Toxic Kidney Beans?

We are, perhaps, uniquely among the earth’s creatures, the worrying animal. We worry away our lives, fearing the future, discontent with the present, unable to take in the idea of dying, unable to sit still.
--Lewis Thomas, The Medusa and the Snail

***********************

Readers, there's something new to worry about. If you want to, that is.

Toxic beans.

I learned about toxic beans thanks to a comment from Mike Vrobel at Dad Cooks Dinner on my Easy White Bean Stew recipe. Mike warned readers to watch out for the toxin phytohemagglutinin, which can be found in kidney beans and certain related white beans (cannellini beans for example, which are cousins of the kidney bean). Mike helpfully directed readers to a page at the US Food and Drug Administration's website discussing details of this risk (link here).

There's an easy fix, fortunately: just boil the beans for ten minutes or more. This breaks down the toxin.

But the possible problem here lies with kidney beans cooked in a crockpot or slow cooker while using the low setting. It's theoretically possible that someone's crockpot, on a low setting, might not bring the beans quite to the boiling point, which wouldn't break down the phytohemagglutinin toxin. You might then be eating toxic beans.

Clear so far?

Now, the easiest way to eliminate any risk of food-borne illness is to cook your food properly and thoroughly. If you're concerned about your crockpot's low setting while making my Easy Bean Soup, just look in on it after a few hours. Is the soup bubbling? If it is, you're in the clear.

However, I want to go further. I want to tackle this "toxic beans" issue on its own merits. Is this really a risk that we should worry about? Because, to be honest, it smells like yet another example of we here at Casual Kitchen call worry porn.

Before we analyze the true risk of toxic beans, I want to make two points. First, this is in no way a criticism of Mike. I love Mike's work. I always enjoy his insightful writing, and I've found significant value from his book Rotisserie Grilling, which is literally the reference guide on the subject. He's a credit to the food blog world, and I'm grateful he brought up this issue--not least because it gave me the idea for today's post.

Second, in no way whatsoever am I saying that the risk of toxic beans isn't real. It is real, as we will soon see. Just rare.

Except that, well, human spontaneous combustion is real too. And also rare. So rare, in fact, that we can safely ignore it as a risk.

So really, the question is: how rare is death or illness due to improperly cooked kidney beans? Is it rare enough that we can ignore it?

Remember, we are surrounded by worries, and surrounded by a system of media designed specifically to grab our attention with worrisome things. As a result--and despite the fact that life in the modern era has never been safer--we are worrying more than ever.

So, it's up to us to choose: will we submit to every seemingly convincing fear tossed at us? (Protip: they will all sound convincing.) Or will we consider the fear rationally and disregard it if it lacks merit?

I'll give you the punchline first. With toxic beans we quickly arrive at an obvious conclusion: disregard for lack of merit.

What's sad and disturbing about this particular worry, however, is how much exaggerated and even incorrect information I found sitting, right there, on the FDA's phytohemagglutinin toxin factsheet.*

We'll get to the errors in a minute. But first, let's look at the prevalence and probabilities behind "kidney bean syndrome." The FDA cites one primary paper citing specific cases of this toxicity from the UK (see bullet point #6 on the factsheet). This study cites, for example, seven examples of "outbreaks" from 1976-1979. If you hunt down the study (a quick Google search uncovers it: link here, see pages 236-237), the cases over this four year period involve a total of just 43 people in the UK, a country where kidney beans are rather popular and which (at the time) had a total population of 55 million. Of these 43 cases, everyone recovered rapidly, and no one died.

So, let's look at these four peak years of 1976-1979, when by far the worst burst of cases occurred, and let's calculate the worst-case odds: a per-annum risk rate of 1.95 x 10 -7, or, roughly, one chance in five million.** Infinitessimal. If you were to consider the longer period of 1976-1980 (the UK study lists an addendum of another small outbreak in 1980), or if you were to consider a related paper that found and studied 50 suspected cases of toxic beans (but confirmed only nine cases) between the fourteen year period of 1976-1989, the probabilities become laughably low.

Sure, this toxin may exist. But it cannot kill you. And unless you were alive in the 1970s in the UK and lottery-winner unlucky, it can't even make you sick. It seems more probable that "kidney bean syndrome" is rooted in some other factor--perhaps some idiosyncrasy with how beans were processed in that era in the UK, or perhaps a combination of processing techniques combined with improper cooking. Most likely the real root cause is simply unknowable.

