gene-environment perspective into curricula (e.g., medical schools).
Identify infrastructure requirements
Similar to interdisciplinary training, the issue here is to consider the kinds of institutional arrangements and resources that are necessary to foster research in these areas. For example, should data analytic, genetic, or social/behavioral science cores be established to provide substantive, methodological or statistical expertise? How can knowledge and expertise gained through research be disseminated to other researchers and, ultimately, into clinical practitioners? How can longitudinal, multidisciplinary studies be established and maintained? Should standards be established for data release and sharing? Would the existence of well-characterized (both genetically and social-behaviorally) pools of volunteer research participants facilitate research? (Attention would need to be given to privacy issues.) What is the best way to assure adequate and appropriate review of interdisciplinary research and training grant applications to the NIH?
Develop methodological primers ("methods tool-kit")
Papers on appropriate methods and data analytic strategies for research at the interface of genes and environments need to be developed and disseminated. Such primers would offer guidance to state-of-the-science conceptual models, measurement instruments, data collection strategies, and analysis techniques. Such primers might include examples of successful interdisciplinary research as well as indicate anticipated difficulties.
Expand research on ethical, legal and social implications
Research on gene-environment interactions is bound to raise ethical, legal and social questions (e.g., how "race" is defined and included in studies; stigmatization of groups) that will need to be addressed. How can we maintain adequate confidentially when genetic information is linked to detailed personal data? How do individuals and groups use scientific or clinical knowledge to make policy or personal decisions? What are the unintended consequences of improved knowledge of gene-environment interactions? How should informed consent be obtained for "add-on" of genetic topics to larger studies? How might such additions affect willingness to participate in studies?