plinary intersection, and it is these agencies that are the primary focus of this committee’s analysis and recommendations. Ultimately, it is the program managers within these agencies—who support research activities within the organization or fund external research at academic institutions, government laboratories, and state agencies—that must be convinced of the value of promoting collaborative research at the intersection of earth science and public health and be provided with recommendations concerning priority research areas and the optimum means for facilitating collaborative research.
All of the cultural barriers that exist in academic institutions have their parallels in funding agencies. In practice, the peer review systems used by most agencies mean that each additional disciplinary area involved in a particular proposal in effect equates to an additional hurdle that must be surpassed, so that more restricted proposals are widely understood to have the highest chance of success. The metrics for assessing the success of the programs overseen by individual managers are almost always based on individual program funding levels and provide little incentive for valuing the “discounted research” that can result from partial funding of cross-program activities. On the broadest scale, the fact that the appropriation levels for these agencies are determined by a range of congressional appropriations committees—which themselves have no intrinsic incentive to promote cross-committee activity—means that a top-down insistence on interdisciplinary research is unlikely. While there is broad, and even in some agencies a pervasive recognition of the merits of interdisciplinary research, the existence of barriers and the lack of incentives mean that the focus here must be on suggesting practical measures, with incentives, that can be taken by individual program managers and their superiors.
To date, support for collaborative research between the relevant agencies has been limited, although increasing (but still minor) support for collaborative research within specific funding agencies suggests that models for such research are being developed, and experiences gained, that should be applicable to cross-agency collaboration. Existing collaborative research models include NSF-based programs such as the Integrative Graduate Education and Research Traineeship (IGERT) and Biocomplexity in the Environment initiatives, and NIH-based collaborative programs such as the NIH Road Map for Interdisciplinary Research.
The classic risk assessment paradigm (see Box 8.3) provides an illustration of the need for cross-agency research support. Two of the key factors in risk characterization are exposure assessment and dose response. Historically, funding for exposure assessment involving environmental science research came from NSF or EPA, whereas funding for dose-response studies came from NIH. It seems obvious that such risk character-