The Obesity Myth

I don't see why it matters if everyone who's overweight loses 2 points or if a fifth of them each lose 10 points. Instead of having one person cut their BMI from 40 to 30, we could have one person go from 40 to 38, another from 38 to 36, ..., and a fifth from 32 to 30. None of them will feel like they accomplished much, but the overall public health benefit will be the same.

The one way that unequal weight losses could be better would be if people who benefit more from losing weight (which would presumably be the people who are the most overweight) tend to lose more than other people. That seems like a plausible outcome for some policies (like junk food taxes), but with other policies (like subsidies on gym memberships) you might get the opposite.

I believe that "encouraging people to get married (or Jesus)" was said sarcastically. There were a couple times he used sarcasm that didn't really come through in the text, but could only be figured out be rereading and considering what message he was trying to convey.

A small movement of the average (27 to 25) could consist of 1/5 of the population losing 10 points, rather than everyone losing 2 points.

But which is more likely to actually happen? Or, more realistically, which is less tremendously unlikely?

Healthier, yes. Thinner, probably not.

Yeah, that's trying to have it three ways at once.

I've been preparing for this moment my entire life: a topical xkcd comic.

Also, if obesity lowers life expectancy and causes serious health problems, I would expect people who are genetically more prone to obesity to have fewer descendants. I also suspect slimmer people probably reproduce more because 1. fitter men are more sexually successful and 2. pregnancy can be complicated by obesity. Since cheap, readily available fatty foods have only been available recently, the obese-prone have never been strongly selected against until now. Basically, fast food is kicking obese people out of the gene pool.

Let's consider, though, why humans can become fat. The body stores energy in the form of fat for later conversion to sugars in the event of insufficient sugars to keep the body functioning.

You know that mammals cannot convert fat to sugar no matter what, right? (OK, there's some limited evidence that glyoxylate cycle enzymes might be present in mammals, but even if they are active, what doesn't seem terribly likely, the primary way to utilize fatty acids is definitely beta oxidation, which burns them for ATP).

To me this suggests that outright starvation is, evolutionarily speaking, a far greater threat than lack of access to particular essential comestibles.

There's an argument that pre-agricultural food had a lot higher micronutrient to calorie ratio, so micronutrients were unlikely to be lacking even on very restricted diet. Modern food is supposedly selected for high calorie content.

Historically being thin was not attractive, especially for women, as evidenced by human art that has long shown women with at least some significant fat stores as the most attractive.

As far as I can tell most art shows women of healthy weight as attractive, very rarely overweight or obese kind.

Probably nothing, although it's a worthy goal to push the elimination of weight as a status signal because even a partial success would result in a lot less suffering for billions of people.

Why would we want to eliminate a status signal that's so easy to manipulate for smart people?

In the mean time, at least LW readers can eliminate from themselves false beliefs that anything but extreme obesity, or extreme skinniness, has anything more than a marginal health effect.

Metabolic syndrome seems real enough. Even if being moderately overweight wasn't a health risk, it seems likely that someone moderately overweight is much more likely to become morbidly obese than someone with healthy weight. And far more people suffer from obesity-related diseases now than ever - so even if it's mostly the extreme cases, if anti-fat interventions can get them to just moderately overweight it's obviously worth it.

Let's consider, though, why humans can become fat. The body stores energy in the form of fat for later conversion to sugars in the event of insufficient sugars to keep the body functioning. Sugar is only part of what the body needs to keep functioning (we also need amino acids we can't synthesize except by deconstructing muscle, trace amounts of vitamins and minerals, and water), but it seems to be the most critical since our bodies are very good at storing it. To me this suggests that outright starvation is, evolutionarily speaking, a far greater threat than lack of access to particular essential comestibles. So our bodies are doing the right thing by making us fat because it protects us against the very real and serious threat of starvation.

That is the standard argument. Maybe it is true. But has it actually been tested? Is it observed that when famine strikes, the fat survive best, or that people with a tendency to fat in times of plenty survive better than the naturally thin in conditions of chronic scarcity? A lot of things with obvious explanations turn out not to be true, and there's surely enough famine in the world to study.

After all, if the standard explanation is true, then we do know how people can make themselves thinner: eat as if food were in chronically short supply. This does not tell us how to make people thinner, short of concentration camp conditions, which certainly did work --- none of the liberated prisoners were fat. But outside of such extremes, making people do anything is difficult.

