Environmental Health Sciences Decision Making: Risk Management, Evidence, and Ethics: Workshop Summary (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1 Approaches to Decision Making When risk assessment was in its infancy in the 1950s and 1960s, the general assumption was that the major cause of a problem could easily be identified and a solution generated. As time has progressed and society and science have faced new problems, this assumption is no longer applicable to the decision-making process. Thus, society is currently at a crossroads in environmental health deci- sion making, and there is a need to carefully examine the current paradigm and think about what science can do to improve the way decisions are made. This chapter highlights current approaches to environmental health decision making and opportunities to improve decision making under complex problems. Human–Environment Network: Challenges to Environmental Health Christopher J. Portier, Ph.D., Director, Office of Risk Assessment Research, NIEHS Risk analysis and risk decision making consist of balancing the needs of science, economics, and society. Science and economics conduct research based on hypotheses and interpret the results for societal application. Society, through government, balances the various scientific and economic outcomes to ultimately decide policy. It is the role of risk analysis to translate among these groups, evaluate the literature so that decisions can be made and implemented, and com- municate to all the stakeholders (Figure 1-1). In most risk analyses, it is assumed that one can manage each risk independently and preserve the public’s health. As time has progressed and society and science have faced new problems, this assumption is no longer applicable to the decision-making process. Humans are not independent of their environments; rather, their interac- tions affect the environment, and the changes they make to the environment can  

10 ENVIRONMENTAL HEALTH SCIENCES DECISION MAKING Communication Science Economics Research Research Test Hypotheses Test Hypotheses Interpretation Interpretation Risk Analysis Translation Evaluation Risk Characterization Assessment Benefit Implementation Assessment Communication Society Decision Cost/Benefit Risk Set Priority Management Manage FIGURE 1-1  A risk decision-making approach is a function of evaluating science and economic information and effectively communicating the outcome to society. SOURCE: Portier, unpublished. 1-1 affect their health. Taking complex human–environment interactions into account requires a new systems approach to environmental health decision making. With regard to human health and the factors that impact it, four overarching categories are basic needs, shelter factors, personal factors, and endogenous factors. All four are social determinants of health and play an interactive role in environ- mental health, yet only endogenous factors cannot be changed by individuals (Figure 1-2). Basic factors are needs that are crucial to survival: having food to eat, water to drink, and clean air to breathe. At the most basic level of human function- ing, these factors are the foundation for everything human beings do. The next level—shelter factors—includes items that, although not directly needed to live, can improve the quality of life, such as physical surroundings, like homes and schools; community; access to health care and hospitals; clean water for recre- ation; and the ability to be employed and earn a living. Personal factors, the third category, are less tangible in nature, such as exerting control over one’s life and making choices or decisions, the feeling of social cohesion, and establishing rela- 

APPROACHES TO DECISION MAKING 11 FIGURE 1-2  Basic needs, shelter factors, personal factors, and endogenous factors inter- act in a holistic network to determine health outcomes. SOURCE: Reproduced with permission from Environmental Health Perspectives. Gohlke, J., and C. Portier. 2007. The forest for the trees: A systems approach to human health 1-2 Low-res Bitmapped research. Environmental Health Perspectives 115(9):1261–1263. tionships. Last are the endogenous determinants, such as genetics, race, ethnicity, and life history, which cannot be changed by individuals themselves. All of these components, whether they are basic needs, shelter factors, per- sonal factors, or endogenous factors, interact with each other, which creates a snapshot in time to determine a person’s health status. The categories are further affected by the state of the physical environment and ultimately impact the health of an individual and his or her relationship with the environment. They need to be looked at from a holistic point of view, not a fragmented one, which is how environmental health decisions have historically been made. The Need for a Holistic Approach to Decision Making What people do in one aspect of the environment can greatly impact other aspects. For example, air quality and climate change impact human health through their interactions. Scientists and doctors have begun to see the earth’s tempera- tures rise and, along with that, a direct effect on human and environmental health. With increases in climate change, the quality of the air may change, resulting in an impact on the ozone level and a reduction in smog clearance. In turn, such 

