Poison Spring (21 page)

Despite Nalco’s paying George Cotsirilos hundreds of thousands of dollars, the Chicago jury found all four IBT executives guilty of fraud. Three of them were given brief jail sentences. Calandra stayed out of prison, indicted but free because his lawyer argued that Calandra had had an aneurysm of the aorta that threatened his life.

In the end, the IBT trial—as dramatic as it was—was a rare moment when such corruption was actually brought to light. For pesticides alone, IBT faked the information the government used for the registration or licensing of at least 212 ingredients. This fact is deeply troubling if you realize that the thousands of products made from these 212 “pesticide active ingredients” represent many acutely toxic and widely used sprays in the United States and throughout the world.
10

In 1981, five years after the IBT scandal became public, EPA calculated some of the damage. Nearly 65 percent of the company’s studies were deemed invalid. Data from eighteen laboratories, or about 20 percent, was considered “questionable or unacceptable for regulatory purposes.” The rate of invalidity of pivotal studies “was so high that reliance on the data is risky.”
11

Some time in late 1981, EPA prepared a “fact sheet” on the IBT disaster in which the agency gave credit to the FDA for the discovery of the fraud. “FDA uncovered problems so serious that, after documentation and review of the reports, they refused to accept further IBT studies and insisted that the 40 or so studies submitted to FDA be replaced,” the report said.

“Raw data” (basic information on the material tested and its effects on the lab animals) was “missing” and “shredded.” For some of the data that survived, it was impossible to make certain that it was authentic. It went on. Those who worked at IBT did not fit their job descriptions. The inspectors found “undocumented changes [in study protocols], [and]several protocols [were] used interchangeably.” The inspectors also found very little information on what the lab animals ate, how much, when, or what the food was mixed with. Not only that, but the lab kept “extra animals for the entire test.” This is because it was common in the IBT for the animals to “die twice.” The lab workers used “gang caging” for the animals.

The fact sheet also summarized the frequency of invalid IBT data. For example, in June 1980, all the two-year rat studies were invalid; in December 1981, the failure rate was 95 percent. In other words, of 41 studies reviewed, 39 were invalid. In general, the percentage of invalid studies in December 1981 ranged from 29 to 95.
12

Far more worrisome than this laundry list of tawdry revelations was this simple fact: for about a quarter of a century, thousands of farm sprays that had been “tested” by IBT ended up in the food and drinking water of hundreds of millions of people in the United States and throughout the world. It is impossible to assess the damage of this massive poisoning, but it must surely have been a considerable factor in human disease and death.

IBT’s betrayal of the public trust was spectacular largely because of its magnitude, involving as it did a company carrying out roughly 40 percent of all tests for industrial agribusiness and the chemical and drug industries. But even as this fraud was going on, other, smaller laboratories were engaged in their own versions of faking data and manipulating science.

Around this time, a female scientist told me of atrocious conditions at a veterinary, toxicology, and entomology research laboratory of the U.S. Department of Agriculture in College Station, Texas. The woman referred to Dimilin “tests” that took place in a USDA lab farm experimenting to see whether Dimilin had any effect on male chickens’ levels of testosterone, a hormone vital for normal reproduction.

Dimilin belongs to a group of chlorinated diphenyl poisons known as “insect growth regulators.” Growth regulation is essentially biological warfare: a chemical like Dimilin poisons (and thus inhibits) an organism’s growth or development, often resulting in the disfigurement or premature death of the organism. When industry (and EPA scientists) say Dimilin “decreases testosterone levels in roosters,” what they really mean is “Dimilin sterilizes roosters.”
13

Dimilin, it turned out, was especially deadly during molting. The compound made it impossible for insects and arthropods such as shrimp, crayfish, lobsters, and crabs to build a new exoskeleton; young animals would die because they had no hard tissue in their bodies for walking, flying, swimming, or protection.

But Dimilin also neutered both insects and warm-blooded animals, and in very small amounts. It takes only 75 parts per trillion of Dimilin in water, for instance, to sterilize mysid shrimp, an organism at the very base of the ocean’s food chain.

