NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on Gender Differences in Susceptibility to Environmental Factors; Setlow VP, Lawson CE, Woods NF, editors. Gender Differences in Susceptibility to Environmental Factors: A Priority Assessment: Workshop Report. Washington (DC): National Academies Press (US); 1998.

Cover of Gender Differences in Susceptibility to Environmental Factors

Gender Differences in Susceptibility to Environmental Factors: A Priority Assessment: Workshop Report.

Show details

Workshop Report

A variety of sources, such as morbidity and mortality data and health care utilization data, point to differences in health status between men and women. Some of these distinctions are thought to be associated also with race, ethnicity, and/or socioeconomic status. While critically important, these factors are not the subject of this report. Other distinctions are thought to be solely based on sex,1 but there is growing awareness that the environment and environmental factors may play a role in creating health status differences between men and women. Various factors, such as genetics and hormones, may account for gender differences in susceptibility to environmental factors. In the development of approaches to disease prevention and health promotion, to behavioral and medical interventions, or to the initiation of research strategies, many have come to realize that special consideration must be given to health effects that are either gender-specific to or are overrepresented in women because of environmental factors such as occupation, behavior, lifestyle, hobbies, reproductive status, or physical activity. This latter series of issues is the focus of this report.

Many activities have been initiated on issues concerning women's health, and many federal agencies now have programs to address various aspects of health outcomes in women. However, there is ample room for newer opportunities for coordination and prioritization of research to answer questions about sex differences in susceptibility to environmental factors or gender variation in disease expression. Identification and clarification of the real or perceived gaps in knowledge may assist policymakers in planning future research initiatives and in interagency coordination.

Thus, in 1996, the Office for Research on Women's Health, National Institutes of Health (NIH), requested that the Institute of Medicine (IOM) conduct a workshop study to review the current research programs of the National Institutes of Health (NIH), the Centers for Disease Control and Prevention (CDC), and the Environmental Protection Agency (EPA) that are devoted to women's health. The purpose of this activity was to identify the state of knowledge regarding gender differences in susceptibility to environmental factors and make recommendations about promising areas of inquiry that may profit from interagency coordination.

In response to this request, the IOM formed a Committee on Gender Differences in Susceptibility to Environmental Factors. The committee included experts in environmental and occupational health and medicine, health sciences policy, epidemiology and public health, risk assessment, endocrinology, immunology, toxicology, and women's health. The committee met twice during the course of the study and held a scientific workshop in May 1997.


The first meeting of the committee had three purposes:

  1. Review the charge of the committee. The phrase “gender differences” implies assessing differences and similarities between men and women; however, discussions with relevant agency representatives indicated a need to focus on women's unique susceptibility. Therefore, the committee refined its charge to focus on this latter aspect: that is, the identification of areas in which research and policy initiatives could address women's differential susceptibility to environmental factors.
  2. Review the existing research related to the topic of the study. The committee reviewed a variety of materials related to environmental health research, gender differences, and environmental susceptibility. An abbreviated list of these materials, resources, and sources of information is in Table 1.
    Based on a review of the literature and of relevant IOM/NRC reports (see Table 2), the committee understood that substantial research work is being done in environmental health research that would have an impact on understanding gender differences and environmental susceptibilities. Key areas of research include work on environmental estrogens, multiple chemical exposures, gender differences in response to toxic substances, allergens, and autoimmune and other immune responses to environmental factors. However, most of this work does not focus on gender differences.
    Against this broader context of research, the committee reviewed the missions, programs, and strategic plans of the sponsoring agencies. Finally, the committee reviewed abstracts of the sponsoring agencies' research portfolios. The goal of this review was to understand what research was currently being conducted and identify areas in which future research would be useful (Appendix A).
  3. Develop an agenda for the workshop. The committee began to outline the areas from which to gather more information. To do this, the committee and staff identified other experts who would provide presentations on various related topics and promising areas of research. Relevant data sought by the committee revealed gaps in knowledge or offered new information that displayed specific gender differences in disease initiation, progression, or outcome.
TABLE 1. Types of Information Received from Sponsoring Agencies and Reviewed by the Committee.


Types of Information Received from Sponsoring Agencies and Reviewed by the Committee.

TABLE 2. NRC Reports with Information and Recommendations Related to Environmental Health Research and Training.


NRC Reports with Information and Recommendations Related to Environmental Health Research and Training.


Because of the extensive information reviewed at the organizational meeting, the committee designed the workshop to review broad aspects of environmental exposure that would be common among women. These exposures were thought to occur in a variety of ways: in different settings (e.g., the home, the workplace); through different routes (e.g., foods), because of different activities (e.g., societal roles, chores, hobbies), or because of unique or critical times in the lifespan. A second design approach for the workshop was to review examples of current research to examine how patterns of susceptibility or differential exposure may be viewed and understood at the level of the individual or the molecular level.

