6

The Deal with Diesel

diesel engines don’t quit. An old truck, even one that is slow to start and then makes an infernal racket, can still rack up hundreds of thousands of miles, even with minimal maintenance. Diesels are also more efficient and economical than gasoline engines, and far less flammable. Basic accounting therefore dictates that most of our high-mileage modes of conveyance—trucks, buses, railroad engines, and mining, farm, excavation and construction equipment—are powered by old, loud diesel machines.

So why doesn’t the United States have more diesels on the road? For those of us over 40, say, the problem with the old diesel engines has always been self-evident: that cloud of black smoke spewing from the exhaust pipes. Younger generations have gotten a bit of a break, because the latest technology is much cleaner-burning (and quieter), but the trade-off of diesels has always been their stubborn indestructibility in exchange for fouling the air and promoting smog in cities across the globe.

Although knowledge of the health problems associated with diesel exhaust has informed the United States’ auto policy for decades, the industry’s product defense enablers are also still on the job, working hard—even after decades of ever-increasing scientific understanding—to convince the public that the black smoke is harmless to our health: saying in effect, who you gonna believe—us or your lying eyes?

People should believe their eyes, although sometimes diesel exhaust emissions can be invisible and still dangerous. Diesel emissions are a complex stew of gases and particles composed of thousands of chemicals, and none of the stuff is good for us. Inhaled in sufficient quantity, long-term exposure is associated with increased risk of stroke, ischemic heart disease, chronic obstructive pulmonary disease, and lung cancer. The gases in this engine exhaust include carbon monoxide, sulfur dioxide, and a collection of nitrogen oxide compounds referred to as NOx (pronounced “knocks”). The nitrogen molecules are important air pollutants by themselves, causing lung and cardiovascular disease while also reacting in the atmosphere to form ozone, more particulates, smog, and acid rain. They also reduce crop yield and play a major role in climate change (this despite for every mile driven, diesels emit less carbon dioxide—a major greenhouse gas—than gas engines). George Washington University scientist Susan Anenberg estimates that in 2015, diesel emissions were responsible for about 175,000 premature deaths globally, including 40,000 in Europe, 39,000 in China, 36,000 in India, and 11,000 in the United States.¹

For decades, the U.S. Environmental Protection Agency has joined regulators in many countries with smog-bound cities to push the engine industry to design new diesels that put out less NOx emissions, use low-sulfur fuel—sulfur is a contributor to the smog problem—and filter out particulates. The results have been dramatic, but not dramatic enough. Thanks to all the trusty old engines still chugging away on roads and rails, there is still too much diesel-related pollution, and its effects are enormous.

It’s worth noting here that diesel engines are also unevenly distributed around the world. The United States has stronger diesel regulations than much of Europe, and Americans have healthier lungs as a direct result. The Global South, where older diesel engines predominate and regulations are much weaker, is where air pollution is most predictably deadly.

The black element in the smoky clouds that billow from exhaust pipes on diesel engines is a particulate matter called DEP, short for diesel exhaust particulates. (A particulate is what it sounds like: a particle floating in and transferred by a larger cloud or liquid.) Extended inhalation of diesel exhaust emissions increases humans’ likelihood of developing lung cancer. Thanks to experiments with rats, we know the culprit is the particulates in the emissions, not the diesel-exhaust gases. Expose rats to diesel exhaust and they get cancer; filter out the particulates and expose the rats to what’s left, and they do not.

Many of these particulates are tiny—less than one micrometer (1μm) in diameter and able to penetrate deep into the lung tissue. They are primarily but not exclusively carbon, and it isn’t clear exactly which of the many chemicals in this particulate soup are responsible for the lung cancer. Until we know that answer, and until the identified chemicals are controlled by the engine technology, regulators correctly focus on reducing exposure to any diesel particulate small enough to breathe, and to the lowest possible level. One important reason for this is that a range of toxic chemicals stick to the small particles and are carried with them deep into the lungs.

Of all the health hazards posed by diesel exhaust, the particulates have been the subject of the most comprehensive disease research. It almost goes without saying that this research has been the target of industry’s intense pushback. Attack, manipulate, re-analyze, delay, delay some more, repeat; it’s been going on for decades. And since the mid-2000s, Volkswagen and other German automakers have led a concerted push to sell a new generation of diesel-powered cars to consumers, a campaign that turns out to have been premised on bogus claims of lower pollution. (That story is the subject of Chapter 10.) This chapter focuses on those cancer-causing particulates in diesel exhaust, which remain the subject of obstruction in dramatic and amazing fashion.

This science of particulate health impacts isn’t easy. Every air-breathing creature on this planet is exposed to diesel particulates—along with lots of other stuff. Precisely attributing the percentage of an increased cancer risk to the particulates, or any other cause, is challenging for plenty of reasons. With urban residents especially, it is exceedingly hard for scientists to accurately reconstruct 30 or more years of environmental exposure, and then to compensate for other causes of lung cancer, including smoking and asbestos. One approach is to follow specific groups of people who are exposed to diesel exhaust as an occupational hazard over a period of many years: truck drivers, railroad employees, bus mechanics, and underground miners are the classic subjects. Compare their risk of lung cancer with the risk demonstrated by the rest of us who experience less exposure, then try to factor in at least some of the variations inherent within population exposure levels—for example, comparing residents who live near streets and highways (and their soot-belching trucks) with residents of the same city who live at a greater distance from the main emission sources. The results of these intricate comparisons can then be used to estimate increased risk of lung cancer posed to the general public by diesel emissions.

For much of the twentieth century, scientists harbored concerns that DEP increased cancer risk,² but in the absence of hard data there was little to be done about it. Then, in the 1980s and continuing in the 1990s, several studies of diesel-exposed workers reported increased risk of lung cancer in the range of 30 to 50 percent. This may seem like a small increase, but from a public health perspective it was seismic. A 30 to 50 percent increase meant thousands more cases annually, many of them resulting in death.

