Thirteen hypotheses of the obesity epidemic

16 August 2023

There is an obesity epidemic in western countries, apparently. No, I’m not providing citations.

Here is a very rough, incomplete list of theories about the cause of the obesity epidemic:

The CICO (calories in, calories out) model. This is the standard model of weight gain and loss. Note, however, that it isn’t in itself a theory about the cause of the obesity epidemic, since presumably the relationship between calories consumed, stored and expended has been true since the dawn of time, whereas the obesity epidemic has only appeared in the last few decades. More fundamentally, it mistakes a mathematical identity for a causal model: while it must be true that people who gain weight are consuming more calories than they expend, that tells us nothing about why they are doing so.
Set point theory. This is in some sense a contradiction of the CICO model. But again, while this might make sense as a driver of weight, it doesn’t in itself explain why set points have been going up.
Food availability. That food is now more abundant than it was previously and we eat more because it’s easier to do so.
Food palatability. That the twin engines of technology and capitalism have made food that’s more appealing than it was previously.
Food variety. Separate from the above, the variety of foods available has made people eat more. You can feel full of your main course, but still find that you have room for dessert.
Meal frequency. We eat more often than we did previously and don’t have sufficient periods of fasting, which affects metabolism. This is part of the thinking behind intermittent fasting.
The insulin hypothesis. That we store fat in response to elevated insulin as a result of eating high quantities of carbohydrates. This was put forward comprehensively by Gary Taubes in Good Calories, Bad Calories.
Separately from above, that insulin metabolism is particularly affected by high quantities of sugars, and we eat more sugary food now than we used to.
Separately from above, that insulin metabolism is particularly affected by fructose, e.g. in high-fructose corn syrup. Robert Lustig has made this argument.
That seed oils and particularly linoleic acid have a damaging effect on metabolism.
That there are pollutants that are causing people to gain weight. Candidates include lithium (as suggested by Slime Mold Time Mold) and microplastics.
That there is some dietary deficiency that people are suffering from. Candidates include potassium, also suggested by Slime Mold Time Mold.
That time spent standing versus time spent being sedentary has some effect independent of calories consumed by the activity alone. So even though we encourage people to burn calories through exercise, it doesn’t help if they spend the rest of their time sitting at their desks or in their cars.

I’m sure I could list a few more with a bit of work. I recall that Seth Roberts had a theory that was something to do with the association between flavour and calories, though I’m not sure I ever understood it properly.

Gary Taubes had a hell of a time when he first published Good Calories, Bad Calories with people saying that he was denying the laws of thermodynamics in rejecting simple interpretations of the CICO model. I feel like the situation has improved somewhat since, at least among people who think give a lot of thought to these things, but how much of that has extended to the general public I don’t know.

One of the theories that I personally think is promising but under-studied is the possibility that time spent sitting versus standing is particularly damaging. It’s quite easy to find studies on calories expended through sitting versus standing, but this is entirely useless and stupid if there is an additional effect on metabolism separately from the calories expended through the activity. You can even find people who you’d think would know better approvingly quoting maths of calories expended, though I appreciate that they’d welcome actually running the experiment.

Considering everything said above, it seems to me that the cause of the obesity epidemic is one of the major open scientific questions that we have to wrestle with. While I might attach different credences to each of these hypotheses, I’m not sure that I could rule any of them out entirely.

Usually when I find myself in a conversation about weight, the other person has an underlying assumption that losing weight is simply a matter of eating less or exercising more. This is intensely frustrating to me. While I wouldn’t want to deny that you might be able to move the needle a bit with that approach, I would like to see much more appreciation that this might not be the relevant lever to pull.

Occasionally they’ve picked up some other particular non-CICO theory and are now convinced that this is the answer. This is a slight improvement since they’ve at least had the experience of changing their mind once. But even then the argument is usually simply in favour of their pet theory rather than weighing up the evidence for multiple theories.

I guess that a standard I’d like to see is something like this: if you hold a strong opinion about obesity but you can’t list at least five competing theories off the top of your head then maybe you should have a bit more humility.