In the laboratory, scientists had discovered unsettling results: feeding roosters and hens food mixed with even tiny amounts of Dimilin made them different birds. The hens grew fat and huge, the roosters became thin and small with practically no secondary sex characteristics left in their combs and wattles and nearly no sex hormone left in their testes. These remarkable discoveries came out into the open thanks to another USDA man, James E. Wright, an entomologist who published the lab’s findings in 1976 in volume 14 of
Environmental Health Perspectives
.

But these findings were not what industry—or its companions in government—wanted to hear. The USDA did not appreciate EPA inspectors checking up on its scientists; The USDA argued that its findings on Dimilin were in the open literature, so how could it be blamed if companies used such findings to register the chemical with the EPA?

This reasoning was disingenuous at best. The bottom line was that the Texas lab was a mess. For starters, the poultry manager had been out following a heart attack, which meant—at the very least—that the birds did not get fed regularly. “The state of malnutrition during the critical growth period, from 0–35 days, appears to have, upon re-examination, retarded the birds’ growth,” the report said. Many of the birds were overcrowded in their cages, and essentially starved.
14

Laurence Chitlik and Salvatore Biscardi, the EPA investigators who visited the USDA’s chicken lab in Texas, found the studies invalid, since the chickens’ starvation probably diminished the Dimilin effect.
15

Their EPA supervisors were also unsettled. The audit revealed “numerous irregularities in laboratory practice which ranged from conditions which unduly stressed the test animals to possible sabotage.”
16

Bizarrely, none of this caused a ripple in the companies that had used the results of the chicken study to support EPA applications to sell Dimilin as an “insect growth regulator”—which now contaminates beef, eggs, goat, pork, milk, mushrooms, lamb, sheep, and soybeans.

And on it went. In January 1985, an EPA biologist named Jack McCann got wind of pesticide testing data that a defunct laboratory had abandoned in a room at the Reading, Pennsylvania, airport. The data, which the EPA depended on to determine whether a chemical was safe, had been produced in a lab run by a toxicologist named Gaston E. Cannon. The lab had gone bankrupt before August 1983, and the airport, which had rented space to the Cannon lab, was threatening to clear the room unless it got its rent. American Standard Bioscience, the company that bought Cannon, was not exactly rushing to secure these “raw data” documents.

Between 1973 and 1983, Cannon had done more than a thousand studies in support of sixty-nine pesticides, among them some of the largest-selling, and acutely toxic, sprays used by American farmers. Cannon’s clients included many American, British, Swiss, and French giants in the global chemical industry, including Union Carbide, DuPont, Zoecon, Mobil Chemical, Rohm and Haas, Pennwalt, Mobay, FMC, Chemargo, Thompson-Hayward, American Hoechst, Ciba-Geigy, Rhône-Poulenc, Uniroyal Chemical, and Velsicol. Cannon also served government institutions including the U.S. Forest Service, the U.S. Department of Agriculture, the Colorado Department of Agriculture, and the New York State Experiment Station.

Yet such was the state of affairs between EPA, the chemical industry, and the testing labs serving that industry that in January 1985, data from these studies—without which hugely profitable pesticides would not have been allowed on the market—was about to be thrown into a Pennsylvania dump.

That February, Jack McCann told Adrian Gross and me that in his opinion, the Cannon lab might be a “mini-IBT.” By using that magic acronym, the humble, soft-spoken wildlife biologist expressed his fears about the enormity of what might be revealed in a rigorous investigation of the Cannon lab.
17

For years, McCann had had a feeling that something was not right at the Cannon lab. In the early 1970s, he warned other EPA scientists that the science being conducted at that testing facility was extremely poor. In August 1983, Gross and an inspector from the FDA had gone to the Cannon lab to audit an animal study the lab had done some years before for the Pennwalt Corporation. That two-year study had examined the effects on mice of a weed killer, disodium endothall; technicians examined mice tissue to see whether endothall caused cancer in the experimental animals.

