5Health Policy

Publication Details

Science should play a critical role in the formulation of public health policy. For a complex and newly developing subject such as environmental justice, however, science often cannot provide policymakers with research data pointing conclusively to a particular solution. Policymakers must therefore try to understand both what the science base is able to offer and its limitations. This understanding will help them weigh the merits of delaying a decision, in the hope of obtaining better data and analysis, or proceeding with a decision on the basis of imperfect knowledge and acknowledging uncertainties about its consequences. Decisionmakers who find themselves in this situation should also be very attentive to the inferences and presumptions with which they are operating; to what level they have assigned the burden of persuasion for or against a given option, and how it was chosen; and to ways of expanding not only the sources of additional information but also the processes and participants involved in arriving at a decision.


The committee believes there is ample evidence that racial and ethnic minorities and residents of low-income communities have a higher incidence of disease and a lower general health status than do majority and more affluent populations (National Center for Health Statistics, 1998b). In addition, the committee also believes there is substantial evidence that these populations also experience higher levels of exposure to potentially harmful environmental stressors (defined in this report to include noise, odors, and particulate matter, as well as toxicants, chemicals, and other pollutants) than other populations, although the committee acknowledges that the evidence is not consistent in either its methodological rigor or its conclusions (Brown, 1995; Szasz and Meuser, 1997). Beyond these beliefs, however, the chain of causality scientifically demonstrating a lack of environmental justice grows weaker.

One limitation is that with a few exceptions, an adequate body of knowledge directly establishing the link between those things researchers believe to be environmental hazards and specific adverse health outcomes typically does not exist. A good start on this research has been made in the occupational health field (Frumkin and Walker, 1997), but much more research is needed to document the nature, scope, and severity of the toxic effects of environmental stressors, particularly when multiple sources or types of potentially harmful substances exist. When causal relationships between environmental hazards and adverse health outcomes are established, risk assessments are needed to evaluate the relative contributions of the various hazards, both in terms of the numbers of people affected and the severity of illness. Judgments can then be made about research and intervention priorities.

A second limitation is that policymakers need better evidence about the precise roles of environmental hazards in creating adverse health outcomes and the burden borne by racial or ethnic minorities or demographically defined populations (Brown, 1995). Although evidence of disproportionate exposure to potentially harmful substances strongly suggests a disappropriate health burden, a variety of other factors might be involved. Clarifying these relationships will require detailed epidemiological studies, often on small, targeted populations, examining the relationships among disease incidence, environmental hazards, and other behaviors or exposures that may cause or exacerbate health problems. Properly disseminated, the research arising from such a broad-based public health perspective will help increase awareness that a given health outcome can be caused by multiple environmental hazards (with cumulative or interactive effects) and that a particular hazard can contribute to multiple adverse outcomes. Therefore, efforts are needed to evaluate total exposures and then to identify the most severe and remediable hazards as a guide to the most effective courses of action.

As discussed in Chapter 3, uncertainties in the scientific analysis of environmental justice issues have several sources, some reflecting limitations of current methodologies, others reflecting the lack of funding for implementing available methodologies. One large gap in current research data is the lack of tools for measuring important elements of environmental health and environmental justice issues, including health status indicators, sociodemographic characteristics, and other lifestyle factors that influence a population's health. Even where demographic and environmental databases do exist, it may be hard to link them. Small sample sizes and the inability to disaggregate data by such factors as race, ethnicity, income, and education further impede rigorous analysis and robust conclusions. In addition, it is very difficult to identify specific environmental hazards and to document their role in specific adverse health outcomes. These problems are even more difficult when there are complex interactions among multiple environmental hazards and long latency periods. Finally, little is known about the differential susceptibilities of various populations to environmental hazards. This suggests that greater attention, resources, and innovation are needed to develop the science base required to inform policymaking with respect to environmental health and justice.

Increased support for environmental health research may encounter political opposition from those who are concerned that the results of such research could increase their legal liability and the associated level of litigation they would face. Opposition to research and extreme skepticism about its results are rewarded if uncertainty leads to paralysis. It is the committee's view that policymakers need a philosophy that is consonant with the inherent difficulty of linking environmental hazards to adverse health outcomes. The populations in question generally have a lower health status than other groups in society and, therefore, are likely to be more susceptible than others to the adverse impacts of environmental hazards. Moreover, they are also likely to be less able to deal effectively with these hazards and their impacts, either in the political arena or by obtaining adequate access to high-quality health care.

