| 摘要 | Executive SummaryOctober is Children’s Health Month, a time when official federal agencies call attention to important efforts and policies designed to advance childhood health and safety. In recognition, Environmental Protection Agency (EPA) Administrator Andrew Wheeler recognized the importance of the occasion in a press statement explaining: “EPA is highlighting the availability of its many programs and resources dedicated to improving air quality, reducing lead exposure, and protecting the health and well-being of children.” The occasion offers the perfect opportunity to assess federal programs that fund research in this field as well as the EPA’s recent move to reduce funding for a number of university-based children’s environmental health centers.
The EPA and the National Institute of Environmental Health Sciences (NIEHS) have jointly funded a number of university-based children’s environmental health centers for more than two decades. The EPA announced in May 2019 that it would cut its half of the funding for the centers starting in July. The announcement prompted a firestorm of criticism from environmental activists, including claims that reduced funding of these centers would undermine children’s health. Lost in this melee is the fact that these centers do not have a measurable impact on children’s health or add much to the body of research on the topic. In fact, many of the centers simply waste taxpayer dollars while funding junk science and environmental activism. Rather than simply cutting the EPA’s half of the funding, the Trump administration should go further and eliminate all funding for these centers and investigate similar NIEHS grant programs.
In 1995, the Clinton EPA, under the leadership of Administrator Carol Browner, published a policy statement calling on EPA assistant administrators to consider unique risks to children when assessing chemical risks. President Bill Clinton followed up two years later by issuing Executive Order 13045, “Protection of Children from Environmental Health Risks and Safety Risks.” The executive order set up an intergovernmental task force and offices at various agencies devoted to children’s environmental health. In 1997, pursuant to the executive order, the EPA began a collaboration with the NIEHS, which is housed at the Department of Health and Human Services’ National Institutes of Health. Since then the EPA and NIEHS have jointly funded, at a 50/50 ratio, several university-based children’s environmental health centers.
The EPA and NIEHS have poured a substantial sum of federal dollars into these centers as well as into other children’s environmental health research programs.NIEHS lists 13 centers on its website that are currently part of the program. However, the EPA/NIEHS “Impact Report” on the centers states that the two agencies have spent more than $300 million to fund 24 different centers since 1997. These centers’ grants only represent a fraction of federal spending for children’s environmental health. For example, a NIEHS newsletter reports that, over the past decade, the agency has spent more than $1 billion in children’s environmental health research, including $100 million in grants during fiscal year 2018 alone. In addition, NIEHS recently announced an expansion of its efforts to include new “children’s environmental health research translation centers” that will focus on outreach efforts related to children’s environmental health rather than conduct research.Even if the EPA were to permanently cut funding for the centers, a substantial portion of funding in the field will likely continue. Accordingly, it makes sense to better evaluate this funding and determine if any of it should continue. It is also worth examining whether other “children’s environmental health” programs funded solely by the National Institutes of Health,the EPA, or other federal agencies should be eliminated as well.
In addition, the National Institutes of Health should halt plans for NIEHS to fund the creation of additional “translational centers,” which will be specifically designed to engage in activism without even attempting to provide any scientific justification.
Taxpayers should not be forced to fund agenda-driven science. If the government spends on any funding for health-related research, it should focus on such things as finding cures and treatment for cancers, heart diseases, and other serious illnesses.Introduction
October is Children’s Health Month, a time when official federal agencies call attention to important efforts and policies designed to advance childhood health and safety. In recognition, Environmental Protection Agency (EPA) Administrator Andrew Wheeler recognized the importance of the occasion in a press statement explaining: “EPA is highlighting the availability of its many programs and resources dedicated to improving air quality, reducing lead exposure, and protecting the health and well-being of children.”1 The occasion offers the perfect opportunity to assess federal programs that fund research in this field as well as the EPA’s recent move to reduce funding for a number of university-based children’s environmental health centers.