12 ENVIRONMENTAL HEALTH SCIENCES DECISION MAKING chronic diseases as asthma may become more prevalent, a situation that places increased burden on the health care system. In addition, higher temperatures play a role in extreme weather events and natural disasters, which can wreak havoc, such as with Hurricane Katrina and Hurricane Rita as well as during the record heat wave in Europe in 2003, which resulted in significant mortality. In the past, scientists and researchers would look for a single cause to account for global climate change, characterize it, and try to find a solution. That approach is no longer sufficient for making an assessment of risk and, in the end, an environmental health decision. Global climate change, as with many other environmental heath issues, is not caused by a single factor; rather, a number of different factors will play an important role in the severity of the health impacts of climate change over the next century. Each factor will have its own set of associated risks. Looking at the complexity of the environment on a global scale, one begins to realize that environmental health decisions have been made in a fragmented way. For example, different types of regulations for chemicals in the air contrib- ute to air quality and climate change. Methyl mercury is regulated differently depending on the context and uses. Carbon dioxide, particulate matter, sulfur dioxide, and other chemicals are regulated independently, without assessing whether making a change to the regulation of one chemical may impact another chemical. Finally, there is the larger question of the impact of these individual regulations on air quality and climate change overall. Risk decision making for environmental health in the future will need to be based on a holistic view of the global network and its interactions. The Global Network What occurs in one environment or country can greatly impact another envi- ronment or country. The United States is not a solitary country immune to the decisions made in other countries—and the converse is also true. The world is becoming increasingly interdependent and, as a result, there are direct impacts of the environment on health. For example, the growing demand by U.S. consum- ers to have access to all types of fresh fruits and vegetables throughout the year carries with it many of the vulnerabilities of greater globalization of the U.S. food supply. Food manufacturing is also taking place in conditions that are pos- sibly raising the risk of zoonotic diseases, which move from animals to humans (Hastein et al., 2006). In addition, while coal consumption is projected to stay static in countries of the Organisation for Economic Co-operation and Develop- ment (OECD), coal use in non-OECD states is projected to increase over the next 50 years (EIA, 2004). Yet OECD states are a contributing factor to the trend, as the manufacturing of goods is being shifted to these countries (EIA, 2007). There is growing recognition that environmental health is global. Environmental issues 

APPROACHES TO DECISION MAKING 13 in every part of the world can therefore have repercussions for the entire planet. If policy makers do not consider this broad picture, they risk making decisions that may be inappropriate to the overall goal of trying to improve human health on the planet. Scientific Direction With regard to the human system and how science addresses its exposure to hazards, a large body of research and testing is being performed, from the population and clinical levels to the molecular level. Although all of this science contributes to understanding the impact of the environment on health, most risk assessment is based on toxicological and epidemiological evidence and not on emerging sciences, such as genetics and toxicogenomics. Scientists and policy makers therefore need to look at the emerging areas of science to find ways to incorporate this research into the environmental health decision-making pro- cess. The ultimate goal would be to use scientific evidence to guide exploration into the environmental causes of disease and remove these hazards from the environment. Through the use of an environment–disease interaction network, Portier’s laboratory has taken approximately 500 compounds and, using genetics, created a linkage system to see how disease and environmental factors may match. Target- ing the metabolic syndrome cluster illustrates the complexity involved in looking at disease and shows how closely related diseases have common etiologies. In these linkages, the chemicals and pharmaceuticals that one expects to cluster do. This type of activity and analysis focuses on genetics in relation to disease, target- ing ideas for research in terms of looking at environmental diseases. Use of these types of networks and the overall use of more tailored, personalized medicine are beginning to guide scientific research. This development is evidenced in the study of biomarkers, which are essen- tially indicators of disease or therapeutic effects that can be measured through dynamic imaging tests, as well as tests on blood, tissue, and other biological samples (FDA, 2006) by the federal government, including the National Insti- tutes of Health, the Environmental Protection Agency (EPA), and the Food and Drug Administration. Furthermore, the National Toxicology Program and the EPA are using high throughput screening to set priorities and move forward with their testing program. However, without applications to the risk assessment field, these new scientific areas may lose funding priority. Science needs to continue to make strides in new research areas and to focus decision making by looking at risks in a global networked capacity that will strengthen the ability to protect public health. 