It is a fundamental principle of statistical decision theory that the interpretation of scientific data requires making a value judgment about the relative importance of avoiding different kinds of errors. In the context of environmental decisionmaking, the possibility of error arises when human activity is suspected of causing ill-effects to people and their surroundings, but the linkage has not or cannot been proven. Society's response to such uncertainty reflects its willingness to tolerate one of two sorts of errors: (1) either failing to protect the public when there actually is a connection between an activity and a harmful result or (2) failing to protect the activity itself when it is not the source of the underlying problem. Which of these two types of errors is preferred is reflected in the manner in which the burden of proof is assigned. Are the activities presumed innocent until proven guilty (to some specified degree of confidence)? Or, rather, is the public presumed worthy of protection until the activities have established their innocence (with some specified degree of confidence)? There is no way of avoiding these choices.

This committee has concluded (as have others studying environmental justice) that social-political institutions tend to place the burden of proof on poor communities. This may not be done maliciously or even overtly. Rather, there are general procedures in place that foster inaction until scientific evidence of a firmer nature than most poor communities can muster has been collected. The procedures reflect, in part, the norms of basic scientific research, which place a premium on avoiding “false positives”—that is, accepting the existence of phenomena that do not actually exist. Although these norms have served the scientific community well, they place poor communities at a disadvantage. Such communities generally lack the ability to generate the scientific studies needed to confirm or disprove their suspicions, to examine the studies conducted by others, and to translate their concerns into terms that scientists employ. The committee urges the medical and scientific communities to make better use of lay observations, to make local communities partners in the design and conduct of scientific studies, to help citizens understand scientific data, and to help local public health officials and health practitioners deal with environmental health problems. If followed, the committee's recommendations should help establish a more equitable sharing of the burden of proof among the public, government agencies, and those engaged in potentially hazardous activities.

The committee also urges the scientific and political communities to recognize that the environmental problems faced by communities of concern are but one among many different problems they confront, including poor nutrition, limited education, minimal political representation, high unemployment, inadequate transportation, poor sanitation, hazardous workplaces, and insufficient vector control. When a community is under stress, primary attention should be given to attempting to address its health concerns in prudent ways. Secondarily, attention should be directed to identifying the accountable parties because holding people accountable for past misdeeds can be a valuable social function and provide incentives for appropriate behavior in the future.


As noted in prior chapters of this report, action, supported by adequate resources, is needed to develop and implement a public health strategy, to improve the science base, and to enhance awareness and understanding of these issues on the part of health professionals, educators, the business community, public officials, and the general public. In addition, policymakers need a more expansive perspective and approach to the development of public policy—and public-sector decisions—that have a bearing on environmental health and environmental justice.

Inevitably, in some situations delaying the decisionmaking process to wait for more data or better research is not a desirable or acceptable course. Delay may mean that a community will forego a significant economic benefit or may suffer further exposure to potentially dangerous environmental hazards. Foreman notes two other reservations about seeking solutions for environmental justice problems predominantly through scientific inquiry. He notes that “science cannot resolve what are ultimately value questions.” In addition, he suggests that “calls for more and better scientific studies may, if successful, simply generate more information than policymaking institutions can reasonably digest given their available resources” (1998, p. 112).

These constraints on public policymaking do not mean that decisionmakers should forego evidence-based problem solving. To the contrary, it means that great attention must be paid to making sure that the decisionmaking process is as open as it reasonably can be to ideas, information, and participants derived from affected or potentially affected communities. In the case of existing environmental hazards, members of the local community may be the best available source of information regarding exposures—including multiple exposures or routes of exposure—or about the interplay of exposures and other potentially influential factors or susceptibilities that exist in the community. They may also be able to identify emerging patterns of illness before the health care delivery system or other possible sources can. Engaging the community not only has the advantage of increasing policymakers' knowledge and understanding of local problems, but can also give voice to community values. Hearing citizens' beliefs and concerns is essential to securing general acceptance of policies and actions. In this context, the community of concern needs to encompass all those likely to be significantly affected by a policy decision—including residents and community groups, business and labor, trade associations, and government and regulatory agencies.