The workshop was held in May 1997. (See Appendixes B and D for a summary of the presentations and the agenda of the meeting.) It was designed to answer the three questions that comprised the charge to the group:

  1. What areas within the existing portfolios of the sponsors are likely to yield information appropriate to this topic? What are the gaps in knowledge that warrant future research?
  2. Are there research strategies and priorities for addressing the gaps in knowledge?
  3. What other strategies, including interagency coordination, might improve the prospect of developing knowledge that will identify gender differences in susceptibility to environmental factors?

The workshop was composed of two panels (see Appendix D). The first panel examined the overall issue of patterns of exposure among women. Presentations included issues related to environmental exposure in the workplace, environmental exposure and nutrition, and multiple environmental exposures over a woman's lifespan. The second panel focused on patterns of susceptibility, with presentations on various clinical and basic research studies. Presentations included epidemiology, gender, and environmental influences on multiple sclerosis; estrogen receptor knockout mouse studies and implications for hormonal differences in susceptibility; gender differences in metabolism and susceptibility to environmental exposures; and gender differences in the occurrence of molecular markers of carcinogenesis. The final panel included speakers and invited participants who discussed how current information is applicable to the three questions that formed the task. Part of the discussion was utilized as an opportunity to examine issues and obstacles to understanding women's differential susceptibility: recognizing variables in research on gender or sex differences; data collection, utilization, and analysis; understanding the relationships among exposure, dose, and effect; and the role of social factors in contributing to health and disease. The final portion of the discussion focused on a review of current federal efforts and resources and the creation of newer opportunities for collaboration among all federal agencies. Subsequent to the workshop, the committee met to discuss the results of the meeting and to outline areas for recommendations and priorities.

This report provides highlights of the workshop and a summary analysis of the research portfolios provided to the committee by the sponsoring agencies. Therefore, the committee's conclusions and recommendations regarding future research and policy directions for this area are based on the review of all materials at the organizing meeting, the analysis of agency portfolios (Appendix A), and the presentations and discussion at the workshop (Appendix B).


The summaries of the formal presentations at the workshop are in Appendix B. Below is a summary of the discussion by the committee, panelists, and workshop participants; it is based on the presentations and subsequent questions raised by those attending the workshop. The group was asked to identify issues related to the three parts of the statement of task and to suggest strategies for interagency coordination.

Discussion Points from the Presentations

One important issue is that of a woman's multiple exposures: that is, in the workplace as well as in the home, and during childrearing and caregiving for elders. For this reason, many participants suggested that exposure histories should be designed to collect more information about women's total experiences and exposures at different points in the lifespan. The group noted that there is growing scientific acceptance of the notion of differential susceptibility, but regulatory standards are, for the most part, still based on data averaged from male populations or experimental animals.

Weight gain and loss is another important issue. Women experience more cycles of fat gain and loss because of dieting behavior and also because of natural phenomena such as pregnancy. If toxicants stored in fat tissue are mobilized during these periods, this could be a significant behavioral/nutritional factor with differential impacts based on gender. Similarly, questions regarding neuroendocrine factors and stress deserve further attention. Regulatory changes may also have an impact on exposures for both men and women. For example, the Delaney Clause, a 1958 amendment to section 409 of the Federal Food, Drug, and Cosmetic Act of 1938, established a zero tolerance for pesticide residues and other known carcinogens in processed food. However, that clause is no longer in effect; therefore, there may be reason to be concerned about the level of toxicants found in food.

Research on “gender differences” and “sex differences” is often conducted and referred to as though they are the same; however, “sex differences” often serves as a proxy for cultural and socioeconomic variables that have little to do with biological differences between men and women. The group contended that the scientific community must do a better job of identifying biological differences in susceptibility, on the one hand, and nonbiological variables, on the other.

These other points were also raised during the workshop discussion:

  • the rising proportion of women who are postmenopausal;
  • the need to look at the role that infectious agents, emerging infections, repeated exposure to childhood infections, and differential exposures among childcare workers (the majority of whom are women) have on differential health outcomes between populations of men and women;
  • ultraviolet radiation (men are more likely to work outdoors, but women are more likely to sunbathe);
  • dietary and environmental estrogens as variables of the hormonal cycle; also, changes in dietary behavior that increase exposure to phytoestrogens;
  • methods for detoxifying the body, such as chelation agents to bind and remove a toxicant, particularly if the detoxifying process makes once-stored agents more bioavailable;
  • the role that race, ethnicity, and culture may have in establishing differential exposures between men and women of a given subpopulation or between women of different racial, ethnic, or cultural groups;
  • accessibility of the raw data from the National Health and Nutrition Examination Survey-III (NHANES-III); these data recently have been made available to researchers.

There was a striking contrast between how the first panel addressed the issue of gender differences, in terms of global, societal, and cultural issues, and the second panel's approach, which focused on biological mechanisms. The group suggested that multidisciplinary research that included consideration of social and cultural factors was needed. Clearly one priority that emerged was the question of multiple or combinational exposures; another was the search for linkages between animal studies and clinical studies. However, at least one participant questioned whether the kind of multidisciplinary studies that would be needed are actually possible, given the fact that researchers from various disciplines would be looking at different points on the causal pathway, measuring with different tools, and obtaining different results.

