Analysis of Agency Research Portfolios
As one of its first tasks, the committee completed an analysis of the research project summaries provided by the National Institutes of Health, the Centers for Disease Control and Prevention, and the Environmental Protection Agency. The analysis was in two parts: an initial assortment into six categories of research being conducted by each agency, followed by a more subjective analysis of each research category across the agencies. These analyses were conducted after the assignment of every project to one of six descriptive categories. The categories and their definitions are as follows:
Environmental exposures (EE), defined as studies or activities that examine specific agents in the environment and their impact on health.
Endocrinology (END), defined as studies that examine the impact of environmental exposures on the hormonal/endocrine systems. This category also includes the impact of specific agents on the endocrine systems.
Long-term chronic care (LTCC), defined as studies that assess the impact of long-term environmental exposures on specific diseases or disabilities.
Molecular biology/basic science (MB/BS), defined as studies that focus on environmental exposures at the molecular level; includes animal studies.
Epidemiology (EPI), defined as studies that examine the impact of environmental exposures on populations.
Prevention (PREV), defined as studies or activities that have as their primary goal to prevent the adverse health consequences of environmental exposures in populations.
The purpose of this analysis was to obtain an overview of research areas supported or not supported by federal agencies. The results of this analysis were
factored into the process of selection of topics for the workshop and the identification of issues for the priority recommendations.
Analysis of Agency Projects
The National Institutes of Health (NIH) provided 121 project summaries; the Centers for Disease Control and Prevention (CDC), 41 summaries; and the Environmental Protection Agency (EPA), 15 summaries for a total of 177 project summaries reviewed. The relative proportion of each category in each of the agencies is shown in Figure A-1. As expected, the types of projects vary by agency and are related to the mission of each.
Analysis of Portfolios by Subject Category
A second, subjective analysis was done based on sorting the projects into the six categories listed above. The results and observations by the committee are below.
Areas highlighted in the portfolios. Research areas represented in the portfolios provided by the agencies include work associated with some of the epidemiological studies from family exposure research; men and women are included. In others, exposure measurements were combined with epidemiological studies as, for example, in the animal models of chemically induced endometriosis. Also, the portfolios contain studies on actual or theoretical biomarkers of exposure.
Gaps/missing areas of research. Few studies in this category deal with environmental exposures and women's health. None of the studies covers the entire lifespan of women: that is, environmental exposures in utero and in children, young women, women of childbearing age, and postmenopausal women. A corollary set of studies would be those that focused on exposures during critical times throughout the life span and during development. An area of increasing interest will be studies on frail elderly women in their 80s and 90s. Weight loss gradually occurs among women in their 80s and 90s, and loss of fat cells exposes their bodies to any toxicants stored in those cells. The same scenario is possible as bone mass is depleted.
Not many studies include environmental exposure measurements, especially in epidemiological studies. In most studies reviewed by the committee, exposure is considered as a variable but not as the main focus. Also, studies are needed that describe the nature of exposure settings (i.e., home, workplace, etc.) in order to develop new treatments or polices for intervention.
Areas highlighted in the portfolios. This category also included the impact of specific agents on the endocrine system. Endocrine-related research in the agency portfolios included a disproportionate amount of research on breast cancer, breast development, and hormonal breast regulation. Pregnancy outcomes, placental effects, marker enzymes, genes, in utero exposures, and the effects on female offspring were also covered in a number of studies. The field of biomarkers is expanding, particularly as it involves drug-metabolizing enzymes. While the research studies on marker enzymes are deemed important by the committee, there is a need to focus research on the importance of several indicators, which, if they occur together, constitute a more significant clinical marker.
The committee was encouraged by the number of ongoing projects focused on exposures to the uterus and the endometrium. Uterine research is important, given the magnitude of the problem of endometriosis and the fact that animal models for the condition are difficult to obtain. (It was noted that the National Institute on Child Health and Human Development recently announced a request for proposals for endometriosis research.)
Projects focused on chemicals study the usual dioxins, PAHs, PCBs, and endocrine disrupters. A number of ongoing projects address the effects of these chemicals at critical times during a woman's development. The committee was impressed by the number and breadth of the projects and the number of chemicals being studied in relationship to the endocrine system.