By 1988, the WHO’s International Agency for Research on Cancer (IARC) had seen enough of the accumulated evidence to classify diesel exhaust as “probably carcinogenic” to humans. That same year, the U.S. National Institute for Occupational Safety and Health (NIOSH), created in 1970 by the same legislation that created OSHA, signed by President Richard Nixon, also recommended that DEP be considered a potential occupational carcinogen. Shortly thereafter, the Mine Safety and Health Administration (MSHA, created in 1977) formally requested that NIOSH perform a risk assessment for exposure to diesel particulates to serve as the basis for a new standard.

About the same time, NIOSH and the National Cancer Institute (NCI) decided it would be worthwhile to study the effects of particulate exposure on one of the most heavily exposed populations, underground miners, who had the highest exposure to diesel emissions of any occupational group. But designing the study was rife with complications. For the purposes of this study, coal and metal mines were excluded, as were any other mines where asbestos, silica, radon, or other carcinogens where present and therefore might confound the results. In underground coal mines, it’s also difficult to distinguish between the respiratory effects of diesel-related particulates and coal dust. Moreover, these mines require forceful ventilation regardless, because any build-up of methane from the coal dust is potentially explosive. Headlines and funerals periodically remind everyone of this fact.

So for once, this was a mining story that was not about coal miners. It focused on the thousands of other miners working in limestone, salt, and other non-coal mines whose exposure to noxious emissions was “just part of the job.” This non-coal non-metal mining population was sufficiently large (more than 10,000 people), and the emissions exposure levels were very high: giant diesel engines powered underground mining equipment, and environmental measurements had been made over decades that enabled the researchers to more accurately estimate actual exposure levels. Moreover, smoking history could be obtained from these workers (or, in the cases of those who had died, family members), so the scientists could control for use of tobacco, another lung carcinogen. Overall, it was an excellent cohort to analyze, and the researchers picked eight mines for the minutest examination: one limestone, one salt, three potash (a potassium-rich salt used for fertilizer), and three trona (a “non-marine evaporite mineral” and the primary source of calcium carbonate in the United States).

Under the best of circumstances, several years are required to conceptualize such a study, lay out in detail how it will proceed, gather the data, and then proceed to final results and interpretation. With the Diesel Miners Study, it took a few years before NIOSH and NCI even had the protocol ready to go. In 1995, they dutifully sent the study protocol out for external review. The mining and diesel equipment industries, both of which had been following developments with the planned study and the circulation of the final protocol, kicked their opposition into high gear. There were no results yet, not even any data collection, but the mere dissemination of the protocol marked a significant step forward for any of the regulatory agencies charged with examining the risks posed by diesel particulates.

The industry pushback was waged by a group of companies working under the auspices of the Methane Awareness Research Group Diesel Coalition (MARG), later repurposed as the Mining Awareness Research Group Diesel Coalition (still MARG—maybe wanting to save the money required for new stationery). MARG embarked on a multiyear campaign to at least delay the study’s progress, if not kill it completely. It’s a familiar first strike in instances where human health interests threaten bottom-line profitability: even quality science is treated with contempt and antagonism. MARG waged an unprecedented, full-court-press legal and political campaign to stop the study of the underground miners. Leading the obstruction was Henry Chajet, an attorney with what was, at the time, one of the main law and lobbying firms in Washington, Patton Boggs. What followed was long and ugly, and it was carefully explored and written up by my former student and then colleague at George Washington University, Celeste Monforton, in an American Journal of Public Health article aptly titled, “Weight of the Evidence or Wait for the Evidence? Protecting Underground Miners from Diesel Particulate Matter.”³

One notable, in fact pioneering, feature of MARG’s campaign is how the mining corporations (along with Navistar International, the heavy equipment and truck manufacturer formerly known as International Harvester) enlisted congressional committees to do the dirty work they could not do on their own. This strategy has been embraced by other industries eager to devalue the science showing the dangerous nature of their products. When there was a Republican majority, the U.S. House of Representatives’ Science Committee (and I use the term “science” advisedly, since most of its Republican members seem to have little regard for the scientific enterprise) led a series of attacks on research that industry didn’t like, scientists that industry didn’t agree with, and scientific agencies and organizations whose pronouncements and reports implicate the products.

In 1998, the Diesel Miners Study was not ready—not even close, as we will see—but the United Mine Workers of America petitioned MSHA for a new standard, pointing out that most earlier studies had found higher cancer risk among diesel-exposed workers, so the health-advocating federal officials at MSHA decided to move ahead and officially propose strengthening the standard protections afforded to underground miners. In response, the mining industry and Navistar International enlisted a slew of product defense experts who could be counted on to highlight and challenge every iota of the existing studies. This was their first tactic in rebuffing and delaying federal action, and it was effective. Some of the methodological flaws they alleged in the earlier studies were arguably fair—but that didn’t mean that the studies’ results and conclusions were wrong. The studies cited by MSHA were unquestionably the most advanced that had been performed to date, and by citing them they were upholding the law as written by Congress at the time of the agency’s creation. The Federal Mine Safety and Health Act of 1977, succinctly known as the Mine Act, requires federal overseers to protect the health and safety of miners and instructs MSHA to consider the “best available evidence” in setting standards.⁴ The agency had identified 47 scientific studies, 41 of which found some degree of association between lung cancer and occupational exposure to diesel particulates. Were these perfect studies? No. But as French Enlightenment writer Voltaire rightly pointed out, the perfect is the enemy of the good.

The product defense industry seemingly does not care for Voltaire. One of their basic strategies is to demand perfection from all research (or at least all research whose findings they don’t like), ignoring the fact that imperfect studies are still useful—and often they are all we have. In 1998, as MSHA sought to strengthen protections for mine workers, the weight of the scientific evidence held that exposure to the exhaust emissions from diesel engines is likely to increase risk of lung cancer. Industry interests held that worker health should take a back seat to a cultural pursuit of more perfect science.