The EPA and the National Institute of Environmental Health Sciences (NIEHS) have jointly funded a number of university-based children’s environmental health centers for more than two decades.2 The EPA announced in May 2019 that it would cut its half of the funding for the centers starting in July.3 The announcement prompted a firestorm of criticism from environmental activists, including claims that reduced funding of these centers would undermine children’s health.4 Lost in this melee is the fact that these centers do not have a measurable impact on children’shealth or add much to the body of research on the topic. In fact, many of the centers simply waste taxpayer dollars while funding junk science and environmental activism. Rather than simply cutting the EPA’s half of the funding, the Trump administration should go further and eliminateall funding for these centers and investigate similar NIEHS grant programs.
Background
In 1995, the Clinton EPA, under the leadership of Administrator Carol Browner, published a policy statement calling on EPA assistant administrators to consider unique risks to children when assessing chemical risks.5 President Bill Clinton followed up two years later by issuing Executive Order 13045, “Protection of Children from Environmental Health Risks and Safety Risks.”6 The executive order set up an intergovernmental task force and offices at various agencies devoted to children’s environmental health. In 1997, pursuant to the executive order, the EPA began a collaboration with the NIEHS, which is housed at the Department of Health and Human Services’ National Institutes of Health (NIH). Since then the EPA and NIEHS have jointly funded, at a 50/50 ratio, several university-based children’s environmental health centers.
The EPA and NIEHS have poured a substantial sum of federal dollars intothese centers as well as into other children’s environmental health research programs. NIEHS lists 13 centers on its website that are currently part of the program.7 However, the EPA/NIEHS “Impact Report” on the centers states that the two agencies have spent more than $300 million to fund 24 different centers since 1997.8 These centers’ grants only represent a fraction of federal spending for children’s environmental health. For example, a NIEHS newsletter reports that, over the past decade, the agency has spent more than $1 billion in children’s environmental health research, including $100 million in grants during fiscal year 2018 alone.9 In addition, NIEHS recently announced an expansion of its efforts to include new “children’s environmental health research translation centers” that will focus on outreach efforts related to children’s environmental health rather than conduct research.10Even if the EPA were to permanently cut funding for the centers, a substantial portion of funding in the field will likely continue. Accordingly, it makes sense to better evaluate this funding and determine if any of it should continue. It is also worth examining whether other “children’s environmental health” programs funded solely by the National Institutes of Health, the EPA, or other federal agencies should be eliminated as well.
Gaming the Science
Since the centers are based at universities, often within science- related schools, one would expect these children’s environmental health centers to be focused on producing independent, objective scientific research. Unfortunately, an examination of their activities reveals otherwise.They tend to focus on generating research to serve an activist-oriented, anti-chemical agenda.
As will be detailed below, many of the research studies these centers produce are essentially nothing more than statistical analyses that attempt to find an association between a chemical and a health effect. Yet, such associations alone do not prove cause-and-effect relationships and can happen by mere chance.
In fact, the chance of generating a statistically significant positive association is more common than one might think. As David Randall and Christopher Welser of the National Association of Scholars detail in their study, The Irreproducibility Crisis of Modern Science, it is common for positive associations to occur by mere chance. They explain that researchers apply a generally accepted standard for determining whether a finding is “statistically significant,” which ensures that positive associations will occur by chance no more than 5 percent of the time. That means that there is aone in 20 chance that a research study will generate a positive—statistically significant—association simply by accident.11The fact that false positives are relatively common has fostered both unintentional bias and scientific mischief, including the propensity for researchers to manipulate the data until it generates a positive finding. As James Mills of the National Institute of Child Health and Human Development lamented back in 1993, in the New England Journal of Medicine: “‘If you torture your data long enough, they will tell you whatever you want to hear’ has become a popular observation in our office.”12 Randall and Welser highlight one outrageous case where Brian Wansink, head of Cornell Uni- versity’s Food and Brand Lab, literally bragged in a blog post that he had schooled one of his students on how to churn data to generate positive results and get them published.13
Even when researchers do not torture data, many positive associations will occur by mere chance or unintentional biases. Stanford Professor of Medicine John Ioannidis demonstrated in a 2005 research article that most published research findings are false positives.He explained:
Simulations show that for most study designs and settings, it is more likely for a research claim to be false than true. Moreover,for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing bias.14
Hence, it is not all that hard to find associations, especially weak and largely inconclusive ones, and publicize those to scare the public and lobby for regulations. That is why we should pay attention to the strength of an association, which is expressed in these studies numerically as arelative risk ratio.