Gaps/missing areas of research. The following areas were identified as deserving of future work:
Solvents: Except for dry cleaning solvents and the usual top half-dozen or so major chemicals of concern, studies on solvents were not well represented in the portfolios.
Neuroendocrine research: A focus on women, depression, and disease was not well represented.
Metals: The spectrum of metals being studied needs to be broadened.
Immunopathology: Of particular importance are diseases with an immunopathological aspect, such as scleroderma (also related to exposure to solvents).
Long-Term Chronic Care
Areas highlighted in the portfolios. While there was only one project in this category, the committee noted that other projects had aspects of this issue contained within them.
Gaps/missing areas of research. All future research in this area should address multiple age groups, unless there is a research issue that occurs during a critical age range.
Molecular Biology/Basic Science
Areas highlighted in the portfolios. The committee noted that basic research studies posed questions about toxicology; however, fewer toxicological studies were present that studied the effects of compounds on biological systems. This may be because studies on whole animals, exposed to teratogens, are difficult to interpret. The committee found that the portfolios in this category were heavily weighted towards research on breast cancer. They also noted a major emphasis on endocrine disrupters, disproportionate (in the committee's opinion) to the diseases caused by those compounds.
Gaps/missing areas of research. The committee noted that recent basic science research findings have expanded the questions posed that need to be answered through toxicologic studies. There are few examples where toxicologists designed mechanistically-oriented studies to provide a more detailed understanding of the effects on systems of environmental exposures. The committee identified several areas for future research. The first is for research on mechanistic approaches, which ask questions about pathways and targets for disruption, as opposed to the toxicants themselves. There is a need for greater interest in the underlying biology and in using compounds to ask questions about biology. Other future needs include a more comprehensive approach to understanding biological mechanisms and how toxicants alter biological functions. In summary, more research is needed which takes advantage of recent advances in molecular and cell biology to address how toxic compounds alter normal physiological processes, including gender-dependent targets.
The committee indicated that current compound-specific research has a somewhat narrow toxicological focus; the committee, however, advocates a broader biological approach. In addition, there is growing interest in the molecular imprinting process that may occur as a result of perinatal exposures. The committee believes this area deserves more thought and attention and considers toxicological questions to be tied to the basic biological questions. Therefore, a concerted effort to conduct research on the molecular imprinting process in perinatal exposures would be productive for the whole field.
Many other organs and organ systems, other than breast development and regulation, need attention. There are also many serious questions in breast research other than those concerned with cancer. The portfolios did not contain much information on the role of dietary effects, particularly regarding phytoestrogens.
The committee noted also many opportunities for the use of animal models that express human genes and those that control genes using human gene regulatory elements. Also, there was not a significant cohort of long-term
molecular studies in animals. Such studies were not represented in the portfolio, even though they may provide the basis for answering important questions about toxicants in systems utilizing genetic technology.
Areas highlighted in the portfolios. The committee commented that beyond the projects placed within the epidemiology category, there were an equal number of project summaries in other categories that had epidemiological components.
Gaps/missing areas of research. The committee noted that the portfolios contained very few projects concerned with environmental exposure. Much is known about the relationship of exposures to disease in women which could lead to prevention. Of critical importance would be new research to describe the nature of exposure settings, in order to develop appropriate interventions. The committee referred to the missing component as "hazard surveillance" or "hazard epidemiology." These type of studies may not be typical for biologically-based agencies, but are important links to biological measurements and analyses. Other missing aspects include differential risks of exposure according to women's work and women's lifestyles, whether or not they are unique to women.
Another area needing attention is a careful assessment of the impact of multiple exposures. Few studies focus on one substance; many focus on multiple substances without understanding the difficulty of such studies. Recent data suggest the need for a more careful look at the synergistic effect of multiple exposures, as it relates to organochlorines and estrogens.
Some windows of opportunity were noted, primarily by the CDC projects on the magnitude of risk. There are some pharmacokinetic, pharmacodynamic, and biological exposure models used in epidemiology that do not currently include gender. The committee suggested that the importance gender may play in these models is inadequately examined and deserves a more intense focus by investigators who choose to develop and examine these models. Statistical methods are another area in need of support in epidemiology, in general, and epidemiological studies of women's risks, in particular.