Because the inspection of this important study was done at least five years after its execution, Gross and his colleague simply sifted through what was left in the lab in the form of written records or “raw data.” They discovered that the science being practiced in the Cannon lab was atrocious. Lab techs kept five mice in every cage, which caused crowding so bad that some mice ate the others. Several of the mice that survived cannibalism had their ears chewed off.
18

Inside the lab, the magnitude and implications of the research problems were beyond anyone’s guess. The Cannon lab was very small, employing at most four scientists and three technicians. Yet for ten years it “tested” numerous industrial chemicals, drugs, pesticides, food additives, and cosmetics. In 1980, for example—in pesticides alone—the Cannon lab agreed to do 86 animal studies for 19 chemical companies and other organizations. And in 1981, Cannon lab started 42 studies for 12 clients. How could a very modest testing facility like the Cannon lab do all that complex and exacting work and do it right?

The government did not see fit to dig into Cannon’s past. In the mid-1980s, the specter of unearthing another IBT was just too much for the shell-shocked bureaucrats of the Reagan EPA. The similarity of the Cannon case with the IBT case probably scared off any EPA investigation, and ultimately, nothing happened. No one went to jail or paid any fines.

In fact, in the days since IBT and Cannon, the system of testing and licensing most pesticides and other chemicals has not changed. For example, in 2012, about 10 million pounds of TCC per year was used to make antibacterial soaps in the United States alone. Even the fraudulent IBT studies that supported the legal use of this chemical showed it to have deleterious testicular effects on rats. Today, scientific studies are raising concerns that TCC is an endocrine disruptor. But TCC’s continued existence more than half a century after the IBT gangsters “tested” it is an indictment of not merely the EPA but also the chemical industry, which continues to defend it.
19

Thus, regardless of when the shoddy work took place and whether it was just mistakes or criminal wrongdoing, the EPA knew of problems that might well have allowed dangerous toxins into the market, did nothing to remedy that failure, and was more concerned with protecting the reputation of companies than with exploring a public health threat. The bigger, underlying issue, of course, is the inherent conflict of interest in a situation in which private labs, paid by chemical companies, do testing that the government relies on to protect the public. Even if there is no conscious bad faith, there is inescapably a powerful incentive to deliver results agreeable to the paying customer.

Finding four IBT officials guilty of fraud was too light a remedy, and it came too late. IBT’s transgressions were committed at a time when IBT served not just Monsanto but several other chemical giants. The EPA knew that. An audit of IBT had “uncovered practices which not only undermine the integrity of the final reports upon which [EPA] has based far-reaching regulatory decisions, but also the integrity of the raw data underlying the studies,” Edwin Johnson wrote in a letter to other senior EPA officials. “In addition, evidence is accumulating which suggests prior knowledge of those practices by the sponsor(s) of the [fraudulent spray] studies. I know that you share my deep concern regarding the seriousness of the regulatory ramifications of these recent findings of falsification of data upon which national and international regulatory decisions have been made.”
20
But none of these damning implications became part of the government’s strategy in pursuing IBT. The Carter administration decided to let the chemical companies off the hook and focus on IBT alone.

The IBT scandal merely revealed what my EPA colleague David Smithson would call “garbage science.” IBT was not a unique case of scientific fraud––it was emblematic of a dark and deviant scientific culture, a “brave new science” with deep roots throughout agribusiness, the chemical industry, universities, and the government. IBT highlighted why labs are essential to the protection of human health: they are the means by which we evaluate the toxins ending up in the food we eat. So lab integrity is essential, which makes the case of IBT and IBT-like behavior in the chemical industry so abhorrent.

And what about the true science that never got done? In the IBT fraud, nearly half of the animal studies testing the potential mutagenic effect of pesticides were tampered with. The EPA and FDA reviewers rightly concluded that those studies were invalid. But the real significance of this classification is that the makers of the pesticides discovered that their pesticides caused gene damage, so they falsified the studies.
21