The group concluded that it is important to gain a better understanding of basic biological mechanisms of pathways leading from exposures to health effects. They also indicated it was important to understand exposures, especially multiple exposures. There was general agreement that the potential exists for more synergy between basic scientists and epidemiologists than either group currently realizes. It was suggested that basic researchers could provide guidance on how to stratify an epidemiological analysis based on biologic factors, and epidemiologists can identify cohorts with specific exposures for further research on such issues as genetic variability. Issues suggested for further understanding include ways to shorten the time lag between basic discoveries, the applicability of laboratory findings in the clinical or epidemiological setting, and the availability of cheap, accurate diagnostic tools for monitoring exposures. In addition, the group pointed out that one desired result of exposure studies is to find ways to prevent or reduce those exposures.

As acute exposures with marked effects are eliminated, the chronic low-level exposures to other diverse factors may become more important. This suggests that nutritional interventions are also worthy of attention, particularly in view of the rising level of obesity in the United States. The group expressed the need for a systemic approach that looks at all these factors in women.

Several participants pointed to the need to adequately fund the fourth iteration of the National Health and Nutrition Examination Survey (NHANES-IV), to be conducted by the National Center for Health Statistics (NCHS) of the CDC. That is an efficient way to achieve comprehensive answers to some of the dietary questions. Money for planning NHANES-IV was known to be available—in fact, that planning is more or less complete—but no money is currently available for follow-up studies on NHANES-III. It was suggested that part of the reason is that follow-up studies would be based on secondary data, and the general opinion was that follow-up proposals do not fare well in the NIH peer review process. The group suggested that there is a need for coordination among agencies that fund such studies as well as among data collection systems generally. Several participants asked for improvement in the NHANES-IV measures of socioeconomic status (SES), occupational data, and environmental exposure measures. Others felt that researchers should be encouraged to submit proposals for secondary analyses of existing NHANES data that include testable hypotheses.

Opportunities for Agency Collaboration

A part of the discussion on existing federal resources and opportunities for interagency collaboration focused on several repositories of tissue and serum samples from earlier NHANES studies, including the availability of anonymous DNA samples from NHANES-III. The group noted that while NCHS had taken some steps to solicit the scientific community's input on the content of NHANES-IV, there was a need for additional input in the design of NHANESV, perhaps through the mechanism of a broader advisory group.

Participants suggested that federal agencies should fund investigators to conduct a variety of secondary analyses that would integrate critical questions concerning gender and susceptibility, utilizing existing data sources; but they also wondered about the source of funding for such integrative studies.

A suggestion was made that one agency should centralize the resources that could be used by the entire research community. The National Center for Research Resources at NIH might be a candidate for this role. Opportunities for public-private cooperation to fund research that is too expensive for either sector alone were suggested as another possible option. Industry consortia have collaborated with EPA in this way, and both EPA and USDA have done cooperative research and development agreements with industry, university, and nonprofit researchers. Participants pointed out that CDC and NIH are perceived to be open to these cooperative efforts; the National Institute of Occupational Safety and Health (NIOSH) was highlighted as having industry staff working in its labs.

In areas where regulation would be a principal outcome, it was viewed as desirable to have a protocol workshop that would design a clinical or epidemiological study acceptable to all stakeholders. Another initiative cited was the development of core components on exposure and diet that were developed for the National Action Plan on Breast Cancer. The group suggested that consensus and protocol workshops are more common in science than in regulation; they emphasized that science should inform regulation.

In concluding the discussion, participants suggested that, in these and other areas, “environment” should be taken to mean more than just chemicals. Biomechanical stress, noise, and even violence in the workplace should also be considered as factors that belong on the list of environmental priorities.



Throughout the course of this activity, one area of confusion centered around the definition of terms. From discussions with the sponsors to discussions among the committee and the invited workshop speakers, it became clear that key terms needed to be defined. For the purpose of promoting useful interagency research strategies additional factors should be included to capture aspects that are related to differences in susceptibility to environmental factors (Table 3). Environmental exposures often implicate biological, chemical, and physical agents. However, in addressing the issue of gender differences in susceptibility to environmental exposures, the inclusion of nutritional, ergonomic (both biomechanical and psychosocial), and behavioral factors as well as specific places such as the workplace and the home in the concept of “environmental” provides a broader base for this type of research. Similarly, the definition of “susceptibility” could include all adverse outcomes except those that are directly related to hypersensitivity or allergy. Finally, the definition of “gender differences” can include aspects that are sophisticated enough to separate genetic and physiobiological differences between men and women from differences in environmental exposure, which in many cases result from the independent and interactive effects of socioeconomic status, employment patterns, and family, role and cultural experiences—sometimes called inherent vs. extrinsic differences.

TABLE 3. Additional Factors to Include in Broader Definitions.