A series of data systems is needed to facilitate both analysis and surveillance. These systems could be used to identify risks, to track risks, and to track the effects of intervention on risks. Most of the NCHS data are used by NCHS descriptively. The committee views the NCHS database as an impressive resource that is not being fully utilized. More knowledge can be leveraged from these data because of the risk information that is collected. Beyond national databases, the committee observed that there are not many population-based studies, longitudinal studies, or cohort studies in the portfolio. The few that were highlighted were internal studies identified as part of the NIH portfolio.
While important work should continue on chemical toxicology and its relationship to women's health, other work needs to be done with a focus not on chemicals, but on psychosocial effects and risks (e.g., on cardiovascular disease). The committee pointed out that the effects of behavior on biology and disease induction are a complex but critical area for future research. Some studies of the cardiovascular demand/control model with a social support component suggest that the combination of demand, control, and support is more important than cigarette smoking in the etiology of cardiovascular disease. The committee urged researchers to look for important risk factors in cardiovascular disease, even though the technology of exposure assessment is complex and utilizes behavioral science measures. Job demand and job control in women versus men was also not among the studies reviewed. The committee noted that Europeans seem more interested in exploring these issues; but even the European studies are controlling for gender rather than focusing on gender.
Finally, the group noted that there is a need to look at environmental allergic disease and ascertain in what ways women are different from men, in order to develop hypotheses for intervention.
Areas highlighted in the portfolios. The committee noted that the prevention portfolio was small; the bulk of it is in CDC. (The committee wondered whether a broader search of the NIH database would have identified any additional prevention projects).
The committee noted that the CDC portfolio generally focused on population-based models of prevention, where either public information or advertising was used in some way so that the media transmitted the information about prevention. Because the analysis of the EPA portfolio did not result in the identification of projects in this area, it was suggested that a definition of prevention linked to regulation might uncover additional EPA prevention research. Comments by EPA representatives indicated that only those projects with a strong research component or women's base were submitted, and that other activities were not included.
Gaps/missing areas of research. The committee suggested that there undoubtedly were more projects in EPA and NIH that benefit society as a whole and not just women. Therefore, the committee urges against the interpretation that there is a lack of prevention-related research in EPA or NIH. Of the other projects reviewed, only one project had risk assessment in the title or in the abstract; it raised the question of the agencies' view of prevention research and the operative definitions and models for prevention research across the agencies. Policy-related research, such as that informing changes in the use of toxic substances or uses of engineering controls, was not evident in the portfolios.
Nonetheless, the committee suggested that it was not surprised that the prevention area is relatively small. The committee endorsed the notion that rushing ahead with preventive interventions without fully understanding the biological effects of these interventions is not good. Prevention research should focus on individual behavioral change, an emphasis on regulation, and widespread dissemination of protective measures that people themselves can employ to limit exposure.
The committee noted that much of the ongoing preventive activities are not gender-specific; opportunities for examining gender-specific factors would be useful and would improve the potential for prevention. Such opportunities should include different gender-specific learning practices or behaviors. The committee also suggested taking advantage of the behavioral science literature that deals with preventive strategies.
In conclusion, the committee suggested that the overall research portfolio is not balanced between exposures and the presumed mechanisms that affect our health. This is particularly true for exposures relevant to women who work. As for epidemiology, some mainstream research issues are being addressed, but some critical issues are missing from the portfolio.
Discussions with the agency representatives revealed three factors that should also modify the interpretation of these analyses:
Definitions of terms are not necessarily consistent across the agencies. Prevention and preventive strategies for prevention, for example, may be categorized by one agency as prevention and by another as intervention. Intervention may occur with a specific drug, but it may also be preventive strategy.
Other research, not particularly focused on women's health, may nevertheless be relevant. This may be the cause of an underestimate of the complete portfolio that is directly applicable to the task.
Research from other components of the agencies may not be represented in the project listings. Because a limited number of NIH components are participating in this study, other relevant work in other components may also be relevant to the task.
These caveats, taken together, suggest that this analysis is, at best, an underestimate of the work that is being done and is a subset of research activities in the agencies. Nonetheless, a review of the projects identified thematic areas in which substantial research is being conducted, as well as areas in which there are gaps in research and opportunity. The committee suggested that the agencies review their research portfolios using identical definitions, expanded to include all possible adverse influences on women's health. Following this suggested expansion, an assessment of the relevant research portfolios should be carried out again.