What is known, however, is this syndrome's freakish level of infrequency. Despite this, the FDA somehow manages to claim on its factsheet that "this syndrome has occurred in the United Kingdom with some regularity." Worse, the FDA page appears to make a factual error in saying "in the seven outbreaks mentioned above, the attack rate was 100%." The 1980 study is not so definitive. Finally, in the primary 1980 study, all but one of the "outbreaks" came from eating raw kidney beans.

Okay. So don't eat raw kidney beans. You never needed to be convinced of that in the first place.

This next part I want to be sure I phrase carefully. On the factsheet under #10 ("Selected Outbreaks") the FDA links to the CDC's website, which provides a search of the term phytohemagglutinin. Here, the CDC website offers two entries, except that neither involve any instance of kidney bean toxicity at all.

The first of the two entries has nothing even to do with beans--it's a rodent study of isobutyl nitrite--something completely unrelated. The second entry, "Outbreaks of Gastrointestinal Illness of Unknown Etiology Associated with Eating Burritos: United States, October 1997-October 1998" essentially says "a very small number of people got sick eating burritos, and we don't really know why, but here are some possible reasons."

Except that the document itself says kidney bean toxicity couldn't have been the reason for these illnesses. Why? Because the questionable burritos contained... pinto beans, which don't contain the toxin in the first place.

I'll state it as clearly as I can: It's one thing for the FDA to use the phrase "with some regularity" in describing a food illness as freakishly rare as kidney bean poisoning. It's another thing entirely to link to "selected outbreaks" that are not actual outbreaks.

Forget worrying about toxic beans, I'm worried about the FDA's ability to inform us appropriately about health risks.

Look, I'm no one special. I'm neither a doctor nor a scientist. And yet I was able to easily identify clear exaggerations and factual inaccuracies on a public FDA factsheet. I expect this kind of shoddy work from, say, The Food Babe's website, not the FDA. This is our own government's food and drug regulatory body--the very people who are supposed to be in charge!

What's depressing about this--to me, at least--is how much time and due diligence it takes to follow up on links cited as evidence of a given health risk, only to find that the links aren't really evidence at all. Will the average reader search out all of the linked and unlinked sources behind the FDA's worry page to see if those links support what the FDA says they support? Do you have the time to do this? No one follows all the links and reads all the studies.***

I expect more from our FDA than a scaremongering factsheet with links to outbreaks that aren't. You should too.

So. Are you looking for more and more things to worry about? Do you think it's a coincidence that you keep finding them?

Readers, what do you think?


Related Posts:
The Cure for Worry Porn
Could Toxins Be Good For You?
Four Incredibly Useful Books on Fallacy and Cognitive Bias
Understand the True Nature of Consumer Retailing
Organic Food, Chemicals, and Worrying About All the Wrong Things
When It Comes To Banning Soda, Marion Nestle Fights Dirty

Footnotes:
* Perspicacious readers will note that the FDA factsheet says two things: "The content on this page is provided for reference purposes only. This content has not been altered or updated since it was archived." and further, that there is a new version of this reference guide (the "Bad Bug Book") available. Sadly, the new text (which you can find here, see page 254 in the PDF for the text on phytohemagglutinin) is substantially the same as the old text, and most importantly, the links and errors I criticize--including the CDC link to "outbreaks that aren't"--remain extant in the document. The errors survived into the new edition and were not checked or corrected.

** I'm making an implicit and extremely conservative assumption of one serving of possibly toxic beans per person per year on average in the UK. We all know the English love their beans.

*** Readers: this brings us a highly effective technique not only to instill fear and worry, but to win any online debate: just link to tons of stuff and pretend it supports your case! The sheer weight of all those links means you win. Who cares that they don’t say what you say they say? No one will check anyway.


How can I support Casual Kitchen?
Easy. Do all your shopping at Amazon.com via the links on this site! You can also link to me or subscribe to my RSS feed. Finally, consider sharing this article, or any other article you particularly enjoyed here, to Facebook, Twitter (follow me @danielckoontz!) or to bookmarking sites like reddit, digg or stumbleupon. I'm deeply grateful to my readers for their ongoing support.