Additional Factors to Include in Broader Definitions.

The discussions involving speakers and workshop participants provided an important step in understanding how to (1) broaden the scope of review on gender differences and susceptibility, (2) identify areas in which there is or is not consensus, and (3) identify issues that need further attention. So valuable was this cross-agency and cross-disciplinary discourse that the committee concluded that there is a need for such a workshop at regular intervals in order to monitor progress and refine priorities for research.

Preliminary review of the research portfolios provided by the sponsoring agencies suggests that they were developed on the basis of the standard definitions. In light of the additional factors identified by the committee, the research portfolios provided by the federal agencies comprise no more than a small subset of activities that the committee believes should be pursued. Although they appear to have been compiled according to divergent criteria, most of the enumerated projects were focused on chemical exposures and reproductive system outcomes.

The committee urges the agencies to expand the review of their research portfolios, using the additional factors to augment the definitions , in order to identify all possible influences upon women's health in an effort to prevent adverse health outcomes. Following this suggested expansion, an assessment of the relevant research portfolios should be conducted.2 The goal of a reclassification effort based on this broadened approach of environmental factors is to develop a common database that will permit the agencies collectively to look at the adequacy of their funding in areas of greatest importance and to design interagency strategies and partnerships.


Research Recommendations

It became clear in the course of this activity that research on women's differences in susceptibility to environmental factors fell into two distinct categories: (1) differences in opportunities for men and women to be exposed to environmental factors, which are often culturally determined, and (2) genetic and physiological differences between men and women, which are biological. While these categories are useful for organizing research priorities, there is also a need for research that will bridge the gap between these two aspects of human life.

I. Priorities with Regard to Research on Exposure


Research should be based on a broader inclusion of factors in the definition of “environmental exposure”.

In addressing gender differences in environmental exposures, it is important to broaden the inclusion of factors in the definition of “environmental” to include not only chemicals, physical agents, and pathogens but also nutritional, ergonomic (both biomechanical and psychosocial), and behavioral factors. In developing areas for research on the human health consequences of exposure to harmful physical and chemical agents in the environment, the relevant domains that should be evaluated are occupational and nonoccupational exposures, residence, physiological parameters, physical activities, and nutrition and diet. In the development of behavioral and medical interventions or the initiation of interagency research strategies and initiatives, special consideration must be given to gender-specific outcomes determined by occupation, behavior, lifestyle, hobbies, reproductive status, and physical activities. Examples of these various factors include but are not limited to:

  • physical agents (e.g., noise, heat, vibration, ionizing radiation, pressure)
  • pathogens (e.g., tuberculosis, HIV, cold viruses, Candida)
  • nutrition (e.g., calorie intake, vitamin intake, ratio of fat in diet)
  • biomechanical (e.g., sustained awkward postures, frequent repetitive movements, work requiring exceptional force and heavy lifting)
  • psychosocial (e.g., job stress, level of demands from and control over work and life, amount of social support)
  • behavioral (e.g., substance abuse, weight fluctuation, smoking).

Population-based studies should include more complete and meaningful occupational data to help develop more accurate information on exposure.

Many studies that are population-based, those investigating specific hypotheses (e.g., Framingham Heart Study, Nurses Health Study, etc.), as well as those designed to estimate population health status (e.g., NHANES) provide an excellent opportunity to gather data on exposure and susceptibility in both men and women. These studies are very useful for these purposes and represent unique resources for data collection. However, in most instances the data do not reflect the full range of potential exposures. This data gap is exemplified by deficient occupational data. A simple category of “occupation,” for example, does not elicit useful information for a field like nursing, in which some workers are exposed to highly toxic chemotherapeutic drugs, while others are exposed to ionizing radiation, HIV and other infectious agents, ergonomic stress and anesthetic gases, violence, or perhaps a range of pediatric infections. Work histories are difficult and time-consuming to collect; hence, some compromise is needed to improve the “occupation” category. Methods for limited occupational history collection in some newer studies need to be validated.

Because national, population-based surveys are so valuable and because they are not often conducted, it is therefore paramount that the data collected be relevant and useful. The committee learned that planning stages for NHANESIV are nearing completion. The possible inclusion of many of these questions and factors would help collect comprehensive information on a variety of factors that modify health outcomes, such as occupation. The NCHS should seek guidance from a broad research community to determine how best to record relevant occupational information in national surveys. Similarly, federally funded, population-based studies should be encouraged to improve exposure information collected by survey instruments.


Occupational exposure studies should adequately characterize and account for the full range of multiple exposures.

