Still No Obesity Paradox

In some studies of large numbers of people, researchers have found that people with the lowest death rates had a body mass index (BMI) solidly in the overweight range. This finding has led to the so-called “obesity paradox”, meaning that while we generally think that being overweight is unhealthy, being a little overweight might be healthier than being of normal weight. A new study shows that the obesity paradox can’t possibly be true.

A paradox means your theory is wrong

When researchers declare something a paradox, it means that some piece of data doesn’t agree with their theories. For example, the French paradox refers to the fact that the French eat a diet high in saturated fat, yet have low rates of cardiovascular disease. The French paradox is easily solved if you discard the idea that saturated fat causes heart disease.

A paradox is nature’s way of telling you that you’re wrong.

The obesity paradox can be characterized as follows:

… numerous studies have documented an obesity paradox in which overweight and obese people with established CV disease, including HTN, HF, CHD, and peripheral arterial disease, have a better prognosis compared with nonoverweight/nonobese patients.

The paradox can be resolved either by concluding that being overweight isn’t really unhealthy, or that the data showing that is wrong.

What could explain the obesity paradox

The main confounding factor in the obesity paradox is likely to be reverse causation, meaning that in this case, poor health causes weight loss. If it did, then the category of normal BMI would include people who were in worse health than people who weighed more.

And we do know that poor health can cause weight loss. Smokers also weigh less than non-smokers, and are in worse health.

New study refutes the obesity paradox

The study looked at over 296,000 people in the UK Biobank database. Only people non-smokers who were healthy at baseline were included.

Result: no obesity paradox. Risk of cardiovascular disease rose starting at a BMI of about 22. Other studies have found that a BMI of about 22 or even lower is the healthiest, so this latest study adds evidence.

Of interest, being underweight had a higher risk than being overweight/obese. That’s likely due to low muscle mass.

The obesity paradox is due to confounding due to smokers and those with ill health being included in the category of normal BMI.

Maximum BMI

Another way to investigate the obesity paradox and to clarify the dangers of obeisty is through the use of lifetime maximum BMI, rather than BMI at time of enrollment.

Using lifetime maximum BMI, not only is there is no obesity paradox, but risks of overweight/obesity are much higher.

Using BMI at survey, an estimated 5.42% of deaths were attributable to the combination of overweight and obesity, whereas using maximum BMI, the attributable risk was substantially greater, at 32.58%.

Being even a little overweight is unhealthy

Well-controlled studies, which exclude smokers and other unhealthy people at baseline, show that the health risks of being overweight rise monotonically with increasing weight. There is no obesity paradox, and being overweight is bad for health. Full stop.

Other studies using maximum lifetime BMI substantially increase our estimates of the detrimental effects of overweight/obesity on health.

For those few people who have a BMI >25 due to a high muscle mass, rest assured that the aspect of high BMI that causes problems is not muscle, but fat, especially visceral fat.

Stay lean, my friends.

PS: For how it decrease body fat and increase muscle mass, see my book, Muscle Up.

PPS: Check out my Supplements Buying Guide for Men.


Leave a Comment:

Aboutlifting says March 19, 2018

Paradox, because damn idiots are seeing that ice cream sales increase with the rate of people drowning….and so they blamed ice cream for drowning people – there you go; mainstream logic goes like that when it comes to saturated fats, bmi, and paradoxes…it has already made up its mind soo it will always be a “paradox”.

Bill says March 20, 2018

My father was a life long smoker and always thin. Under pressure from my mum in his 50’s he stopped smoking for a while and put on about 9 kg with no change is his diet ! I thought it was bizarre at the time. However when he started smoking again the added weight dropped back off.

So smoking was definitely keeping him thinner…But it killed him with throat cancer…

Thanks for the reminder PD that sometimes the paradoxes found by science just reflect very poorly thought through research.

Ole says March 22, 2018

James P Watson has some interesting points regarding the obesity epidemic. Bascially he is claiming that we should all supplement with vitamin C and avoid HFCS like the plaque:

Steve Parker, M.D. says March 26, 2018

It’s unfortunate that the Biobank study didn’t look at all-cause mortality instead of just cardiovascular disease and death. This leaves open the possibility that being overweight reduces risk of death from cancer, infections, chronic respiratory disease, etc. (I don’t think that’s the case, just playing devil’s advocate.). The Biobank probably has the data. Perhaps we’ll see it in a future report.

    peter connor says March 31, 2018

    Yes, that is rather a large omission…One theory behind the paradox was that slightly overweight people are just healthier in general…This study does not debunk that…

      P. D. Mangan says April 1, 2018

      It would be extremely difficult if not impossible for a group to have more cardiovascular disease than another group, and yet be healthier than the other group with a lower death rate. The study shows no protective effect of obesity against cardiovascular disease and that in fact obesity increases risk of CVD. “Our findings strongly reaffirm that being overweight heightens the risk of CVD and that individual should seek to keep their weights as close to recommended levels to lessen their risks of CVD.” The study on maximum lifetime BMI shows that the risks of obesity to all-cause mortality have been greatly underestimated.

Add Your Reply