14 ENVIRONMENTAL HEALTH SCIENCES DECISION MAKING Alternatives Assessment as a Strategy for Decision Making Mary O’Brien, Ph.D., Oregon Toxics Alliance The Need to Overcome Obstacles The presumed goal of environmental health science decision making is to produce fewer harms to human health and the environment. However, there is a disconnect between environmental health science and decision making for environmental health. The field has made advances in knowledge of toxicology, structure activity relationships, cumulative impacts, animal and human develop- ment, and the nature and amount of toxic chemicals to which all living beings on earth are being exposed, but there is not a clear transformation of that science into health outcomes. The disconnect may be directly attributable to decision-making processes. Too often in the scientific community, among many other professions, there are many obstacles to good decision making, including a narrow power base that leads to narrow decision making and the fact that human nature is often strongly habit based, and decisions are made in ways that stifle creativity and ingenuity. Several key elements in current environmental health decision-making pro- cesses contribute to less than optimal environmental health outcomes: • The assumption that lack of toxicological evidence equals lack of risk. An example is the ongoing substitution of one brominated fire retardant for another, each of which has presented different environmental health problems. • Lack of training of chemists and engineers in innovation for green engi- neering rather than merely for function and cost. • The permitting of technologies and chemicals without consideration of alternatives. • Lack of permitter training or authority to help or require the applicant to consider greener, even life saving alternatives. Decision Making Using Alternatives Assessment In contrast to many current environmental health decision-making processes, alternatives assessment involves four essential elements: 1. The first element is responsiveness to early warnings of health damage. There is no shortage of toxicologists, epidemiologists, physicians, neighborhood residents, or workers who have offered early warnings of environmental health damage. But some policy makers have too often justified “no responsive action,” claiming that the risk has not been fully characterized. One compelling example has been the weak response of the United States to climate change despite envi- 

APPROACHES TO DECISION MAKING 15 ronmental health scientists’ warnings of its trajectory toward massive species extinction, resource wars, starvation, and disease. In order to move past this and work toward producing fewer harms, several questions are useful to ask in the face of the uncertainty surrounding early warn- ings. For example, is the uncertainty about the degree of harm or the existence of the harm? In other words, is the argument in a given situation truly about no harm resulting or about the precise amount or acceptability of harm? Does a claim of “no harm and therefore no change needed” seem warranted on the basis of past analogous experiences? And if health impacts are acknowledged, are pro- nouncements of acceptable risk resulting in a failure to search for or implement reasonable alternatives? 2. A second element is the engagement of diverse perspectives in the devel- opment and examination of reasonable alternatives for producing fewer harms. Joint examination of reasonable options is the most scientific process for decision making because the science brought by one sector will be held to the light of replicability or accuracy by other sectors. It is democratic, including both those who stand to gain money or health or both under particular options and those who stand to lose health or money or both under those same options; those who defend; and those who innovate. The human tendency to centralize and retain power resists transparent, equi- table, innovative decision making. This is precisely why diverse participation and transparency need to be mandated, as they are, for example, in the nation’s National Environmental Policy Act regulations. 3. A third essential element is giving the benefit of the doubt to nature and public health. Different types of doubt exist for different types of environmental health decisions, so it is important to ask some questions about perceived or claimed uncertainty. For example, who is the beneficiary or beneficiaries of con- tinued uncertainty? What incentives exist to resolve uncertainty in favor of the status quo? Alternatively, can incentives be offered to favor environmental health advances in the midst of uncertainty about the precise nature of harms? One strategy is imposing a deadline when alternatives must be instituted for a given technology to be put in place or milestones when a particular issue or decision has to come under review again. For example, the goal of a 90 percent reduction in auto emissions from pre-1968 levels by 1975 led to the development of the catalytic converter, which has been considered one of the greatest environ- mental successes of the past century (Palucka, 2004). 4. A fourth critical element is the monitoring of results for successes and new early warnings, which becomes the foundation for improvements in deci- sions. Monitoring should be a central and early component of decision making. If the likelihood of monitoring for an outcome is low but the environmental stakes are high, then the initial decision should build in time-certain reevaluation in response to monitoring. Incorporating responsiveness to early scientific warnings, entertainment of 