To explore the role of multiple exposures adequately, it is necessary to look at combinations and integrative effects over the full workday, the full working career, and the full lifetime. Many working men and women have a second (or even a third) job, each of which may have its own distinct set of exposures. Additionally, most women go home to “another shift” in which they have additional exposures, such as household chemicals and the emotional demands of childrearing (Broersen, et al., 1996). The combined impact of the demands of work and home for women was illustrated in a study of Swedish Volvo factory workers. This study showed that men and women had the same level of stress on the job; but when they went home at the end of the day, the men's stress levels went down, while the women's went up (Frankenhaeuser, 1989). Other studies highlight the need to study interaction and combined exposures between home and work life (van Dormolen, et al., 1990). The issue of multiple exposures to men and women is a critical and complex problem for population studies. Studies that examine effects of long-term exposures in the working population and susceptible groups (e.g., older workers, partially disabled workers) deserve special attention.


Research should examine gender differences in susceptibility to environmental factors over the entire lifespan as well as during critical exposure periods such as fetal development.

Presentations during the workshop demonstrated that both susceptibility and severity could vary over the lifespan. For example, early pregnancy is an obvious period of susceptibility of the fetus to teratogens; exposure of the growing child is also poorly understood as to the long-term health effects later in life; place of residence prior to age 15 is a major factor in multiple sclerosis (Kurtzke and Page, 1997); and smoking before 16 is a major factor in lung and breast cancer (Devesa, et al., 1995). Adolescence, in general, is poorly understood with regard to susceptibility to environmental exposure. There is evidence from both human and animal studies that initiation of mammary carcinogenesis originates in undifferentiated structures of the mammary gland during early adulthood. This model has been extensively developed by Russo and Russo (1997). Susceptibility before and after menopause is emerging as an important issue, as the number of women over 55 continues to increase. These and many other critical periods in the lifespan need further research to link the impact of exposures upon health outcomes.


The development and use of appropriate animal models is encouraged.

Some animal studies have examined differences in physiological response between younger and older animals. It is worth noting that a national resource already exists in this area: a colony of elderly rats supported by the National Institute on Aging. Although these rats are useful for many types of studies, they are hard to use for toxicology studies. The most direct animal model for human conditions may be nonhuman primates. As recently reported, a few federally supported facilities maintain large and expensive colonies of aging chimpanzees (Roush, 1997); utilization of baboons for relevant studies on aging may be feasible. Feline, canine, and porcine models are also useful for research on menopause-related questions. Further research is needed to develop more accurate animal models for a variety of other measurers of susceptibility studies.


Studies are needed that identify the cultural and historical factors that account for the distribution of exposures between men and women.

The issue of susceptibility by gender, masking what is really different exposure by gender, needs to be clarified. In other words, historical and cultural factors may have accounted for differential health outcomes. As such, the impacts of changes in our societal norm that may or may not have long-term health consequences need to be studied. Recent changes in our society include, for example, the influx of women in the workforce in all occupations and at all levels. In like manner, men are participating in homemaking and childrearing duties. Among the additional examples of changes in our society values that may impact on health outcomes are the growing numbers of grade school children and teenagers who smoke (IOM, 1994a) and the divergence in age of first pregnancies (increasing numbers of teenage mothers and increasing numbers of women who have their first child after age 30). It would be useful for social scientists to unravel the impacts and implications of these changes on health outcomes. Equally useful would be investigations on whether men and women respond differently to the same exposures or whether they actually have different exposures. The result might be to identify a kind of “socioeconomic marker” of susceptibility, similar to the genetic markers discussed in 10, below.

II. Priorities with Regard to Basic Research


Basic research on gender and susceptibility to environmental factors should focus on (1) the biological basis for differences and (2) the contribution of environmental factors to the risk of disease from the same exposure.

Mechanistic studies of hormone-dependent and hormone-independent processes or pathways should be investigated, beginning with animal/cellular models and followed by clinical research that translates the molecular-level information into the impact on human health. Here, too, however, the broader inclusion of additional factors in definitions of “exposure” and “susceptibility” comes into play. For example, stress was cited as a complicating factor in several presentations, and there have been several studies of the effect of stress on epinephrine levels in men (see Appendix B). Yet, relatively little is known about different responses and effects in women. It is known, however, that cortical responses modulate the immune response and that women's greater immune response contributes to differences in the appearance and severity of autoimmune disease. These types of examples suggest that the result may be differences in the prevalence of certain diseases (e.g., multiple sclerosis, lupus) between men and women and among women of different races and ethnicities (e.g., lupus is more prevalent among African Americans). However, differences in prevalence may not always be a predictor of health effect. For example, although fewer men than women contract multiple sclerosis, the disease is more severe in men than women, and the 10-year mortality rate is higher.

It may be that gender differences in neuroendocrine or immune responses, as well as differential exposures, will explain some of the gender differences in chronic diseases such as diabetes. Differential susceptibilities related to gender may ultimately prove important in understanding poorly defined syndromes, such as multiple chemical sensitivity. Consequently, research should address all these dimensions of physiology—hormonal modulations as well as neuroendocrine response—and the changes that men and women undergo over their lifespans.


Priority should be given to studies of human diseases that are manifested differently in men and women or in which gender modulates susceptibility to environmental factors.