The industries’ product defense teams also took another tack, arguing that the upcoming Diesel Miners Study was a good reason to delay any new MSHA standards for the underground mines. This took nerve: they were praising the value of the study they were in fact trying to sabotage at every opportunity. But the product defense industry has always had plenty of nerve. In this instance, industry argued that the underground mining study “has the potential to fill in many knowledge gaps … [and would] offer definitive data on the actual mining population … not a biased view of various academic studies.” The campaign was even able to insert the following language in the report that accompanied the 1999 House Appropriations Bill: “The [congressional] Committee believes that the promulgation of a proposed rule on diesel exhaust should be informed by the ongoing NIOSH/NCI study of Lung Cancer and Diesel Exhaust among Non-Metal Miners.” By praising a study that would be completed in the distant future (at which point they’d continue trying to kill it), mining and diesel equipment companies got Congress to parrot the industry talking points that called for a stay on any new protections for mine workers. (MARG’s success in the halls of Congress shouldn’t surprise us. I don’t know what the group was spending in the 1990s, but in 2011 alone the group spent $120,000 on lobbying federal lawmakers.⁵)

To top it off, the trolling obstructionists of MARG even stated in public hearings and in writing that they were cooperating with the NIOSH and NCI researchers and, along with the rest of us, eagerly awaiting the results. As Monforton wrote in her closely researched takedown, “These public remarks and written comments neglected to mention their relentless efforts to halt the study.”³

On one front, MARG lobbied congressional committees to stifle the study. On a second, it touted the same study as a great reason to delay new regulations. On a third, it took the offending agencies to federal court with a blizzard of technical claims. And the only claim that eventually succeeded at the U.S. Court of Appeals was a nitpicking one: that NIOSH had failed to file their Board of Scientific Counselors’ charter with the appropriate congressional oversight committee. Of course, this point was wholly irrelevant from a scientific perspective, but it demonstrates that no gambit was too irrelevant or trivial to push to the limit. The appellate court instructed the district court “to determine an appropriate remedy” for the filing error—a pro forma ruling, but one that gave the then-chairman of the House Education and Workforce Committee, William Goodling (R-VA), the opportunity he needed to intervene on behalf of his seemingly favored constituents. Since Goodling’s Committee had jurisdiction over NIOSH, Goodling asked the judge to require NIOSH to provide all data and drafts from the Diesel Miners Study to his committee for review.

In 2000—four years after the initial circulation of the Diesel Miners Study protocol, and a full decade after the study had been first proposed—MARG went venue shopping and found a sympathetic judge who instructed NIOSH to give to the committee “all draft reports, publications, and draft results or risk notification materials prepared in connection with the Diesel Study, for review and approval prior to finalization and release and/or publication and distribution of such materials.” To be clear: a team of government scientists was told to submit its work, midstream, to a team of elected nonscientist politicians so that the politicians could make sure it was all to their liking. As it happened, I had spent a year working as a Robert Wood Johnson Foundation Health Policy Fellow on that same congressional committee a few years earlier, and I can attest from personal experience that the committee staff had absolutely no ability to understand, let alone evaluate, the material Goodling was able to acquire. But a good-faith review had never been the point of this request; the point was to give the material to those at the mining companies, who could in turn gain some leverage to further delay the research.

On January 19, 2001, the day before President George W. Bush was inaugurated and still many years before the mining study would be completed, the Clinton-era MSHA issued new federal regulations requiring a gradual lowering of the permissible level of diesel particulates in all underground mines, metal, nonmetal, and coal. Estimated compliance cost per mine: $128,000 per year. For any reasonably profitable mining company, that is not a budget buster. For many, it’s barely a budget consideration. Nevertheless, and as is typically the case whenever the government issues new health standards, the various impacted industries raised all possible objections in the “public comment” period leading up to the rule’s implementation. And of course, once the regulations were issued, these same industries took the agency to court over them. In this case, the mining industry also received serious relief from the incoming Bush administration—so much so that the ink was not dry on the first legal petition before MSHA delayed enforcement of the exposure limit for one year.⁶

When that new date passed in 2002, the federal regulators again backed away from enforcement. The following year it issued a new set of exemptions and weakened standards. And in November 2003, the industries caught an even bigger break when NIOSH and the National Cancer Institute held a public meeting to discuss where things stood with their long-awaited study of diesel in underground mines.

With MARG and other industry representatives present in full force, NIOSH and NCI announced that data collection for the Miners Study was nearly complete and that analyses were already under way. Industry reps proceeded to request copies of the government’s PowerPoint presentations, which were provided as a matter of courtesy—one that would backfire.

Despite the fact that the slides provided by government scientists clearly noted that the depicted data sets and analyses were incomplete—with no information whatsoever about duration of exposure, level of exposure, and other key factors—they were ripe for cherry-picking in the hands of MARG. And that’s exactly what MARG’s hired gun, an epidemiologist named Gerald Chase, did. His subsequent interpretation claimed that the data from the shared PowerPoint “demonstrates that the initial review of data from the NIOSH study … does not show any excess lung cancers above the expected rate for the general population.”

