Photo close-up of bacon sizzling in a pan.Some of you might have heard that bacon was rated as being as carcinogenic as smoking by the World Health Organization.

No. That did not happen.

And this is a good case for learning some modern critical thinking skills.

I’ll spoil the surprise by quoting them directly:

No, processed meat has been classified in the same category as causes of cancer such as tobacco smoking and asbestos (IARC Group 1, carcinogenic to humans), but this does NOT mean that they are all equally dangerous. The IARC classifications describe the strength of the scientific evidence about an agent being a cause of cancer, rather than assessing the level of risk.

In other words, all they said is that we are certain that “processed meats” (i.e. chemically treated meats) do cause cancer (in fact, just one cancer: colorectal cancer). They did not say it was all that bad a cause of it—certainly nowhere near as bad as smoking is of an assortment of other cancers (not only of the lung), which is dozens of times deadlier compared to an average consumption of processed meat—and most people are average consumers.

So…

First Rule of Critical Thinking Club Is: Always go to the original source and read what it actually says. The media should never be trusted to get a story right. Even less so some rando on twitter.

Second Rule of Critical Thinking Club Is: Never buy any alarmism about risk until you know how to compare the newly claimed risk to risks you already accept.

What do I mean by that?

What Risk Is That Again?

Everything kills you. Literally. Everything. It just needs quantity or time. And if you save yourself from one thing, something else will get you. Because we are mortal.

Heart disease, for example, has always remained the leading cause of death, always roughly in the ballpark of a fourth of all causes of death over the last forty years or so. One reason it always stays around that level is that if you stop or prevent all other causes of death (e.g. cancer; accident; war), your heart still fails you. Eventually it stops working. Thus we will probably never change the rate of death by heart disease very much. In fact, we would substantially increase it if we cure all other causes of death. For if nothing else can kill us but our hearts giving out from wear, then our hearts will kill us all. In that quasi-utopian world, 100% of people will die of heart disease. Yet that would actually be a sign of amazing medical progress!

Risk therefore has to be understood in this context: (1) some things are far riskier than others and (2) something will kill you eventually. So the goal is reasonable life extension: not foolishly reducing your life span (or adding decades of misery from chronic illnesses caused by risky behaviors) but also not foolishly extending it by failing to enjoy the life you have. For more years of a shitty life is worth less than less years of an excellent one. So when you get to choose, the choice is obvious. Adding one or two years to your life is not worth it if you must add decades of misery to obtain it; whereas losing a decade or more off your life, when you could be comparably happy avoiding that loss, is probably always foolish. In between is a wide grey area where you have to decide how much risk you want to take, and for what gain. It’s about how you want to live your life.

For example. We drive. Constantly. Even though we are highly likely to die from it. In the U.S., cars kill more people than guns (though guns are catching up). So we are comfortable with the risk of driving. Which is a death rate of about 10 per 100,000 per year. That’s one kind of risk measure (and you have to compare risks with the same measures). Another is to count the years off of your average life expectancy without the risk. For driving regularly, that’s in the vicinity of 1 to 2 years (more for heavier or riskier drivers).

Oh, and you also must consider, that that doesn’t always mean you get those years back if you don’t drive. Studies don’t always control for all possibilities (often because they can’t). But that’s important to do. Because if you don’t drive, you will do something else. Maybe, say, walk. Or ride a bicycle. Which also kill you. Though the death rate for these is way lower than for driving, part of the reason for that is because the population as a whole spends more hours in cars than walking or cycling. So increasing your hours on either, as you might have to do when you give up driving, will increase your death rate well above the national average.

So you can’t fully escape by avoiding driving. Even if you wanted to avoid the driving. And most people don’t. They are content with the risk.

But we shouldn’t use this as a rationalization to accept intolerably stupid risks. “I’ll keep smoking because we all die from something” is a phenomenally stupid thing to say. Not only because smoking shortens your life by at least ten years—which is a lot. But also because smoking will ruin even the life you have. My mother stopped smoking before she was forty. A decade later, emphysema still got her. Resulting in another ten and likely twenty years of considerable misery and physical disability. If you’ve ever seen what that does to a person, really seen it and lived with it…you’d quit cold turkey. Smoking is stupid. In precisely the way driving is not. Even though both kill you. One kills you at a rate five to ten times higher. And adds on top of that an equally greater risk of enduring decades of misery and disability as well.

Cut the smoking. Seriously.

So How Bad Is Bacon?

Well, okay. How does processed meat compare on this scale of concern? (And ordinary red meat, too, if it is ever upgraded in causal certainty.)

Third Rule of Critical Thinking Club Is: Always ask about any alarmist risk claim how the risk varies with exposure.

On average. Smoking sucks ten years off your life. It is thus five to ten times deadlier than ordinary driving. Smoking is also comparably more likely to disable you in some awful way for a decade or more before it kills you. But again. Five to ten times. Sixth grade math tells us: this is an increased risk of 400% to 900%. Think about that for a moment.

What is the increased risk from eating moderate amounts of processed meat? That’s 50 grams a day, or about 2 ounces, roughly four slices of fried bacon, every single day, for your whole life. Drum roll please… 18%. Hmm. Bacon, 18%. Smoking, 900%. Your eyes should be rolling by now. Generally any time someone tells you a risk increase that is less than 100%, you can probably ignore it. Because risks that low are rarely significant compared to the kinds of risks you happily accept already.

