How do experts determine the safe level of a chemical?

“The main rule in toxicology is that ‘the dose makes the poison’. At some level, every chemical becomes toxic, but there are safe levels below that,” wrote Bruce Ames, who is the creator of the Ames Test which determines if a chemical is mutagenic.

Welcome to California, home of chemophobia and flawed risk assessment. Photo by the author.

A Prop 65 sign in a Starbuck’s Coffee outlet. Photo by the author.

Ames says that in the 1970s the prevailing thinking was that “we should assume that even low doses might cause cancer, even though we lacked the methods for measuring carcinogenic effects at low levels.” The assumption has never left, one need only to look at the ever-present Proposition 65 signs or listen to Vani Hari (aka the Food Babe).

At the time experts also assumed that:

  1.  only a small proportion of chemicals would have carcinogenic potential
  2. testing at a high dose would not produce a carcinogenic effect unique to the high dose; and
  3. carcinogens were likely to be synthetic industrial chemicals.
    It is time to take account of information indicating that all three assumptions are wrong. – Bruce Ames, 2005.  (my emphasis)

Ames points out that our test for carcinogenicity of feeding animals near-fatal doses of the chemical is flawed because, “High doses can cause chronic wounding of tissues, cell death, and consequent chronic cell division of neighboring cells, which would otherwise not divide.”

How should a “safe” level be arrived at?

The basic steps to arriving at a safe level are:
  1. Determine a Point of Departure:

    This means to review the scientific data available on the toxicity of a compound and select the most sensitive endpoint. So if a chemical causes liver toxicity at a concentration of 1 mg/kg, kidney toxicity at 50 mg/kg and stomach ulcers at 0.1 mg/kg – the 0.1 mg/kg would be selected as the point of departure because if you pick a concentration that prevents stomach ulcers, you will by design also protect against the liver and kidney toxicity (because you need higher concentrations of the chemical to cause those). Furthermore, typically you are looking to pick a NOAEL (No Observable Adverse Effect Level) as a Point of Departure (POD), as this is the highest concentration of a “substance at which there are no biologically significant increases in frequency or severity of any effects in the exposed humans or animals.” (International Council on Harmonisation, 2011)

2. Determine how many modifying factors or uncertainty factors you should use.

The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) appendix 3 gives examples of the modifying factors to use, depending on what kind of study was conducted to determine the POD. Modifying (or uncertainty) factors provide a cushion to human exposure based on factors like which animal was used for the study, the duration of the study and whether the POD is a “No Observable Adverse Effect Level” (NOAEL) or LOAEL.

The “safe” level is really a concentration that would be highly unlikely to cause an adverse effect in even the most sensitive individuals. Using the modifying factors (in step 2 of appendix 3), this concentration results in a very conservative value. These “safe” levels are referred to as PDE (Permissible Daily Exposure), ADI (Acceptable Daily Intake), RfD (Reference Dose) and other things depending on the agency that is generating them, but they all mean the same thing: the level that would not be expected to produce an adverse effect. These values are expressed as either mg/day (where an adult body weight of between 50 and 70 kg is used as a “typical” body weight) or expressed as mg/kg body weight/day.

That’s it. The equations used, and the modifying factors suggested also differ slightly between agencies, but the general concept remains the same.
So when a safe level is determined by toxicologists using best available science, and regulators arbitrarily increase the safety factors, Schnell correctly notes, “the general public commonly misinterprets those bureaucratically generated ‘safe’ levels of exposure as legitimately established thresholds of effect…”
As Frank Schnell, who is a Board Certified PhD in Toxicology, explained, “If you’re standing near the rim of the Grand Canyon admiring the view, you’re probably safe. Nevertheless, as improbable as it is, it’s not entirely impossible that a very strong gust of wind might blow you over the edge. To make sure that you were safe, even under very windy conditions, you could step back ten paces or so–that’s what regulators call a ‘safety factor.’ But, to imagine that stepping back 100 paces, or even a mile, would make you even more safe under implausible conditions (a tornado?) would be not only misguided, but counterproductive, as well, because then you couldn’t see the Grand Canyon, at all.”


