Just about everyone I meet is surprised to learn that the chemicals in use all around them undergo almost no safety testing whatsoever. They assume that government agencies are carefully testing every chemical used and regulating any that might be dangerous. As you’re about to see, the truth is something quite different from public perception.

What are chemicals tested for?

When it comes to chemical testing, most of the time companies needn’t prove their chemicals are actually safe for human health or the environment. They simply need to submit data on how it will be used and different acute toxicity levels, usually by giving high doses to lab animals. In essence, it’s more a matter of testifying to the fact that normal chemical use won’t lead to every child within a 20-mile radius dropping dead, or it won’t kill every living thing in a given area.

But when it comes to actual long-term safety or a chemical’s impact on human health, companies are only required to conduct safety testing AFTER evidence of explicit harm has surfaced, and only if the EPA requires it.

The inadequacies in chemical safety testing

A very small percentage of the chemicals we’re swimming in – in our air, our water, and our food – have been tested for all the things that they can do to us.”

– Rick Hind, legislative director at Greenpeace (Engber, 2013)

It took the EPA 11 years to develop tests to screen chemicals, and this was done “under the eye of corporate lawyers.” (Shabecoff & Shabecoff, 2010, p. 51) It shows. The ‘safe until proven toxic’ standard means that under normal circumstances, chemicals come onto the market with almost no information about potential health effects. “The manufacturer is under no obligation to study a chemical’s effects on people or the environment,” note Philip and Alice Shabecoff. “The company is required only to identify the chemical and its structure and to offer an estimate of its production volume, a number they’re allowed to change later.” (ibid, p. 47) Since this information carries no real-world consequences, companies will typically provide the most self-serving numbers and information possible.

Whether or not a company chooses to include any rudimentary screening tests they’ve conducted is entirely up to them, and as you might imagine, companies aren’t in any hurry to turn up potential health consequences. History has shown that when companies do know about potential harm from their product, they’ll typically hide it, which isn’t very hard in a voluntary, self-reporting regulation system. And since the actual structure of the chemical is typically hidden behind patent laws and considered a proprietary secret, the EPA has little to go on except for what the company provides.

This means that the safety data on chemicals could best be described as one big black hole. Since so much is unknown, the EPA is often forced to rely on rough computer models of molecularly similar chemicals in order to guess at how it might behave – if such computer models are even available in the first place. This process leaves a number of glaring problems:

  • Chemical testing standards entirely sidestep the biggest and most prevalent problem, which is the effect of low-dose exposure to toxins over extended periods of time. This is the way most people are exposed and where independent research is turning up serious effects, yet regulators focus exclusively on acute toxicity; i.e., how much you’d have to be exposed to in a single setting to be killed or seriously injured.
  • Both regulators and the industry itself typically only consider a compound’s active ingredient when it comes to safety. Yet chemicals can behave quite differently when mixed with other substances or natural elements in the environment, which is what happens in virtually every case. There’s no effort devoted towards how these chemicals might interact with one another.
  • One analysis found that even basic information was missing on 92% of the chemicals produced in the greatest volume in the S. (Shabecoff & Shabecoff, 2010, p. 46)
  • The various industrial chemicals, which are those used in the manufacturing process as opposed to the product (but which often inevitably make their way into the environment) are almost never tested for safety. (ibid)
  • Of the top 3,000 chemicals most widely produced in the S., only 12 have been “adequately tested” for their effects on the developing brain, according to doctors. (Szabo, 4-25-2011)
  • Of the 80,000 or so chemicals used in S. consumer products and industrial processes, only 300 of those have ever undergone significant testing, and just 5 have ever been restricted. (Fischetti, 2010)


Who conducts chemical testing?

Back in 1972, revisions were made to the Federal Insecticide, Fungicide & Rodenticide Act that changed the way chemical research was done. It required companies to shoulder the burden of such research, with the idea that taxpayers shouldn’t be required to cover the cost. Yet like many things in government, it wasn’t completely thought through. The unintended consequence was that now all studies submitted to the EPA, by law, had to come from the manufacturer of that product. In fact, the EPA is expressly prohibited from conducting its own safety tests on a chemical.