In its incomplete state, the PowerPoint presentation—a summary of the aspects of the study, not the study itself—demonstrated no such thing, but MARG used the conclusion to argue that the new MSHA standards must be revisited. Not surprisingly, President Bush’s appointees at MSHA complied, establishing a 45-day public comment period on Chase’s MARG-funded report. Scientists at NIOSH were handcuffed by the earlier legislation requiring their communications be submitted first to the House of Representatives for review: in fact, a standing order issued by the federal judge dictated that all public remarks from NIOSH had to be submitted to the House 90 days in advance. As Monforton wrote in her summary of the process, “The government scientists most capable of responding to the MARG-sponsored report were excluded from the process.”³

Even the most cursory reading of this story demonstrates that the industry’s campaign to kill the regulation, even after they’d first taken the teeth out of it, was skillfully orchestrated by Washington lobbyists well acquainted with the ways of both Congress and federal agencies. The business of delaying the key study during the late 1990s was masterfully executed. In the years that followed, the industry had cultivated new friendships in the Department of Labor, which oversees MSHA and OSHA, both regulatory agencies. In contrast, NIOSH and NCI, as scientific research agencies, have traditionally been shielded from interference by political appointees. So, yes, I’m prepared to believe that MSHA knew that NIOSH could not legally defend its own work. I know that as a result of all these lobbyist “sound science” shenanigans, top-drawer legal and lobbying work, and good electoral fortune, an alliance of mining firms, led by the MARG Diesel Coalition, successfully used the courts and Congress to game the regulatory system for a decade. Other industries can boast about uncertainty campaigns that have lasted much longer, but MARG certainly deserves dishonorable mention.

But put a big asterisk on the prize, because all that obstruction notwithstanding, new modified standards did nevertheless go into effect over the next few years, under the auspices of the Bush administration. An agency can delay only so long; if the Bush administration’s MSHA had wanted to kill the diesel rule, it would have had to go through the same standard-setting process as the agency did to enact the rule in the first place. The new legal limits on diesel particulates were still too high to eliminate cancer risk, but they did reduce the permissible exposure levels by about 80 percent. Michael Wright, head of health and safety for the United Steelworkers, which represents many miners in metal mines, characterized the improvement cannily: “It used to be like working in the tail pipe of a bus; now it’s like working four or five feet behind the bus.”

So in the end, MARG failed to stop the new MSHA standards, and it also failed to stop the Diesel Miners Study. In 2010, a somewhat unbelievable twenty years after planning for the underground-mining study began, the first results were in, and a series of papers was finally published.⁷ (One of the lead scientists estimated that the industry’s delaying tactics accounted for five of those years.) Once again, the pertinent industries and their prominent product defense consultants geared up for the fight. The initial published papers aimed only to report on the methods employed in the study; MARG’s experts, of course, judged them to be all wrong and incomplete.⁸ The mining lobbyists headed to court, and a federal district judge in Louisiana gave them (and the House Committee on Education and the Workforce) the right to a 90-day review of the study’s results before any publication. The Justice Department appealed, and the Court of Appeals in the Fifth Circuit stayed the order. But by then the 90 days had passed, and MARG could no longer turn to the court for help. More importantly, the Democrats had gained control of the House of Representatives and were eager to see the studies published.

That didn’t stop the mining companies, though. Their attorney, the aforementioned Chajet from Patton Boggs, made a last-ditch effort to slow publication of any future reports. Since he evidently didn’t know in which publication the studies were going to be published, he sent a threatening note to the editors of at least four scientific journals, warning against “publication or other distribution” of data and draft documents, threatening unspecified “consequences” if they failed to comply.⁹

This attempt to exert prior restraint, reminiscent of Richard Nixon’s ill-fated efforts to stop the Pentagon Papers, received widespread coverage within the field and went over badly with the scientific community. The journal Science quoted Trevor Ogden, editor of the journal Annals of Occupational Hygiene: “Despite our attempts to be neutral on various controversies, this journal has more frequently been accused of being on the employers’ side. However, I am disgusted by the many actions being taken to delay [the Diesel Miners Study] publications and prevent their being open to public examination.”¹⁰

No one should be surprised to learn that the published research confirmed the link between exposure to diesel particulates and risk of lung cancer. Overall, covered miners in the subject period were about 25 percent more likely to develop lung cancer than non-miners living in the same state, and the most heavily exposed miners, those who had worked underground for longer periods, were three times more likely to get lung cancer than the least exposed.⁷ I don’t imagine the industry and its lawyers and consultants were surprised, either. The earlier studies showing the association may not have been as methodologically strong as the Diesel Miners Study, but nor were they fly-by-night jobs.

While the mining industry took the lead in opposing the research, they weren’t the only industry with a financial stake in the results of the Diesel Miners Study. The diesel engine manufacturers had been closely following developments, too. Even before the findings were published, this industry, led by Navistar International, had commissioned product defense scientists to produce a series of “weight of the evidence” literature reviews.¹¹ Again, to no one’s surprise, these studies added no new evidence, but they criticized the studies others had done, and they all pretty much reached the same set of conclusions: There isn’t sufficient evidence linking particulates with lung cancer in humans. The studies that found increased cancer in diesel-exposed workers were flawed. Cancer in lab animals exposed to diesel particulates? The rats developed cancer because their lungs were overloaded with dust, not because particulates cause cancer, because they don’t. Keep moving. Nothing to see here.

The publication of the state-of-the-art Diesel Miners Study results was a big setback for such an argument. There had been problems in the earlier research, but this much more authoritative study was a game changer. Moreover, in the same time frame as this study, the National Institutes of Health initiated funding of a study of a different cohort of workers with occupational exposure to diesel emissions: the trucking industry, whose drivers and mechanics were subjected to lower emissions levels than the underground miners, but still significant exposure. And this was a much larger group of workers; the study focused on 30,000 of them. It was designed and run by a team of public health scientists from Harvard, Tufts, and the University of California–Berkeley. Their results also found increased risk associated with cumulative exposure to diesel particulates.¹²

Both reports were major news items, or at least as major as occupational epidemiological studies ever are. The twin results also prompted a reclassification of diesel particulates by the World Health Organization’s cancer-study wing, the International Agency for Research on Cancer (IARC). Instead of “probably carcinogenic to humans,” diesel exhaust particulates were now designated as a full-blown human carcinogen.¹³ This change would trigger additional regulation and likely opened the engine manufacturers to lawsuits. The announcement shook the industry and prompted a wave of industry-sponsored studies meant to produce different conclusions. (One of them, by Volkswagen, is part of the Dieselgate scandal detailed in Chapter 10.)