Post to Twitter

For Mice and Men, Dose Doth Make the Poison


Image by ky_olsen via Flickr

My latest Green Chain column in today’s Lake County Record-Bee:

Every day, I make my wife and myself a cup of coffee. Should I be arrested for spousal abuse? I am serving her a phenol-laced liquid, containing 826 volatile chemical substances, 16 of which are known by the state of California to cause cancer. One cup of this hot and astoundingly delicious pick-me-up contains at least 10 milligrams of known carcinogens including: caffeic acid, catechol, furfural, hydroquinone and hydrogen peroxide.[1]

In one cup, my wife and I take in more carcinogens than we would from one year’s worth of pesticide residue on fruits and vegetables. [2]

Let’s be clear: we are talking about food from plants, not just coffee; you can find naturally occurring carcinogenic chemicals in all kinds of food. Honey contains benzyl acetate. Orange juice and black peppers harbor d-limonene. Brussels sprouts, cabbages, cauliflower, collard greens, and horseradishes contain allyl isothiocyanate. And neochlorogenic acid lurks in apples, apricots, broccoli, Brussels sprouts, cabbage, cherries, coffee, kale, peaches, and pears. These are but a few; the list goes on. Whether the plant was grown without any synthetic pesticides or fertilizers is not the issue.

Just as some plants grow spines to hinder grazing, plants produce their own chemical pesticides, to combat predators and competitors. No human put them there. These natural pesticides help the plant ward off insects and animals and even other plants. That is why you will find chemicals such as allyl isothiocyanate and/or neochlorogenic acid in apples, apricots, broccoli, Brussels sprouts, cabbages, cherries, coffee, collard greens, horseradishes, kale, peaches, and pears. The plants themselves developed the chemicals.

Researchers apply the Ames test to determine if a chemical has mutagenic (potentially cancer causing) properties. Developed in the 1970s, the Ames test doses bacteria, which reproduce rapidly, with the chemical being tested to see if mutations result. At that time scientists assumed only a small number of substances would cause cancer. Instead about half of the chemicals tested, whether man-made or natural, turn up positive as being rodent carcinogens. [3] So, Dr. Ames (the man who developed the cancer tests) notes we need to “rethink what we’re doing with animal cancer testing.”

“We’re eating natural pesticides,” Dr. Ames points out, “And we eat roughly 1,500 milligrams of them per day. We eat 0.09 milligrams of synthetic pesticide residues.” [4] In other words, each day we eat over 16,600 times more natural pesticide than synthetic.

Exposure to pesticides isn’t the same as toxicity because the toxicity of a substance depends on the amount. Even that chemical which our life needs, dihydrogen oxide (H2O, water), can be poisonous if you drink too much of it. As Paracelsus, the so-called father of toxicology, noted, “All things are poison, and nothing is without poison; only the dose permits something not to be poisonous.” [5] Or, as it’s paraphrased, “Dose makes the poison.”

About a month ago in the original Peet’s Coffeehouse in Berkeley, I stood behind a woman quizzing the barista if Peet’s used chemicals to produce its decaffeinated coffee. (Never mind that the Swiss Water Process uses water, a chemical composed of two hydrogen atoms bonded to one oxygen atom.) The barista assured her the levels of the chemicals used were too low to be of concern (“Dose makes the poison”). I pointed out that coffee already has 16 chemicals known to be carcinogenic; why worry about the minuscule amount of synthetic ones. She frowned at me. I think her next purchase was to be a chemical-free chemistry set for her grandson. (You think I made that up? “Chemistry 60” with its “60 fun activities with no chemicals” costs $21.88 on [6]. Don’t the makers know that water is…oh never mind.)

The moral of this story is eating fruits and vegetables that have many of these chemicals is much healthier for you than avoiding them. The jury remains deadlocked on the coffee.

[1] Ames, Bruce N., M Profet, AND Lois Swirsky Gold, Proceedings of the National Academy of Sciences, Vol. 87, pp. 7777-7781, October 1990, Medical Sciences, “Dietary pesticides (99.99% all natural)

[2] Dr. Bruce Ames, Reason Magazine, Of Mice and Men (

[3] Ames writes in, “The main rule in toxicology is that ‘the dose makes the poison‘. At some level, every chemical becomes toxic, but there are safe levels below that.

“In contrast to that rule, a scientific consensus evolved in the 1970s that we should treat carcinogens differently, that we should assume that even low doses might cause cancer, even though we lacked the methods for measuring carcinogenic effects at low levels. In large part, this assumption was based on the idea that mutagens – chemicals that cause changes in DNA – are carcinogens and that the risk of mutations was directly related to the number of mutagens introduced into a cell.

It was also assumed that:

1. only a small proportion of chemicals would have carcinogenic potential;

2. testing at a high dose would not produce a carcinogenic effect unique to the high dose; and

3. carcinogens were likely to be synthetic industrial chemicals.

It is time to take account of information indicating that all three assumptions are wrong.”

[4] Ibid



Post to Twitter