16 ENVIRONMENTAL HEALTH SCIENCES DECISION MAKING diverse solutions, the favoring of environmental health amid uncertainty, and monitoring into environmental health decision making will increase the odds of optimal human health outcomes. Beyond Precaution Bernard D. Goldstein, M.D., Professor, Department of Environmental and Occupational Health, University of Pittsburgh Graduate School of Public Health Public Health Approach Many environmental health decisions have been made from what others have described as fragmented, narrow, “reductionist” approaches. With the recent resurgence of awareness in many environmental areas, such as global climate change, the timing is right to refocus efforts on making decisions that benefit the health of the public by addressing issues holistically. Three approaches are necessary to make environmental health decisions from a holistic or public health perspective: bipartisan or global environmental policy, a systems-based approach, and science that is focused on answering the most important questions. A holistic decision making process can be best achieved by implementing a public health approach, which is defined by three key factors. First, policy mak- ers need to take responsibility for all outcomes, whether good or bad. Second, they must utilize the core public health functions of assessment, assurance, and policy development. Third, the right target needs to be set to link the research agenda to the data needs, so that the research is relevant to both the scientist and the policy maker. Trying to solve a problem without using a systems or public health approach often creates a secondary problem or unintended consequence. For example, most people in public health are familiar with the replacement of microbially contaminated surface drinking water with arsenic-contaminated groundwater in Bangladesh. Furthermore, central to a public health approach is engagement of the public, as solutions need to be relevant from a community perspective. Fail- ure to provide the public with reasons to care about an environmental issue can therefore result in a policy failure. Caution in Applying the Precautionary Principle The precautionary principle, developed as a result of the need for action in the face of scientific uncertainty, is an important part of environmental health decision making that is advocated by the European Union (EU). Through their adoption of this principle, the EU is currently undertaking a reevaluation of the relationship between individuals and their community and, in a broader sense, 

APPROACHES TO DECISION MAKING 17 their consortium of nations. The main tenet of this principle, the need to act in the face of scientific uncertainty, is articulated in the Rio Declaration on Envi- ronment and Development (United Nations, 1992). According to the European Commission, this principle should be applied whenever the “scientific data are insufficient, inconclusive, or uncertain and where a preliminary scientific evalu- ation shows that potentially dangerous effects for the environment and human, animal or plant health can be reasonably feared” (EU, 2008). While attention should be paid to this idea, there is also the need to step back and examine what the implementation of the precautionary principle means to the overall practice of public health. If policy makers are going to rely on precaution, then they need to authorize research to monitor and ensure that a precautionary approach is nec- essary and has succeeded. Furthermore, the research agenda should be linked to objectives of data needs and data quality. One of the primary arguments for using the precautionary principle in order to act in the face of uncertainty implies that without this principle there is an absence of action in the face of scientific uncertainty. This, however, is not the case. For example, in the 1970s, the United States banned the manufacture of polychlorinated biphenyls (PCBs) despite strong opposition by industry and clamor that there was no scientific evidence showing harm caused by these chemicals. Industry continues to take this position, yet because of the ban, humans and the environment have much lower levels of PCBs than they did at the time of the ban. The lower levels of PCBs are a result of actions taken despite scientific uncertainty, but without stated recourse to the precautionary principle. One therefore needs to question what this principle adds to already existing pub- lic health concepts. Supporters of the precautionary principle sometimes assume that scientists speak with one voice about environmental policy decisions. This is also not the case—nor, in U.S. society, do scientists unilaterally make policy decisions. The EU actions on aflatoxins illustrate why the precautionary principle may not always be used for its appropriate purpose of protecting public health and the environment. It is scientifically accepted that aflatoxins are a family of toxins capable of producing liver disease and liver cancer. The fungus producing afla- toxins is widely distributed in foods, particularly among groundnuts and cereals, and is known to grow in wet climates and places with delayed harvesting. The EU has used the precautionary principle to enforce the most stringent standard in the world for aflatoxin. This standard has resulted in favoring European growers to the exclusion of $700 million a year worth of sub-Saharan African products (Majone, 2002). A less stringent standard, whose public health significance has been rejected by the Joint Expert Committee on Food Additives of the World Health Organization/Food and Agriculture Organization of the United Nations, would result in a difference in risk of developing liver cancer of less than one case a year in Europe (Majone, 2002). This example and many others from the agriculture sector, including greenhouse implications of noncompetitive farming 