On top of all that bad news, there was more coming out of OSHA (which I was at that point leading on behalf of the Obama administration). Once the Diesel Miners Study was finished, it was OSHA’s job to use those results to try to reduce workplace exposures, even though we had no exposure standard we required employers to meet. And our concern wasn’t limited to just cancer risk; as noted earlier in this discussion, diesel emissions had been linked to numerous other diseases and conditions as well. Collaborating with our sister agency MSHA, we issued a hazard alert that advised workers and employers that prolonged exposure to diesel exhaust emissions can increase the risk of cardiovascular, cardiopulmonary, and respiratory disease, as well as lung cancer.¹⁴

Now facing these federal challenges to their existing ways of business, the vehicle and engine manufacturers took over from the mining industry as the leaders in challenging the emissions science. Responding to IARC’s “carcinogenic” classification, the manufacturers made two arguments. One was valid and understandable but not relevant; the other, in my view, was simply wrong.

The legitimate point was made by the title of one of the industry’s publications: “Evaluation of carcinogenic hazard of diesel engine exhaust needs to consider revolutionary changes in diesel technology.”¹⁵ For more than a decade, while downplaying the health problems, the engine industry had been making remarkable progress in improving diesel technology, lowering the potency of the emissions and therefore the dangers resulting from extensive exposures to both particulates and the nitrogen oxides. Each generation of diesel engine, especially the large ones designed for trucks and trains and heavy equipment, has been less-polluting than the previous one. The advances had reduced particulate exposures by 98 percent in just three decades, and therefore reducing the carcinogenic potential of the exposures as well. The EPA, along with the State of California (a force to reckon with, powerful enough to issue and enforce its own anti-pollution rules) and European regulators had all insisted on improvements. The industry now argued that the Diesel Miners Study—and all of the earlier ones, for that matter—that had prompted the IARC’s new designation and MSHA’s new standards had studied only workers exposed to the old technology. True enough, but the refutation was simple: Those old, dirty diesel machines didn’t just disappear. The outmoded engines giving off particulate emissions were (and are) in use everywhere, spewing soot and, according to the best available evidence, causing cancer. And as I discuss later in this chapter, the rat studies launched at least in part to counter the IARC classification don’t provide the reassurance the industry proclaims.

The engine manufacturers’ second argument was the default, all-purpose objection we have seen time and again: the studies were weak science that yielded wrong results. The trade association representing diesel polluters insisted on access to the raw data for the studies on truckers and underground miners, essentially asserting that the researchers who did all the work—designing the study and collecting and analyzing the data—got the results wrong. By employing this tactic, industry aimed to do more than just criticize or rework the data analysis performed by government scientists: it also aimed to remove the observational studies of health effects of air pollution from consideration in regulatory settings. Game, set, match. Should the data be delivered, the industry insists on the prerogative of commissioning one or more re-analyses by outside parties, which is to say their own scientists and consultants who specialize in product defense, and manufacturing uncertainty, under the guise of “sound science.”

Eventually the diesel industry did bring their own re-analysis team, as we will see, but first the researchers involved in both studies agreed to share their materials with the Health Effects Institute (HEI). This is a unique institution, chartered by Congress in 1980, that receives half of its funding from the EPA and half from industry (including the engine manufacturers). The institute does science while steering clear of policy pronouncements. And it generally conducts its studies with rigor and transparency.

In its relatively short history, HEI has been no stranger to arbitrating differences between public and industry interests, particularly as they relate to air pollution. The landmark Six Cities Study, conducted by scientists at the Harvard School of Public Health, found that residents in Steubenville, Ohio, the city in the study with the most air pollution, had a 25 percent higher “all-cause” (overall) mortality risk than residents of the study’s least polluted city: Portage, Wisconsin. The study found that air pollution particularly increased the likelihood of dying from lung cancer or cardiopulmonary disease.¹⁶ In 2000, HEI confirmed the Harvard scientists’ findings, and in turn the EPA used the study as the basis for regulations under the Clean Air Act aimed at reducing health effects associated with air pollution. Initial projections estimated that the new rule would prevent 15,000 premature deaths, 250,000 asthma attacks, 60,000 cases of bronchitis, and 9,000 hospital admissions every year.¹⁷

As was the case with the Six Cities Study, HEI’s rigorous, independent re-analysis confirmed the findings of the Diesel Miners Study, asserting it “provided results and data that provide a useful basis for quantitative risk assessment of exposure in particular to older diesel engine exhaust.”¹⁸

That should have ended the controversy, but, of course, it didn’t. The offended industries still wanted the raw data and could obtain them thanks to Senator Richard Shelby (R-AL). The sordid history of this episode came to a head with a mere four lines of text, no more, cleverly attached by Senator Shelby to the 920-page omnibus appropriation bill for fiscal year 1999. Called the Data Access Act, it is also known as the Shelby Amendment (I always like to give credit where credit is due.) This amendment guarantees public access, by way of the Freedom of Information Act (FOIA), to “all data produced” by federally funded research scientists employed by nonprofit institutions. The Shelby Amendment was an open invitation for anyone to use the Freedom of Information Act to harass scientists, question their work, muddy the waters, delay action, and perhaps even steal intellectual property. The idea was to institutionalize these strategies—to construct bureaucratic mechanisms with which corporate interests can question the science underlying not just regulation but virtually any “‘information’” disseminated by federal agencies as well. It would be the very triumph of uncertainty.

Even more galling: the disclosure stipulation does not pertain to private studies paid for by corporations. According to the logic of this legislation, industry should be free to dredge and manipulate the data of government-funded work, but federal agencies and outside groups should not be free to reanalyze industry-sponsored research submitted to the government agencies during the regulatory process. Agencies are legally required to consider these critiques in regulatory proceedings, essentially weighting the scale to favor one set of analyses over the other. Do we require better evidence of the hidden agenda behind the Shelby Amendment?