Gender plays an obvious role in susceptibility to reproductive tract disorders, which deserve continued attention among both males and females. But gender also appears to play a significant role in modulating susceptibility to nonreproductive tract disorders. One obvious example is the role of estrogen in women's susceptibility to lung cancer resulting from exposure to cigarette smoke or some compromise of autoimmunity. On the other hand, estrogen can protect against bone loss, loss of vascular function, and possibly brain degeneration. Therefore, losses or supplements of this hormone may have profound gender-specific impacts. In some cases these differences appear to protect women, as in gastric ulcers and certain infectious diseases. The genetic and physiological mechanisms underlying these differences need to be studied and understood in order to identify effective procedures for prevention, intervention, or therapy.

Other examples of major public health problems which are manifested differently in men and women, or where environmental factors are manifested differently in women, include heart disease, pulmonary disease, autoimmune disorders, mental illness, and arthritis. This list of health problems is far from complete, nor is it currently possible to set priorities among them. The identification of all such diseases and the setting of actual priorities among them should be part of any effort to set overall research priorities.


Research should examine the impact of metabolic differences between men and women, as well as neuroendocrine, immune, and hormonal differences.

There are differences in metabolic processes between genders. As indicated in the workshop summary (Appendix B), the herbicide 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) induced liver cancer in females sprague-Dawley rats, but not in males. Hormonal components are involved in this gender difference. On the other hand, there was a significant overlap and interaction between metabolism and susceptibility genes in the study of markers of exposure and susceptibility to carcinogenesis. However, we do not know if it is possible to extrapolate the effect of this toxicant upon humans. The role of gender differences in these interactions has not been clearly identified, and the whole field deserves further study.


Research should seek to characterize genetic markers of susceptibility.

Genetic polymorphisms and mutations play a potential role in ethnic as well as gender differences with regard to susceptibility to environmental factors. Research that identifies specific genes or combinations of genes that are reliable predictors of susceptibility could produce enormous benefits for both prevention and diagnosis. Once identified, the genes could be studied in cellular or animal systems in order to study mechanisms and effects. A recent new initiative by NIEHS to collect susceptibility genes for large-scale studies on how these genes vary from person to person is a major step in this area (NIEHS, 1997). Information that is developed from this project will contribute greatly toward knowledge of the genetics contribution to susceptibility to environmental factors. However, as in any research on human subjects, great emphasis must be placed on the protection and privacy of the subjects. There are many instances in which existing databases may be of use to researchers without jeopardizing the privacy of individuals. For a fuller discussion of issues related to privacy, see the IOM report, Assessing Genetic Risks (IOM, 1994b).


There is a critical need for translational research to bridge the gaps among cellular, animal, and human systems.

When epidemiologists, for example, identify a relationship between risk factors and disease that suggests gender-specific differential susceptibility, researchers then need to study the basis of that relationship at the molecular and cellular level. In addition, there is much to be learned from animal models, not only from the knockout mice (as described in the workshop summary) but also from gene insertion and substitution techniques (e.g., so-called “knock in” mice and “hit and run” experiments).

At the same time, however, new techniques are needed to validate animal models of human disorders; assistance in translating those validation techniques into simple tests that can be used in the field are equally necessary. For these reasons, basic researchers should look for the broader implications of the mechanistic research conducted at the molecular, cellular, and animal levels. In particular, sponsoring agencies should encourage the development of animal models that are directly relevant to (1) gender differences in susceptibility and severity and (2) human exposures and diseases.

III. Policy Recommendations


An annual workshop should be held to encourage and promote opportunities for interagency collaboration.

Research priorities for improving the understanding of gender susceptibility to environmental factors occur in at least three different major areas of research: worker health and safety, women's health research, and environmental health. Each area is supported by a separate funding stream. Synergy can be achieved through a careful review of both the established priorities and missions of the agencies with the cross-disciplinary research needs to identify areas of priority. One way of identifying those areas of natural congruence would be through workshops jointly sponsored by relevant agencies at regular intervals. The workshops would also help monitor progress and refine priorities for long-term studies in this area. The joint sponsorship of this IOM review was an encouraging sign. Presentations by the sponsoring agencies' representatives indicated that other jointly funded efforts are occurring also. The committee encourages these partnerships and believes that more interagency cooperation would advance the research base in this area. Therefore, the committee believes that the proposal for an annual workshop would significantly encourage the development of joint activities.

Suggestions for agenda topics for such a workshop abound. One would be to develop common definitions and categories to guide a broader review of current research activities; others would be to develop a government-wide research priority list (subject to annual revision) and invite proposals for an interagency initiative. The committee suggests neither a particular administrative structure nor a particular lead agency; rather, it encourages agencies to cooperate in setting priorities, funding extramural research, and/or conducting intramural research with multidisciplinary interests.


Agencies should work together to make necessary institutional changes.

Many of the research priorities described are interdisciplinary in nature and will require interdisciplinary peer review. Such interdisciplinary review groups are more challenging to conduct; hence, all sponsoring agencies should work together to meet the peer review challenge.