This does depend, though, on the base rate. For example, smoking’s +900% is the increased risk of death, period. That’s bad. But bacon’s +18% is really only the increased risk of colorectal cancer. Not “of death.” By anything. Much less the cancer. But the base rate of colorectal cancer is already really small. If it were huge—like, say, anywhere near the risk level of dying by driving—an 18% increase would be a matter of some concern. It would translate into a lot more knocked off your average expected lifespan than a year or two. But this is not the case here. Even if all colorectal cancer were fatal (it isn’t, though it’s bad enough to suffer through it, so I’ll consider that fact of little concern to your calculations), you will almost certainly never get it. No matter how much processed meat you eat (unless you eat such prodigious quantities as to be self-evidently foolish).

In fact, remember, whereas smoking has zero health benefits, even WHO admits “Eating meat has known health benefits.” As does alcohol (e.g. moderate wine consumption), yet alcohol causes far more cancer than processed meat. In fact, WHO found, 600,000 deaths per year worldwide vs. 34,000 deaths per year worldwide.

…Wait. Did you just hear a record player scratch to a halt?

You should have.

That’s right. WHO found that only 34,000 deaths each year are caused by eating processed meats. Um. Huh. Over 50 million people die each year. And they are freaking out over 34,000? Oh shit. Don’t look at the number killed by driving! It makes the top ten, at about 1 million deaths a year. Driving could be, by this measure, almost thirty times more likely to kill you than eating bacon. Translation: You should hardly give a fuck.

(Indeed, the variance could be greater. Because worldwide, eating processed meat may be more common than driving.)

So don’t worry about your bacon. Eat it in as much moderation as it already made sense to. And while you’re at it, maybe buy it from as ethical a meat producer as you can find (Niman Ranch, for example) or from states (e.g. California) or countries (e.g. Australia) that are pushing more humane husbandry laws. That’s a reasonable way to live.

That’s why WHO admits “consumption of processed meat was associated with small increases in the risk of cancer.” Unlike smoking. Oh so very much unlike smoking. And though they couldn’t prove red meat caused cancer, they did find that if it does, then its impact rate is 17% per 100 grams (so, same risk, at twice the consumption; or half the risk, at the same level of consumption). So even at worst, it’s equally trivial. They think “diets high in red meat could be responsible for 50,000 cancer deaths per year worldwide.” High consumption. And that only gets you to twenty times less lethal than driving (depending on how common heavy red meat diets are relative to driving worldwide, which I don’t know).

Shit. Why is this even news?

As WHO puts it:

These numbers contrast with about 1 million cancer deaths per year globally due to tobacco smoking, 600,000 per year due to alcohol consumption, and more than 200,000 per year due to air pollution.

This may look like smoking is as lethal as driving, but being the worldwide count, this will depend on variances in traffic safety, and on things like hours driven vs. cigs smoked. And, remember, that stat does not include the years of misery caused by debilitating diseases, which are also 400% to 900% more likely for smokers. And this is only counting cancer deaths from smoking, and thus leaves out, say, death by respiratory failure from terminal emphysema. Meanwhile, if as many people eat processed meats as are exposed to air pollution worldwide, it certainly looks like we should be a lot more worried about curbing air pollution than avoiding bacon. And that 200,000 is again only cancer deaths. Not respiratory failures, which are actually a more common mortality outcome from air pollution.

Perspective. We all need a good dose.

What This Tells Us about Critical Thinking

All three of these rules can be generalized beyond risk decisions to a level of greater abstraction in principles of reasoning:

  1. Always check the primary source of a claim. Because claims become distorted pretty quickly. Almost instantly in fact. A researcher’s own press release—sometimes even the abstract for their study—can already begin the distortion; media only make it worse; alarmist twitter randos, more so.
  2. Always make sure you are clear on what is being measured, and how it is being measured, and how that measure compares to everyday things. Because mistaking or not understanding these things is a major cause of error.
  3. Always ask how an effect varies with exposure to its cause. Because more or less of a thing can be good or bad. And outcomes change with inputs. Simply assuming the inputs never change, or can’t, is another major cause of error.

There are other aspects to this.

  • Many cognitive biases innate to our brains concern an inability to accurately evaluate risk. And you have the same lousy brain as everyone else, so you aren’t exempt.
  • People fall victim to standard fallacies of logic in their own thinking far more often than they admit (e.g. rationalizing a smoking habit). And people means you. You aren’t that special. You are just as prone to this as anyone.
  • And decisions always require a better grasp of probability reasoning. Because expected risks and gains are all measures of probability. Yet we are rarely taught how to do this properly. Even scientists suck at it (as a famous mammography question showed). And if they suck at it, you probably suck at it.

So you might want to do something about that. Likewise with testing your own logic more. And getting more familiar with how broken your brain is.

Shit. I didn’t even intend this to become an advertisement for my online Critical Thinking Class starting next week. But it did occur to me halfway through. Hey. You might be interested in that.

🙂

Discover more from Richard Carrier Blogs

Subscribe now to keep reading and get access to the full archive.

Continue reading