According to published accounts at the time, the motive behind the Shelby scheme was the displeasure of the corporations most responsible for air pollution—mainly oil companies and coal-burning electric utilities—that did not have access to the raw data at the heart of the otherwise bulletproof Six Cities Study, science that now possessed the seal of approval from HEI. That was the reporting at the time, but it should not come as a complete surprise that recent research using the documents discovered through a certain high-profile court case has pinpointed the actual origins of the Data Access Act. Hint: they were not the usual air-pollution suspects. No, the Shelby Amendment was dreamed up by an industry so besmirched in the public eye that few legislators would support anything it sponsored—an industry that needed other industries to front for it. That’s right: Big Tobacco used air pollution controversy as a cover to advance its own agenda, having realized long before that reanalyzing a study’s raw data in order to change its conclusion was a particularly effective way to neutralize dangerous science.¹⁹ This discovery about tobacco’s culpability didn’t surprise me. The cigarette manufacturers are often lurking in the background in the campaign to manufacture doubt.

One final updating note: The Shelby Amendment did not turn out to be a complete victory for industry. After receiving thousands of comments and complaints from scientists, research institutions, and public policy experts, the Clinton-era Office of Management and Budget (OMB) interpreted the famous four lines rather narrowly, “limiting requested data to published or cited research used by the federal government in developing legally binding agency actions.” This gave government-funded scientists, including the ones at NCI and NIOSH, the ability to limit data sharing so re-analyses could be accomplished without jeopardizing the confidentiality of study subjects.²⁰ This may have been only a temporary setback; polluting industries want far more—and have been unrelenting in their efforts (as I detail in Chapter 13).

Even though they had signed on to the process, the diesel exhaust industry (the engine makers, trucking companies, auto and truck manufacturers, and the petroleum industry) rejected the findings of the HEI evaluation of the Diesel Miners Study. Roger McClellan, one of the industry’s long-time consultants and coauthor of numerous negative reviews of emissions science, issued a new report disputing HEI’s conclusion about the mining study’s validity. He was commissioned by the industry, specifically a trade organization called the Industrial Minerals Association–North America, but McClellan was paid through its law firm, Crowell and Moring.²¹ (This legal maneuver was perfected back in the early days of Big Tobacco. When money flows through a law firm, all communications between the parties can be protected from discovery in a lawsuit.)

Thanks to the Shelby Amendment, the diesel industry was also able to work out a limited data-sharing arrangement with the NIOSH and NCI researchers, obtaining their raw data but pledging to maintain the confidentiality of the individual study subjects. For the re-analysis job, industry hired Exponent, one of the leading product defense firms, along with several other scientists who had done similar works for other industries that wanted to defend their dangerous products. And the diesel exhaust industry got exactly what it wanted: a group of new papers re-analyzing the mining study (with no new data, of course) and exonerating the diesel particulates of any crimes against human health.²²

As should be clear by now, I am always suspicious of these post hoc analyses, or re-analyses, undertaken by investigators who already know the study’s results. When studies are first proposed, the investigator must lay out the data analysis plans in advance. But once you know a study’s results, it is easy to design a re-analysis to make those results—if they are positive—go away. Change some parameters, select new cut-off points between categories, and statistically significant differences suddenly disappear, estimates of risk are suddenly reduced. Such alchemy (known in the trade as data dredging) is rather easily accomplished by a competent (if unethical) statistics craftsman, whereas the opposite—turning insignificance into significance—is far more difficult.

As I write these words in late 2019, seven years after the findings of the Diesel Miners Study were finally published, the industries’ effort to re-analyze the investigation is still going strong. We now stand at five critiques and counting. This work doesn’t come cheap. I’m not privy to the hard numbers, but these papers no doubt cost hundreds of thousands of dollars at a minimum. But how is this a problem, since the work is supported by many of the world’s largest and most powerful corporations? Here is the funding acknowledgment for just one of the five studies:

A coalition of trade organizations working through the Truck and Engine Manufacturers Association (EMA):

American Petroleum Institute (API)

European Automobile Manufacturers Association (ACEA)

American Trucking Association (ATA)

International Organization of Motor Vehicle Manufacturers (OICA)

Alliance of Automobile Manufacturers (Alliance)

European Research Group on Environment and Health in the Transport Sector (EUGT)

Association of Equipment Manufacturers (AEM)

Association of American Railroads (AAR)

European Association of Internal Combustion Engine Manufacturers (EUROMOT)²³

Where will all this reanalysis and revisionist “sound science” lead? I’m confident what to expect because we’ve seen this drama before—multiple times and with many of the same actors. First, the product defense consultants will continue to produce impressive-looking (at least to nonexperts) studies exonerating diesel emissions. I had to laugh when I read one paper from 2012 concluding that “the weight of evidence is considered inadequate to confirm the diesel–lung cancer hypothesis.” That paper’s lead author was John F. Gamble, who discloses that his work was supported by “CONCAWE (CONservation of Clean Air and Water in Europe), a European trade association of oil companies working on environmental, health, and safety issues in refining and distribution.”²⁴ In 1998, Gamble, an Exxon employee at the time, published a similar critique of the Six Cities Study of air pollution, asserting that “the weight of evidence suggests there is no substantive basis for concluding that a cause-effect relationship exists between long-term ambient PM_2.5 [very small particulates] and increased mortality.”²⁵ In fact, hundreds of studies have since demonstrated that Gamble was mistaken about the deadly impact of PM_2.5 exposure but he is still carrying water for the oil industry.