Interagency cooperation and collaboration are, however, not limited to project review. Mechanisms and programs are already in place and attempt this kind of “cross-boundary” coordination; they may provide models for future collaboration. For example, an intra-NIH advisory committee, convened by the Office of Research on Women's Health, involves all the NIH institutes. Members of the Interagency Task Force on Women and the Environment, established by the Department of Health and Human Services (HHS), come from Cabinet-level departments. Experience indicates that these groups work best when they have high-level and consistent participation.


Current sponsors should make every effort to expand the roster of agencies conducting or funding research on gender and the environment.

The scope of this project was limited to the civilian agencies that sponsored it. Nonetheless, many, if not all, research-based federal agencies have missions and programs that support various aspects of women's health and environmental health. The Department of Defense (DOD), for example, has a number of research programs that can contribute to the development of new knowledge regarding gender differences and differential susceptibility. Servicewomen are increasingly exposed to the same environmental risks as their male counterparts. This type of research could become a high priority in DOD's long-term planning. In addition, DOD's classification of jobs through its Military Occupational Specialties system could be the basis for a more specific occupational analysis than is available in most other civilian-based data sets. DOD already supports some analysis in this area; it might prove fruitful to hold discussions between DOD and NIOSH or NIEHS with regard to joint support of work-related research. DOD already has women's health programs and, therefore, has a built-in interest in gender-specific research. More importantly, DOD has a peer review system already in place; this would facilitate review of meritorious research proposals on topics of interest to this project. Another equally important partner is the Department of Veterans Affairs (VA) which will have a growing interest in gender differences and differential susceptibility as more former servicewomen become eligible for medical care in VA hospitals.


Sufficient interest and opportunity exist for agencies to invest in prospective research projects that focus on both gender differences and the environment. These investments should be flexible with regard to funding mechanisms and should provide continuity for long-term investigations.

Many of the recommended research objectives could be accomplished by giving the combined issues of gender and environment higher priority within agency budgets. However, agencies should also look for innovative opportunities to match their funding mechanisms with other types of research support. Traditional investigatior-initiated R01 grants at NIH would suffice for many research proposals; but the use of other mechanisms such as project grants, core support, subcontracts, and cooperative agreements should not be ignored. These other mechanisms can be tailored to specific agency needs and priorities and can provide support over a relatively long period of time. This long-term support is vital for continued surveillance of human populations and multiyear funding of laboratory and animal resources. In such cases, continuity of funding, possibly through bridging mechanisms, can be extremely important. Funding versatility and continuity of support are vital for providing the necessary infrastructure to encourage research on gender differences and susceptibility.


Opportunities for cofunding and for public/private cooperation with university, nonprofit, and industry groups should be sought.

While the preeminent federal role is to create new knowledge, the line between basic and applied research is sometimes blurred. Within this gray area, however, cooperative relationships with the private sector can produce great mutual benefits (NAS/NAE/IOM, 1995). Gender differences in susceptibility are not only important to many government agencies but also to many private organizations willing to form research and funding partnerships. Some of these private-sector groups include foundations, universities and university consortia, labor unions, and industry (particularly drug companies). Partnerships with private industry, such as pharmaceutical companies, for research in gender differences and susceptibility could produce data and information that may have both long-term biomedical significance and short-term value for product development and marketing. A recent example of such a partnership occurred in the Women's Health Initiative (WHI).3 A $9 million “add-on” study, completely funded by the private sector, now conducts research on the effect of hormone replacement therapy on cognition and Alzheimer's disease. Because the WHI involves a sizable population of women of all races and lifestyles, the data derived from this “add-on” study will aid the WHI, the private-sector sponsor, and the women who participate.

In the arena of innovative cooperative efforts, more attention should be given to partnerships with nontraditional partners, such as nongovernmental organizations (NGOs). NGOs are the site for some of the most innovative work that is being done on linking gender and environment, including a focus on women's health outcomes.


Strategies for utilizing national health surveys and databases should be developed. The broader public health community should be encouraged to find ways to improve and broaden such utilization in the future.

Large-scale health surveys and their databases, particularly longitudinal and cross-sectional data from study populations, contain potentially valuable information that could be used to study environmental issues. CDC alone has data from its National Center for Health Statistics (NCHS), National Health and Nutrition Examination Survey (NHANES), National Health Interview Survey, National Ambulatory Medical Care Survey, and National Disability Study. Other studies that examine specific cohorts include the ongoing Nurses Health Study and the Six City Study. These surveys and databases are well known to a narrow constituency but not so well known to the broader scientific community. Given the significant investment that has already been made in collecting this information, it would be very cost-efficient to invest a bit more for further analyses.

Access, however, remains a critical barrier to greater utilization of these resources. Although CDC is increasing its efforts to make data more accessible by providing tapes and CD-ROMs, the results are still not user-friendly, nor are the data easily exported into common statistical data analysis software. All federal agencies should be encouraged to develop pathways for easier public access to the information.