If the relative risk is one, the study reports no association between a chemical and any health effects. When the number is less than one, the research indicates that the chemical exposure may have positive health effects, which is what drug researchers seek when conducting drug trials. On the other hand, if the relative risk number is higher than one, the research indicates that the chemical may adversely impact health, and the higher the number, the stronger the association. Risk ratios of close to one and even two or three are generally considered weak and not particularly compelling for drawing cause-and- effect conclusions. They can easily occur by mere chance.
Mount Sinai School of Medicine epidemiologist Paolo Boffetta explains:
Although any measure of risk would follow a continuousdistribution and there are no predefined values that separate “strong” from “moderate” or “weak” associations, relative risks below 3 are considered moderate or weak.15Such weak associations alone do not offer much compelling evidence of actual health impacts. Boffetta explains further that weak associations raise the prospect that one of three serious problems with the research may be in play: “chance, bias or confounding factors.”16 He notes that some weak associations might be telling only under certain circumstances.
Specifically he says:
Identifying the causal nature of a weak association is not impossible, but requires large, well-planned, and well-conducted studies and supporting evidence from molecular and experimental studies.17
Yet, activist researchers are willing to draw conclusions and sound alarms even when they find weak associations without a larger, stronger body of research to support them. To cover themselves, they include qualifiers, such as the findings “suggest” or “link” a chemical with a health problem or they “may” demonstrate a connection. They will often note serious limitations with their research and rationalize why it is somehow still compelling. Onceone study finds a link, it will be cited in other studies to build yet more “evidence” with weak statistical associations.18 But just as you cannot build a sound structure in the sand, you cannot build a firm scientific body of evidence with a series of weak and largely meaningless research findings. Unfortunately, much of what the government is funding in the name of “children’s environmental health” involves a seemingly endless web of weak studies cross-citing one another and then being cited as justification for government action.
The Children’s Environmental Health Impact Report
The 2017 EPA/NIEHS Impact Report on the children’s health centers provides a telling profile of what these entities are truly about. On the surface, it seems impressive as the report defines the “problem” with worrisome statistics related to children’s environmental health. That is followed by an overview of myriad research studies and discussion of the alleged public health impacts the centers have achieved. But many of their statistics and research claims fall apart under scrutiny, and the reported impact involves mostly political activism rather than validated health benefits.
Statistical Shenanigans. The report opens with infographics that contain lots of statistics designed to impressand worry readers, but the statistics carry little weight, and many are not particularly relevant to the research and claims included in the rest of the report.For example, one infographic states: “Approximately 1,600 premature births per year in the U.S. are attributable to air pollution.”19 This claim is referenced with a study that lists Leonardo Trasande of New York University as the lead author,20 a researcher who often combines his research withpro-regulation advocacy efforts.21 This study is based on the assumption that outdoor air pollution—particularly levels of airborne particles smaller than 2.5 micrometers in diameter, known as PM2.5—increases premature birth rates. The study’s stated objective is described as: “Objective: We aimed to estimate burden of PTB [premature birth] in the United States and economic costs attributable to PM2.5 exposure in 2010.”22 Hence, the goal is not to determine if PM2.5 causes premature birth rates, it is to measure the economic impact of a presumedcause-and-effect relationship.
To achieve this end, the authors simply pulled data from the EPA and other sources that fit their narrative. They then modified that data “to obtain an estimate that better represents the true relative risk.” Apparently, they wanted to make the relative risk estimates— which were quite low and largelyinconclusive at 1.04 to 1.16—stronger. Yet even with their modifications to the data, the relative risk numbers they developed are still too low to draw a meaningful conclusions. Remember, numbers below 3 are weak and not particularly compelling, and the numbers in this study ranged between 1 and 1.4.23
Trasande et al acknowledge other problems with their study in the section where they discuss “limitations”:
The specific components of outdoor air pollution that contribute to prematurity and other adverse birth outcomes remain elusive, as do the mechanisms by which they produce effects. Although it is true that some studies to date have failed to find significant associations with adverse outcomes, exposure imprecision may have biased those estimates (Fleiss and Shrout 1977); others may have had modest statistical power to detect significant differences in prematurity.Therefore, some may argue that the scientific evidence for air pollution has not reached the threshold for causation.24
In other words, the research literature does not show or even theorize how air pollution in the United States— which is relatively low and declining25—could affect premature birth rates. Trasande et al also admit that the other statistical research in this field is either contradictory or equally weak.