18 ENVIRONMENTAL HEALTH SCIENCES DECISION MAKING and methane production, call into question the primary motives for using the precautionary principle. The question could be asked if it is really preventing or reducing risk in the face of uncertainty, or if it is being used to build trade pro- tection walls around Europe. The use of precautionary principle also calls into question how society determines what a significant health benefit is as well as which issue has priority over another. All of these questions need careful analysis and exploration before applying the precautionary principle. Connecting Science and Policy A systems approach is critical to addressing environmental justice. Three truisms are that there are more environmental hazards in disadvantaged com- munities, that there are more individuals with poor health in disadvantaged communities, and that individuals with poor health tend be more susceptible to environmental pollutants. To scientifically evaluate the potential for environmen- tal health problems, the focus should be on the people who are most at risk from environmental pollutants, in whom effects are most likely to be seen. Performing such studies requires community involvement and a new approach to environmen- tal health research and decision making—one that looks at the entire system and not just the individual chemical. Ultimately, such an approach connects science and policy. Session Discussion: Applying a Systems Approach There can be a fundamental conflict between science and society. The sci- entist wants to make decisions using all the available evidence in as rigorous and objective a manner as possible; society is constantly grappling with ways to express what is an acceptable and unacceptable risk. The fundamental trade-off that is occurring between what society wants and what science wants is one that challenges the decision-making process. Some participants suggested that new approaches or understandings of the conflict may help improve risk assessment and environmental health decision making. The central theme of the three presentations was the need for a systems approach to environmental health decision making that encompasses the increas- ing complexity of the environment and avoids overburdening the thought process behind decisions. This theme generated ample discussion by the panel and par- ticipants about how to implement such an approach. O’Brien suggested that in order to avoid some of the complexities and not get bogged down, policy makers need to think much more broadly about a decision. Goldstein suggested that addi- tional attention needs to focus on the data quality objective and that complexity can be reduced, as can the overall length of the decision-making process, if the research has an objective and is not being undertaken just because a topic can be researched. Taking this a step further, Portier added that there may be a need to 

APPROACHES TO DECISION MAKING 19 revisit certain issues and decisions in order to see whether the scientific process is using information in a way that addresses the complexity. While many participants echoed the usefulness of a systems approach to deci- sion making, questions arose regarding the implementation of such an approach. For example, does a systems approach include additional steps or further data analysis in the decision-making process compared with the original method? Participants questioned what criteria need to be developed to determine how to incorporate this new information. Is there a standard, and if so, what is it? Portier countered that it wasn’t necessary to quantify all the data and links in the systems approach, but rather it is sufficient to know qualitatively what the links may be and how they could impact the decision-making process. Goldstein simplified this even more by explaining that a systems approach does not need to include complexity unless it adds value to the decision. A second challenge to the systems approach is whether the use of complex- ity itself as a tool to obfuscate and thwart a particular argument could occur by various stakeholders. While this is possible, O’Brien noted that a systems approach should engage many diverse sectors in an ongoing conversation on a particular problem, essentially a collaborative process. To this point and based on his experience, Goldstein raised questions as to how to make the outcome a consensus that moves the science forward. He also pointed out the importance of a policy maker or science agency deciding that there has been sufficient input solicited and adequate data gathered to make a decision. O’Brien countered that the collaborative group itself needs to be linked to action and that there is a need to put limits on the data process. She further suggested that milestones could be set for revisiting decisions, if needed.