So, first is the inevitable parade of mercenary studies and re-analyses, but while the hired guns are reanalyzing the benchmark Diesel Miners Study for the umpteenth time, new scientists are publishing new studies, involving new cohorts in different parts of the globe, linking this exhaust with lung cancer. Four years after the IARC classification of diesel emissions as a carcinogen, two European panels, the Nordic Expert Group for Criteria Documentation of Health Risks from Chemicals and the Dutch Expert Committee on Occupational Safety together, issued a document reviewing hundreds of studies on the effects of exposure to diesel. This group had the advantage of a few extra years of new research and access to the concerns and commentaries published by industry experts. Their conclusion: “There is extensive epidemiological evidence for an association between occupational exposure to diesel exhaust and lung cancer.”²⁶

Second, and only a few years from now, I predict, the industry will simply give up and acknowledge that emissions from their old engines cause lung cancer, and probably bladder cancer as well.²⁷ All those expensive, disingenuous studies will be seen the same way we look at the tobacco industry’s sponsored studies of decades ago: attempts by dirty industry to sow scientific confusion. That impressive list of trade associations is fighting a rear-guard action, destined to lose. My guess is that the firms will persevere for as long as possible in an attempt to limit legal liability, since, after all, thousands of individuals have likely been sickened by the diesel exhaust from the many old engines still in use on our roads and in underground mines.

The good news: diesel engines keep getting cleaner and cleaner, remarkably so. They put out fewer emissions, and all our lungs are better for it. Some engine companies, like Cummins Engine, are not only developing methods to dramatically reduce emissions, they have embraced the public health regulations that EPA and other agencies have issued that require reduced emissions. This has been a good business decision for them, and I don’t say that cynically. They recognize that they can do well by doing good, and they are playing a very positive role in cleaning the air.

As much as we would like to, we can’t yet conclude that the soot coming out of the new engines doesn’t cause cancer. It would take years to conduct the epidemiologic studies on the long-term effects of exposure to new diesel engines, of course. Cancer from environmental exposures doesn’t appear until decades after first exposure, and the new engines haven’t been around for very long, so any cancer seen now would be unlikely to have been caused by exhaust from the new engines.

So, for the moment, the best we can do is animal studies, and while the results are promising, they are not the exoneration the industry is claiming. After IARC concluded diesel particulates were a human carcinogen, HEI and the industry contracted with the Lovelace Respiratory Research Institute (which comes up again in Chapter 10), a private nonprofit laboratory in New Mexico that had done some of the earlier studies showing that exposure to high levels of DEP gave rats lung cancer. The scientists at Lovelace built exposure chambers in which rats would breathe exhaust from new heavy-duty diesel engines. The exposed rats were no more likely to develop cancer than ones who just breathed filtered air. While this is certainly good news, I don’t think it is the proof the industry claims it is. To understand why, you need some background in how these studies are designed.

The operating theory underlying animal cancer studies is that if a chemical causes cancer in animals, it is likely to cause it in humans as well. When toxicologists set up animal studies, they try to give the animals as high a dose as possible that won’t cause acute or immediate effects, because they can’t perform a study large enough to detect the effect of a smaller dose. Let’s say a component of air pollution causes cancer in one in every thousand people exposed at the level most of us are breathing it. That would be considered a public health disaster and eliminating that substance would save countless lives. But to prove that exposure level caused cancer in lab animals, you’d need a study involving thousands of animals. Instead, toxicologists generally give very high doses to a far smaller number of animals and compare the rate of cases with that found in a group of unexposed animals. If the difference is statistically significant (and you still need a pretty sizable group to find statistical significance) then you can say the material is an animal carcinogen.

The fatal flaw with the Lovelace studies of the new technology engine exhaust is that the particulate exposure level tested was very low. The old engines put out 100 times more soot than the new ones; the rats in the new engine studies were exposed at low levels, and, happily, didn’t develop cancer. From a public health perspective, however, the Lovelace investigators asked the wrong question, or at least used the wrong exposure levels. We aren’t interested in the effects of exposure to the exhaust from a single engine, since no one is exposed to only one engine. We want to know if the accumulation of particles from thousands of engines in our environment increases cancer risk, in other words, do higher levels of new engine diesel particulates cause cancer. And even if the particulate levels in our urban air are lower than in the past, partially because these new engines are much improved, it is still necessary to expose the rats to higher amounts of emissions, or use many, many more rats to rule out a small but important increased risk of cancer. So, while the Lovelace study shows that the engines are safer than the old ones, they don’t prove what is coming out of these cleaner engines is truly safe.

The improvements in new diesel engines should not deter federal agencies like EPA, MSHA, OSHA, and the other regulators from continuing to push for cleaner air. MSHA, for one, hasn’t given up trying to increase protection for underground miners, or at least it didn’t under the Obama administration. In July 2016, the agency reopened its rule-making process—asking stakeholders to provide information on a range of subjects necessary for it to update its 15-year-old standard. The response from the mining industry was vociferous and not surprising: Don’t do it. An attorney from the aforementioned Crowell and Moring, representing Murray Energy (the same firm responsible for the Crandall Canyon mine disaster, where six miners and three rescuers were killed²⁸) and the Bituminous Coal Operators’ Association, wrote: “the Companies want to state categorically that they believe the current MSHA diesel-exhaust related rules are more than amply protective of the health of their employees. There is no need to engage in a new rulemaking on this issue for underground coal mines.”²⁹

The coal operators can relax. David Zatezalo, former coal company CEO and the Trump administration’s appointee to head MSHA, essentially put the whole process into deep freeze, extending the information-collection period (remember, it opened in July 2016) to March 26, 2019. That ridiculous delay guarantees that no new standard will even be proposed until after the 2020 presidential election.