The Impact Report includes two other alarming claims in the infographic that can be tackled together. One states: “60% of acute respiratory infectionsin children worldwide are related to environmental conditions.” The other states: “Air pollution contributes to 600,000 deaths worldwide in children under 5 years old.” Both claims may reflect some truth but they have little, if anything, to do with the chemicals discussed in the report or children’s environmental health in the United States. The authors should have known this, but stating things this way helps sound alarms to continue the flow of federal dollars.
Specifically, both of these claims cite a United Nations report that quantifies deaths and health effects related to “environmental threats” primarily in poor nations, including inadequate disinfection of water supplies,insect-borne diseases like malaria, poor air quality related to burning wood and other biofuels in homes without proper ventilation, food borne illnesses related to poor sanitation, and chemicals. Basically, these numbers largely reflect problems in poor nations directly related to poverty and the lack of modern sanitation. They have little relationship to exposure to trace chemicals from consumerproducts in the United States, which is the focus of the children’s environmental health centers. It is highly misleading to use these figures to demonstrate the need to spend money to research U.S. risks, which are completely different.26Questionable Research. The report then follows with sections on “health outcomes” addressing how chemicals might adversely affect or cause asthma, birth defects, cancer, immunity, neurodevelopment, obesity, and reproduction. A subsequent section addresses “environmental exposures,” including things like arsenic, plastics, lead, pesticides, and secondhand tobacco smoke. Much of what is included in each section simply ignores the larger body of evidence to focus on the myriad statistical studies that “suggest” certain chemicals cause adverse health effects. These sections do not address the public health benefits that many of the products they demonize provide to public health.
The section on the chemical Bisphenol A (BPA) is illustrative. Without any regard to its many valuable uses, BPA has been under attack for decades by environmental activists who work in tandem with these federally funded researchers to continue to produceand publicize numerous studies of questionable value. While many studies involve weak and largely meaningless statistical associations, they are usefulto activists who sound false alarms and keep the issue in the news.
BPA is used to make hard, clear plastics and resins that line food packaging to prevent rust and the development of pathogens in food. BPA has enormous value in securing a safe foodsupply27 and improving public health and safety thanks to its use in medical devices, among other uses.28 Humans can consume small traces of BPA from packaging, but the body quickly passes it out via urination.
The Impact Report lists a handful of BPA-related studies conducted by the children’s environmental health centers, noting that, “BPA may contribute to childhood obesity.”29 Yet these studies simply generated inconsistent findings and statistical associations that donot demonstrate cause-and-effect relationships. In addition, they all relied on relatively small samples and highly questionable exposure estimates. In particular, these studies use a handful of one-time measurements, or “spot” measurements, of BPA levels found in the study participants’ urine samples. For example, a study might develop a BPA exposure estimate for a child using a couple of measurements of BPA in the mother’s urine when pregnant and a couple of measurements of BPA levels in the child’s urine years later. But such “spot” measurements cannot accurately measure the long- term BPA exposures of any of the subjects. BPA levels can varysubstantially over time—even over just a few hours—because the body metabolizes it relatively quickly.30 Accordingly, a few spot measurements are poor proxies for actual exposures over several years.