Another barrier to access is the issue of confidentiality of stored tissue samples. Many states store DNA samples for long periods after their initial use. In its report, Assessing Genetic Risks (IOM, 1994b) the committee stated that:

Later access to DNA samples or to the profiles for other purposes should be permitted only when … b) the data are to be anonymously studied. … In general, regardless of the purpose for which it was compiled, this information should be accorded at least the confidentiality that is accorded to medical records.

This committee suggests that NCHS and CDC take the lead in conducting a strategic review of their current portfolio of population-based health surveys and databases and develop guidelines for public use of the data for purposes other than national extrapolations. A general review of these data resources and the development of strategies to fully use them would comprise an appropriate agenda for an interagency workshop (see III-1 above). Some specific issues to be addressed at such a workshop might include privacy, confidentiality, informed consent, ownership of samples and data, and the identification of new uses and users.

Once the current survey portfolio is known, a review of the types of data it contains would be the next step. In some cases the data may have limited application, because they are incomplete or noncomparable or because they were gathered with survey methodologies that are now outdated. After identifying the strengths and weaknesses of the current portfolio of surveys, agencies should ask the broader public health community to help find ways to improve and broaden their usefulness in the future. The goal should be to expand the utility of the information and broaden the community of users, without compromising the original public health purposes of the surveys. With advice from the user community, ways should be explored for improving and integrating these surveys and databases and for advertising their availability to a broader community of interested researchers.

The IOM report, Toxicology and Environmental Health Information Resources (IOM, 1997), makes similar recommendations. Many of that report's recommendations focus on the need to heighten awareness of health information resources, to analyze user profiles, and to simplify the navigation into and through the databases for toxicology and environmental health. These recommendations should be expanded and applied to broader fields of health data and information in order to assist the development of new knowledge for gender differences and susceptibility.


Strategies should be developed to identify, protect, and utilize other irreplaceable research resources.

DNA, serum, and tissue samples have been collected in conjunction with some of the health surveys described above. While these resources were collected for specific purposes, the committee suggests they may have broader uses. Geneticists, for example, would be interested in the large population of “normal” DNA contained in such collections. Such uses would need to be consistent with current ethical standards and laws governing confidentiality, informed consent, and privacy.

The National Cancer Institute supports a number of cancer registries whose resources are generally available to interested researchers. Additional resources of this sort are maintained by private nonprofit groups: the American Type Culture Collection, for example, collects human normal and tumor cell lines as well as microbial cell lines; the Human Biological Data Interchange collects longitudinal samples from families with autoimmune disorders.

An inventory of these resources would be useful and might reveal the need for (and sponsorship of) new components or new collections that would make these existing surveys and databases more useful and productive. A fuller description of the issues, problems, and features of successful resource sharing is contained in the IOM report on Resource Sharing in Biomedical Research (IOM, 1996b).


A summary of the committee's recommended priorities that were discussed above is displayed in Table 4.

TABLE 4. Summary of Recommended Priorities.


Summary of Recommended Priorities.

The committee believes that these recommendations, taken together, provide a good beginning for the identification of research priorities and interagency initiatives that are well within the mission of the various agencies involved. As additional resources become available, it is hoped that research priorities in this area can be implemented. Opportunities abound, and the new knowledge that can accrue will be valuable for all women throughout their lives.

For the purpose of this report, sex is generally used to designate chromosomal or biologic phenomena linked to having one or two X chromosomes, whereas gender is used when referring to the social expression of living as a man or woman.

Subsequent to the workshop, the committee was made aware of a listing of federal projects related to women and environmental factors, developed by the Office on Women's Health, of the Department of Health and Human Services. However, this listing did not specifically highlight research designated in this area.

The Women's Health Initiative, an activity centered in NIH, is focusing on the major causes of death, disability, and frailty in postmenopausal women. The overall goal of the WHI is to reduce coronary heart disease, breast and colorectal cancer, and osteoporotic fractures among postmenopausal women through prevention/intervention strategies and risk factor identification. WHI is a 15-year effort that is budgeted for $625 million.



For the purpose of this report, sex is generally used to designate chromosomal or biologic phenomena linked to having one or two X chromosomes, whereas gender is used when referring to the social expression of living as a man or woman.


Subsequent to the workshop, the committee was made aware of a listing of federal projects related to women and environmental factors, developed by the Office on Women's Health, of the Department of Health and Human Services. However, this listing did not specifically highlight research designated in this area.


The Women's Health Initiative, an activity centered in NIH, is focusing on the major causes of death, disability, and frailty in postmenopausal women. The overall goal of the WHI is to reduce coronary heart disease, breast and colorectal cancer, and osteoporotic fractures among postmenopausal women through prevention/intervention strategies and risk factor identification. WHI is a 15-year effort that is budgeted for $625 million.

Copyright © 1998, National Academy of Sciences.
Bookshelf ID: NBK100875
PubReader format: click here to try


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.3M)
  • Disable Glossary Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...