I recently came across an article published by Dan Gardner, an environmental consultant for Tactix in Ottawa, with the provocative title „There are harmful chemicals all around us. And this is fine„.

Gardner’s basic line of reasoning is that „the dose makes the poison„, doses equal a „grain of sand in a football field“ and, ergo, everything is peachy.

Not it’s not. And we need to discuss those issues with more than just bumper-sticker depth.

But please head over to the article to check for yourself. I’ll provide some quotes below, but it might indeed be helpful to first read the original text.

The dose makes the poison! Praise Paracelsus!

“The dose makes the poison” is a very old saying credited to Paracelsus, the 16th-century proto-scientist, that expresses the core insight of toxicology.

Oh, no. Not that one again… I definitely need to write something more extensive in the near future about the actual context of Paracelsus‘ third defense, and why such mutilated out-of-context oversimplifications are not helpful for modern risk assessment. But for now, let me just discuss the issues relevant for Gardner’s article.

Of course I agree with the basic assumption, i.e. that the dose is a crucial determinant of toxicity and hence chemical risks. And I also agree that this is not always adequately considered when discussing chemical risks, especially in the news media. Detecting a chemical doesn’t necessarily imply risk.

But that does not at all imply that detecting a chemical at intuitively low concentrations (more on that below) allows the a priori conclusion that this is nothing to worry about.

Let’s acknowledge a couple of things: toxicology has advanced tremendously in the roughly 500 years since Paracelsus wrote his third defense. Chemical use and production have also changed by orders of magnitude. And people certainly expect a better health protection than in medieval times. Modern-day chemical risk assessment thus wrestles with a myriad of complex issues, such as exposures to chemical cocktails comprising dozens or even hundreds of chemicals, the realization that the timing of exposure is critical, the influence of genetic predisposition, non-monotonous dose-response curves and the discovery of epigenetic mechanisms, to name but a few.

All of that makes it fiendishly complex to determine the „safe dose“ of a chemical.

It then becomes obvious that the article misses a crucial point: for many, or even most, chemicals the „safe level“, even if it exists, is unknown or contested. Given the lack of toxicological knowledge for many chemicals, and given that chemical hazards are sometimes deliberately miscommunicated, it is indeed simple common sense to avoid exposure as much as reasonably possible. Broadly painting such efforts as „chemophobia“ strikes me not only as condescending, but also as ignorant.

Intuitive toxicology should be avoided. Agreed. But then why base the article’s arguments on it?

A central argument of the article is that „intuitive toxicology“ in which „Any detectable amount was too much“ is not helpful to assess present pollution patterns.

Agreed. But then why use purely intuitive reasoning in order to try to persuade the reader that chemical exposure is mostly harmless? The article reads:

Today, thanks to science and technology, we can detect inconceivably tiny quantities. A part per million. Or billion. Or even trillion. How little is one part per trillion? The inimitable science communicator and chemist Joe Schwarcz has a vivid illustration: Imagine a football field surrounded by six-metre-high boards and filled to the brim with sand. Buried in there somewhere is one particular grain of sand. That is one part per trillion. To understand information such as that, “intuitive toxicology” is worse than useless.

My apologies for the long quote.

So, Gardner argues that we should let our intuition decide that „a grain of sand in a football field“ cannot cause harm. This appeal to intuition falls flat simply because several modern chemicals are specifically optimized to cause biological effects at incredibly low levels, for example hormones such as the anti-contraceptive ethinylestradiol or insecticides such as chlorpyrifos. Even pollutants such as TCDD-like dioxins or natural toxins such as botulinus toxin are equally potent. The latter has an acute toxicity (LD50) of just 1ng/kg bodyweight. Which is exactly the 1 grain of sand in Gardner’s football field.

Of course, I am not saying that every pollutant is as toxic as botox. But I am certainly saying that the toxicity of chemicals spans orders of magnitude and oversimplistic generalizations that appeal to intuitive feelings are not helpful.

For some chemicals the only safe dose is indeed zero

Gardner takes the argument that low doses are harmless one step further. He writes

To the chemophobe, the danger posed by a chemical isn’t conditional on amount. It’s inherent in the substance. Poison is poison. The only safe exposure is zero. But if that were true, we’d all be dead.

Once more, Gardner underestimates the complexities of chemical exposure and toxicological properties which simply do not allow for such broad-sweeping generalizations.

For many chemicals one could indeed postulate a „safe exposure“. But the fact is that regulatory agencies regulate a whole group of chemicals, genotoxic carcinogens such as benzo[a]pyrene, indeed under the assumption that no safe level exists. Risk assessment then pivots to the question of risk acceptance. Which is a societal discussion, driven by personal value judgements for which there is no absolute „right“ or „wrong“.“

„The dual-edged nature of all substances“

Again, this oversimplification, in order to force the issue towards a conclusion that was drawn long before the article was written.

Sure, chemicals such as potassium chloride, the example used in the article, are essential. But exposure to any amount of e.g. lead has absolutely no health benefits. Actually, the European Food Safety Authority concluded already in 2010 that “there is no evidence for a threshold for critical lead-induced effects.” In other words: indeed, the only safe lead exposure level is zero.

To summarize, I would argue that, instead of using flawed broad-sweeping generalizations, we need to acknowledge the complexity of the issue. Chemophilia is as problematic as chemophobia. We need to discuss chemical risks not in terms of absolutes, but in terms of slippery slopes.