Using these questionable methods, the studies highlighted in the Impact Report managed to tease out associations between allegedly high BPA exposures and obesity among children, but those studies are highly inconsistent. They are all over the map, more akin to a fishing expedition than an objective scientific discovery process. For example:One study found associations between prenatal exposures to BPA and obesity later in life,31while another found the opposite—prenatal exposure was associated with lower body fat among children later in life.32
One study found no association between BPA and obesity in boys.33 Another alleged BPA impacted metabolic hormones that could impact boys’ weight.34
One study found that exposures of BPA measured in a child’s urine correlated with obesity,35while another reported no such association.”36Numerous governmental and other research bodies around the world have dismissed such small studies withdisparate findings as not useful when conducting a review of the literature. After a thorough review of the full body of research, they all have concluded that BPA presents no significant health risks at current human exposure levels and that its benefits exceed any risks. These include the U.S. Food and Drug Administration (FDA),37 the European Food Safety Authority,38 Health Canada,39 the Japanese National Institute of Advanced Industrial Science and Technology,40 and theU.S. National Toxicology Program (NTP).41
Yet, NIEHS and the EPA have continued to fund small studies of questionable value. In addition to having small sample sizes and producing weak statistical associations, most of these studies do not follow good laboratory practice (GLP) standards. GLP involves applying internationally recognized methodsof ensuring data quality control when conducting scientific research,which reduces risks associated with contaminated samples, researcher bias, and data churning, among other problems.42 As a result, activist researchers have been able to generate scary headlines and push the EPA and NIEHS to fund yet more dubious studies.
In 2012, in an attempt to resolve disagreements and controversies, theNIEHS, NTP, and the FDA formed the Consortium Linking Academic and Regulatory Insights on BPA Toxicity (CLARITY-BPA), a joint effort between regulatory agencies and academics to “address research gaps.”43 It included a core study conducted by FDA researchers and 13 grantee university-based studies, all of which would be GLP compliant. The grantees were to use the same samples and data developed in the core study to promote consistency.
The core study, which involved rodent testing, was the largest of its kind. It found that at current human exposure levels, BPA is unlikely to cause adverse health effects. The FDA statement on the research concluded:
Although a comprehensive review of this report, along with future data from other CLARITY-BPA research, will be conducted as part of our continued assessment of BPA safety, our initial review supports our determination that currently authorized uses of BPA continue to be safe for consumers. The report also builds upon the already extensive data collected in the FDA’s 2014 assessment of the safety of BPA.44
The grantee studies are not all yet released, but the core study underscored reasons to doubt much of the university research conducted in the past.However, some grantees disagree and continue to make claims about BPA risks that belie the larger body of science.45
In any case, the CLARITY study was never really necessary because it was already well known that human exposures to BPA via consumer products are simply too low to have any health effects, particularly because humans metabolize BPA quickly and pass it out of the body before itcan have any impact. The EPA has estimated that a safe human dose is0.05 milligrams per kilogram of body weight per day, which agency researchers derived based on levels found safe for rodents and then extrapolated that to a safe level for humans.46 As Michael A. Kamrin, professor emeritus of toxicology at Michigan State University, pointed out in 2004, consumers are most likely exposed to BPA at levels that are 100 to 1,000 times lower than the EPA’s excessively cautious estimated safe exposure levels. He further noted that the research on BPA also shows that exposure levels per body weight are similar for adults and children, which indicates that infant exposure is not significantly higher.47 The European Food Safety Authority has made similar observations, noting that current BPA exposure levels pose little risk to children and even infants.48
Nonetheless, BPA is one of several chemicals in the Impact Reportdeemed to be dangerous, because the researchers maintain it is one of many synthetic chemicals that are so-called “endocrine disrupters.” The Impact Report explains that such chemicals “disrupt” human endocrine systems in a way that causes illnesses ranging from cancer to obesity to developmental issues. Yet synthetic chemicals like BPA are “weakly estrogenic,” which means that simply they are not potent enough—and human exposure is not significant enough—to produce health effects.49 Humans are regularly exposed to naturally forming estrogen- mimicking compounds produced by plants—so-called phytoestrogens—in our everyday diet, and these are much more potent and exposure is much higher. Yet we suffer no ill effects because none of those chemicals, like BPA, are as potent as human hormones.