“It’s the fox guarding the henhouse,” says Ramon Seidler, a former senior research scientist who was in charge of the GMO Biosafety Research Program at the EPA. “And the fox is the one collecting the eggs and bringing them to the regulators.” (Morris, 2015)

The problems with industry-funded chemical research

Greg Loarie, a staff attorney for Earthjustice, says industry funded studies are “an inherent problem because it’s very easy to spin these things in a million directions. There are ways in which you can downplay the negative and prejudice the outcome.” As Alex Morris adds, “In fact, the greatest indication of what a study will find is often who is conducting or financing it.” He says that while doing research for his own article on the effect of neonic pesticides on bees, “A press contact at Syngenta sent me studies that ostensibly showed that neonics were not harming bees: The first was conducted by Syngenta employees; the second was funded by Bayer.” (ibid, p. 54)

Most chemical testing is conducted through for-profit contract research organizations, or CROs. This collective of companies make up a multi-billion dollar a year industry that exists solely to spit out junk science, producing studies that will say whatever corporations need them to say. One of these for-hire research companies, the Weinberg Group, is entirely up-front about the type of research they’re producing: “Our company (has) one goal in mind – creating the outcome our client desires.” They openly boast about having successfully defended Agent Orange and asbestos. They’ve twisted the science on phthalates and bisphenol-A, as well as chlorine-based compounds. In a pitch to Du Pont, they offered to defend perfluorooctanoic acid (which even the company knew was harmful) by saying they would “reshape the debate… to establish not only that PFDA is safe…but it offers real health benefits.” (Shabecoff & Shabecoff, 2010, p. 174)

This type of deceit should be criminal. We lock people up on felony charges for producing knock-off t-shirts or handbags. Yet apparently, the thriving industry of companies producing knock off science that poisons and kills children is perfectly acceptable. It tells you a little something about where America’s priorities lie, and it seems to be wherever corporate interests happen to be.

This endeavor is also quite lucrative for the scientists involved. A staff member for the NASDAQ traded consulting firm Exponent can earn more than $1 million a year – wages far higher than that typically paid in academia, which explains why so many scientists are willing to sell out. Dr. Dennis Paustenbach raked in more than $2 million consulting for companies on chromium over a period of 20 years. When asked whether he thought this posed a conflict of interest, he remarked, “Of course not. Frankly, in the world of conflict of interest in the scientific community, the amount of money is irrelevant.” Hard data would strongly disagree with his assessment. Another professional, Dr. Philip Cole, who was recruited as a paid expert witness for PG & E, was paid $600 just to watch the Erin Brockovich film. (Shabecoff & Shabecoff, 2010, pp. 178-80)

Biologist Tyrone Hayes offers a glimpse into this world of corporate research, as he describes being approached by the group Eco Risk to do work on behalf of Syngenta’s corporate predecessor, Novartis. The job: study atrazine, which at the time was up for reapproval with the EPA. When his studies turned up something, the company turned a blind eye. He was told to repeat his studies, but wasn’t given any funding to do so.

Hayes says that scientists at EcoRisk suggested he apply statistical manipulations that would make his findings disappear – a dishonest yet quite common technique used in the corporate world to slant research, which we’ll discuss further in a moment. Summing up his experience in corporate science, he says that “It’s just like if I was an artist and you told me to paint you this color, and I paint you that color and you buy the painting and I’m done with it.” (Slater, 2012, p. 47) He found this sort of pre-fabricated junk science off-putting, and so he resigned from Eco Risk and became one of the industry’s most outspoken critics.

What he describes is fairly typical. It’s common practice throughout the industry to never publish studies which show a negative result, which means that studies like his typically never see the light of day. Selectively choosing what studies to publish and promote and which ones to discard is a way of hiding harmful effects from the general public. The researcher conducting the study is typically under contract, which means he doesn’t own the right to his work and it’s up to the company funding it to decide what to do with it. (Hayes, too, had signed a clause disallowing him from publishing his studies without company approval.) The only time the truth ever gets out is if a scientist decides to go rogue and start funding his own private research, as Hayes did. “The evidence is just now trickling in of just how bad (this problem) is,” says John Scoggins of the Fred Hutchinson Cancer Research Institute in Seattle. (Ehrenberg, 2008)

When Hayes went rogue and quit, the company started a campaign to discredit his findings. They also hired a new set of scientists via EcoRisk to produce studies suggesting atrazine posed no environmental risk at EPA allowable levels. They also hired longtime Washington insider Jim Tozzi to use the Data Quality Act he himself created to question the validity of Haye’s studies and challenge any proposed regulations.

This type of slanted research is what regulatory agencies then base their actions upon. For example, when it came to the EPA’s position on the pesticide atrazine, a report by the Huffington Post Investigative Fund found that less than 20% of the papers the EPA relied on in making its decision were peer-reviewed, and at least half were done by scientists who had a financial stake in the product. (Slater, 2012)