See also Intermittent Fasting (eat only every other day), reputed to also extend life.

Being fat is unhealthy because it does leave you more susceptible to disease. The American health care system wastes billions of dollars each year on diseases and conditions that are a direct cause of obesity. Diabetes, heart disease, high blood pressure, and cardiovascular disease are just a few of the conditions that could be completely avoided by millions of Americans if they just ate/exercised better. It seems that easy access to high calorie junk food and absolutely no exercise (never taking the stairs) is just too much of a temptation for many people... and in evolutionary speak, is a good indicator of that person's genes.

Obesity rarely occurs in nature because obese animals are targeted quickly by predators and quickly dispatched. The fat rabbit cannot escape the lean, muscular leopard chasing after it. In the event of a massive human starvation scenario I would argue that the fat people would be killed off first (possibly by the lean/muscular humans or predators) as the lean humans would require less resources and would be more capable of gathering food for the tribe. What good are obese/severely obese people in the hunter/gatherer world?

Think about all the benefits staying fit yields. It is easier to sit down, get up, row a boat, hunt for food, have sex, run, participate in sports, attract mates (good genes), defend your tribe, etc.

There's another, related explanation, which is superstimuli. A lot of modern processed food is developed to maximize its taste impact. If there's a feedback between loop between good taste and higher consumption, widepsread and rising obesity may be the result of processed foods breaking this feedback by tasting too good. "Low-fat" may also exacerbate the problem, since fat helps signal satiety, so high-sugar, low-fat foods provide calories without curbing hunger much.

A genetic or evolutionary explanation seems unlikely because obesity is distributed non-randomly throughout society. It affects certain regions, subcultures, and classes very differently. This could be partly explained by the tendency (via affluence, culture, whatever) of certain groups to avoid eating highly processed high-calorie foods. It may be explained simply by net calorie consumption, but I'd be surprised if that itself is not correlated with the consumption of high-calorie processed foods.

It seems fairly obvious that you're right about why humans become fat -- i.e., because in our evolutionary past, it was adaptive to do so when there was a lot of food around -- but I don't see how you get from that point to "doctors are ... playing the status game and consider thin beautiful" rather than "medical research has shown that obese people are at a higher risk from certain health problems". It's perfectly possible, and seems quite probable, that gaining weight is adaptive only a) up to a certain point, which was extremely rarely reached until recently, and/or b) under the actual conditions it was adaptive to -- that is, when the weight is highly likely to come off again during a period of famine.

Nobody disputes the obvious fact that it is possible to make anyone lose weight - starvation is one classic method that has been around for a long time. But that isn't the problem. We know how to make a person thinner, but we don't know how to make people thinner, because people like to eat fatty foods, they like to be sedentary, they dislike dieting and often don't have the willpower for it. Having an effective diet is only part of a solution if people are resistant to it.

Saying calorie restriction is a way to make people thinner is like saying quitting cigarettes is a way to lower cancer rates: not true enough.

Yes: diet and exercise. That's exactly how you make people thinner. The problem -- there are competing truths here -- is that diet and exercise isn't the easy solution it seems to be. How do you implement calorie restriction and exercise? This is what Paul is talking about when he says we don't know. (But I think his choice of words suggests that the "diet and exercise" message is debunked or something.)

Re: "There aren't any health differences to speak of for people between BMIs of about 20 and 35"

Uh, reality check!

There not any doubt that health is affected over this range (imagine here the BMI of a single individual varying over this range) but the problem -- to mediate again -- is that BMI is not a reliable measure of unhealthy fat, so you can't accurately compare the health of two different people by their BMIs, or really have a good gauge of your own health from your current BMI. Paul Campos is exploiting this confusion. It would be incorrect (and intellectually dishonest) to connect this message about BMI with the message that it's just as healthy to be 30 pounds overweight.

Some discussion of the connection between health risks and BMI here and the limitations of BMI as a measure here.

I haven't read the book, so I don't know if Paul Campos actually asserts that being fat is healthy, and unrelated to diet and exercise, or if he is deliberately misleading in order to whip up interest in his book.

Yet there's at least one true message: for people who have a healthy weight but a high BMI (for their height, thanks Psychohistorian), they can consider themselves healthy because they're not overweight.

After all the serious thoughtful comments, this one just makes my day. :D

Even the typing style of this message makes it sound irrational. Were you joking?