Phytoestrogens, for example, are found in legumes, with a particularly high level found in soy. Exposure to natural phytoestrogens from food is 100,000 to 1 million times higher than exposure to estrogen-mimicking substances found in BPA, according to data from a 1999 National Academy of Sciences study.50 Researcher Jonathan Tolman noted: “Given the huge relative disparity between the exposure to phytoestrogens as compared to BPA concentrations, the risk of BPA in consumer products appears to be about the same as a tablespoon of soy milk.”51 We havelittle to fear from soy milk, so we have even less to fear from BPA and similar synthetic compounds.Yet BPA and other chemicals deemed “endocrine disrupters,” are a central focus of many children’s environmental health centers, expending millions of taxpayer dollars to study the issue ad nauseum, despite the fact that it’s highly unlikely to add any value to the already extensive body of researchon this topic. By demonizing these products, such efforts can eventually lead to bans and regulations that undermine many of the important benefits those products provide to society. And BPA is just one of many chemicals covered by these groups according to the Impact Report.
Community Outreach/Activism. In addition to questionable statistics and research, the Impact Report also details children’s environmental health center outreach programs. According to NIEHS, each center sets a Community Outreach Translation Core (COTC) to translate “basic research findings into intervention and prevention methods to enhance awareness among communities, health care professionals, and policymakers of environmentally related diseases and health conditions.”52 Each COTC “develops, implements, and evaluates strategies to translate and apply the Center’s scientific findings into information that can be used to protectthe health of children.” While this might sound scientific, it is a facade for engaging in environmental activism rather than advancing science or its application.
Use of the term “translate” appears to be designed to build credibility for COTCs by attempting to link them to an area of study within the biomedical field known as “translational science.” Yet COTCs do not conform to the definitions of translational science.
According to Christopher P. Austin, director of the National Center for Advancing Translational Sciences at the National Institutes of Health, translational science is “the field of investigation which seeks to understand the scientific and operational principles underlying each step of the translational process.”53 He explains:
[T]ranslation is the process of turning observations in the laboratory, clinic, and community into interventions that improve the health of individuals and the public—from diagnostics and therapeutics to medical procedures and behavioural changes.54
However, the children’s environmental health centers’ Community Outreach Translation Cores do not translate scientific findings because they rarely bother to produce science first, so there is no science to translate. Often,they assert highly questionable risks and fail to mention new risks their advice might create. Most important, they cannot be sure to promote public health—which is the goal—because they lack a firm scientific basis. Rather, as examples in the case studies will show, COTCs often start with ideologically derived, predetermined conclusions—sometimes related to the area of study and sometimes not—and then design outreach programs around those conclusions. The research component is secondary and tainted with bias; it has become little more than a smokescreen to legitimize activism.The Impact Report, for example, details many supposed efforts to translate science in ways that educate the public in order to improve public health. But the programs involved do more to mislead and alarm than to inform, and they follow up with calls for government regulation. For example, the report highlights the creation of “culturally appropriate” brochures that “inform” people about the dangers of toxic chemicals to their health and then urges them to take political action. After a few sentences alleging that “toxic chemicals” are causing health problems, one brochure reads: “Support policies that prevent pollution: We need policies that identify existing toxic substances, phase out their use, and replace them with alternatives that are safer forhuman health and the environment.”55 It then offers advice for healthy living that includes such things as: “drive less,” “don’t spray pesticides,” “don’t use chemical tick-and-flea collars, flea baths, or flea dips,” “take off your shoes,” (so you do not track chemicals around the house), “don’t dry-clean your clothes,” “choose glass, stainless steel or ceramic” (to avoid plastics), “select flame retardant-free foam products,” eat organic food, andso on.56
There is no science presented and no discussion of the benefits we trade off to avoid such “toxic” products and activities. In fact, such advice may increase public health risks, which is the opposite of the objective of translation of science. For example, telling people not to use flea and tick control for animals may mean more risk to both pets and humans.57 Fleas carry serious diseases, including typhus and the bubonic plague, while ticks can transmit Lyme disease, Babesiosis (similar to malaria), Rocky Mountain spotted fever, and many other diseases that affect pets and humans. Urging people to eat organic food—which is not any healthier or less risky than conventional produce— may discourage people from eating enoug |