Two case studies illustrate how effective stakeholder collaboration can lead to a broadly satisfactory resolution of a major environmental problem, and how its absence can result in stalemate and widespread dissatisfaction. In California, diverse interests were able to agree on long-range goals and strategies for preserving a valuable estuary (see Box 5-1). In Illinois, the attempt to impose a solution on a community that had been excluded from the decisionmaking process left the parties entangled in litigation that served environment and health policy needs poorly (see Box 5-2).

Box Icon

BOX 5-1

SUCCESSFULLY ENGAGING STAKEHOLDERS: SAN FRANCISCO BAY/DELTA ACCORD. Declaring “a major victory of consensus over confrontation” on December 14, 1994, California Governor Pete Wilson and Cabinet-level federal officials announced the signing (more...)

Box Icon

BOX 5-2

INSUFFICIENT STAKEHOLDER COLLABORATION: GRANITE CITY, ILLINOIS. When stakeholders are not included early in the decisionmaking process, they are more likely to oppose the risk management decision and block its implementation. This has been happening in (more...)

The Environmental Protection Agency's (EPA's) National Environmental Justice Advisory Council (1996; see Box 5-3) has developed a thoughtful and comprehensive plan, the Model Plan for Public Participation, which emphasizes planning and inclusiveness. Models and experiences from other areas of public policy are also available (National Research Council, 1996), but any model must be applied flexibly, must be adapted to specific local interests and concerns, and must be evaluated to ensure its effectiveness.

Box Icon

BOX 5-3

HIGHLIGHTS FROM THE NATIONAL ENVIRONMENTAL JUSTICE ADVISORY COUNCIL'S PUBLIC PARTICIPATION CHECKLIST. Use the following guiding principles in setting up all public meetings: Maintain honesty and integrity throughout the process;

Policymakers should not shy away from the inclusion of community participants on the grounds that the issues are too technical or difficult for the general public to understand or respond to. The research literature generally finds that motivated lay audiences can understand many environmental issues, especially if a conscientious effort is made to determine their beliefs and concerns, and to develop appropriate communications (Fischhoff et al., 1997; National Research Council, 1989c, 1996). Even potentially esoteric material such as risk assessments or cost-benefit analyses can and should be understandable if presented in policy-relevant terms. When citizens object to the conclusions of an analysis, a careful examination is needed to see whether they fail to understand its premises or, instead, understand it all too well and object to how it summarizes existing studies, treats uncertainties, or defines risk (Fischhoff et al., 1981; Glickman and Gough, 1990).


The committee concludes that concerns about environmental health and environmental justice are legitimate and should be taken seriously, even if the information and data related to these concerns still lack some of the rigorous scientific attributes that policymakers desire. Policymakers cannot assume that these concerns are without merit. However, policymakers should also recognize that many other considerations go into decisions such as choosing a site for a new manufacturing plant or solid waste facility, removing an alleged hazard, or imposing expensive environmental controls. Decisions of any significant consequence will almost always involve choices and trade-offs. Given the current state of knowledge, the committee believes that policymakers should be attentive to potential environmental hazards and adverse health outcomes and should be meticulous about including affected communities in the decisionmaking process.

Requiring residents of a potentially affected community to prove definitively that alleged adverse health outcomes are linked to environmental hazards may put effective participation beyond their means. Conversely, accepting assertions of environmental injustice without reservation could lead to actions that adversely affect other interests without significantly improving a community's environment. Knowing the content and limits of the best available science will encourage the framing of reasonable inferences and help in determining the confidence to be placed in such inferences and the appropriate burden of rebuttal.

The committee believes that its recommendations are consistent with the collaborative view of risk management that has emerged from the series of National Research Council reports on this topic (National Research Council, 1983, 1989c, 1996b). The committee saw much goodwill in the communities it visited, which should allow for humane solutions to otherwise seemingly intractable environmental problems. The committee believes that lasting, workable arrangements can be created between communities of concern and others who have the capacity to use science and the law to help minority or disadvantaged populations protect themselves and their communities from harmful environmental stressors. Among those who will need to be involved in such arrangements are legislatures, regulatory agencies, business and industry, institutions of higher education, and the medical establishment.

Recommendation 4. In instances in which the science base is incomplete with respect to environmental health and justice issues, the committee urges policymakers to exercise caution on behalf of the affected communities, particularly those that have the least access to medical, political, and economic resources, by taking reasonable precautions to safeguard against or minimize adverse health outcomes.