At the EPA, an unbelievable development in 2018 put much of America’s cleaner-air progress in jeopardy. This story began in 2010, when the latest EPA diesel emissions standards took full effect. In order to skirt the new rules, a small number of diesel truck dealers came up with the idea of buying engineless tractor truck bodies from the various manufacturers, installing rebuilt engines from the bad old days, then claiming that these “gliders” predated the latest EPA emissions standards and should therefore be exempt from those standards. (A few such trucks that had been retrofitted with rebuilt diesels had been around for years—they were a good way to take advantage of an engine that could be salvaged after a crash had ruined the rest of the truck—but after 2010 they offered a very tempting way to get around the new emissions standards.) Business started booming, from fewer than 1,000 gliders sold in 2010 to 10,000 five years later. EPA investigators became alarmed by the shape of the curve on that graph—it would seriously imperil the nation’s atmosphere—and announced that henceforth no manufacturer could produce more than 300 gliders annually. With such a restriction in place, total annual production would be limited to a small fraction of the quarter-million trucks sold each year in the United States—if the businesses survived at all.

In 2015, House Representative Diane Black (R-TN) introduced legislation to annul the Obama administration’s limit and expand the loophole for as many gliders as the market would bear. That legislation went nowhere, but in 2017, following the election of President Trump, Black went straight to Scott Pruitt, the new head of the EPA, and requested that he overrule his professional staff and allow unlimited glider production.

And this is where the story gets really good. Black attached to her request a 2016 study produced by Tennessee Technological University concluding that the rebuilt old engines were as clean-burning as the very latest diesels. Soon enough it was discovered that this study had been funded by the Fitzgerald family business in Tennessee, which also promised to build a new research center at Tennessee Tech. Various Fitzgerald interests also contributed at least $225,000 to Black’s unsuccessful campaign for governor. (During his own campaign, Donald Trump visited one of the Fitzgerald dealerships, which sells baseball caps emblazoned with the slogan “Make Trucks Great Again.”)

In November 2017, EPA chief Pruitt announced that he was proposing an exemption to the 300-unit limit for three manufacturers, one each in Indiana, Iowa, and Tennessee (Fitzgerald Glider Kits, the largest manufacturer).³⁰ He noted it was in response to a request from Fitzgerald, who also provided a copy of a letter from the president of Tennessee Tech, touting the results of the Tennessee Tech report (although never mentioning that Fitzgerald made payments to the University).³¹ Only days later, professional EPA staff produced an analysis that challenged any conclusion that the gliders are no more polluting than the new systems. Quoting New York Times reporter Eric Lipton whose work revealed much of this subterfuge, “The analysis said EPA tests found that the Fitzgerald trucks emitted nitrogen oxide levels during highway operations that were 43 times as high as those from trucks with modern emissions control systems. The air pollution from these glider trucks was so bad that one year’s worth of truck sales was estimated to release 13 times as much nitrogen oxide as all of the Volkswagen diesel cars with fraudulent emissions controls, a scheme that resulted in a criminal case against the company and more than $4 billion in fines.”³² The EPA report noted that its testing equipment broke down in urban traffic. “The filters were overloaded with particulate matter,” the report stated. In the accompanying photo, a filter that had been white when new was solid black. Chet France, former director of assessment and standards at the EPA’s Office of Transportation and Air Quality, noted that there are enough indestructible diesel engines in salvage yards to support a prosperous glider market for decades.

As the story unfolded in 2018, a study published in the Journal of the American Medical³³ Association (JAMA) found that repealing the “glider” restrictions would yield 41,000 premature deaths over a decade and 900,000 cases of cases of respiratory ailments. Some of the faculty at Tennessee Tech were concerned that the questionable report, which had gained national notoriety, reflected badly on them and their school. The interim dean of the College of Engineering wrote that the report had been primarily the work of a graduate student. His letter continued, “No qualified, credentialed engineering faculty member (1) oversaw the testing, (2) verified the data or calculations of the graduate student, (3) wrote or reviewed the final report submitted to Fitzgerald, or (4) wrote or reviewed the letter submitted to Diane Black with the far-fetched, scientifically implausible claim, that remanufactured truck engines met or exceeded the performance of modern, pollution-controlled engines with regards to emissions.”

Within days, the university’s president advised the EPA that “experts within the university have questioned the methodology and accuracy” of the study and advised the agency not to take the report under consideration. The EPA replied that the glider decision “only noted the existence of the study” but did not rely on it. Pruitt claimed it did not have the authority to regulate the production of the rebuilt gliders. That was patently false, as demonstrated by the eventual decision of his successor to reverse Pruitt’s decision without taking any public input.³²

Putting aside the health and environmental impacts of deregulating gliders, the surprising aspect of this story is how Pruitt, with the support of the Trump White House, was willing to go against the wishes of some major players in the U.S. economy: trucking companies and engine manufacturers, both of which lost out as “new” gliders used junkyard engines rather than expensive new engines. Anti-regulatory zealots, like self-proclaimed junkman Steven Milloy (so called because he’s the proprietor of junkscience.com and author of a book titled Green Hell: How Environmentalists Plan to Control Your Life), who long advocated for Big Tobacco, seemed to be more able to advance their positions than large corporations. I attribute that to what I saw as President Trump’s desire to reverse anything the Obama administration did—if Obama supported it, Trump would do the opposite, no matter what the consequences.

Once Pruitt was forced to step down because of multiple ethical scandals, the glider deregulation could no longer survive. Tennessee Tech sent another letter to the EPA reporting that it had conducted an internal investigation that found that the key statements in that first letter were inaccurate. Andrew Wheeler, a former corporate lobbyist and a more traditional Republican operative who replaced Pruitt, killed the initiative. Did the embarrassment of using phony results from a study paid for by the firm that stood most to benefit influence Wheeler’s decision? I think not, since they have embraced many other studies, equally bogus. I suspect the large firms that make and use new, cleaner diesel engines pointed out to him that not only would all these new gliders pollute the air and kill many Americans, it hurt their bottom lines. While the first argument should have swayed him, it is far more likely the latter did. Either way, he made the right decision, and killed the glider deregulation.