Stunning account of the length epidemiologists go when your study goes against what they believe to be true

(Bob M) #1

This is long, but stunningly revealing:

Obesity wars

Basically, the author and her colleagues had the temerity to publish findings on the effect of being overweight (better than being normal weight) or obese (not as bad as previously thought) that were scientifically sound. Harvard was not happy.

The extent to which this was attacked is revealing.

(A fool and his bacon are soon parted) #2

Walter Willett’s tactics bear a striking resemblance to those Ancel Keys used against John Yudkin. Since Willett and Keys were friends, this is perhaps not particularly surprising.

(Michael - When reality fails to meet expectations, the problem is not reality.) #3

‘Scientists’ are humans. All too so sometimes. The objective search for truth often takes a back seat to the subjective justification of your own pet notion. How many nutritional ‘scientists’ still think keto is nonsense or dangerous and eating a few hundred grams of carbs per day the mainline to health and longevity?

(Gregory - You can teach an old dog new tricks.) #4

I’m reminded of a study, I’ll try to find a link, where all cause mortality in elderly ( 65 and older ) was lower for the group that had higher LDLc…

Part of the conclusion was that they could not account for this, and they would have to dig deeper…

(A fool and his bacon are soon parted) #5

I believe this was true of both the Nurses’ Study and the Women’s Health Initiative. And possibly of the Framingham Study as well, though no doubt the Harvard School of Public Heath would deny that.

(Joey) #6

Thanks for sharing this link.

Malcom Kendrick covered this finding that the “overweight” group (BMI-defined) has been found to have the lowest mortality (as compared to “underweight,” “normal,” and “obese”) in his wonderful book “Doctored Data.” Well worth a read if you haven’t already gotten to it.

(Butter Withaspoon) #7

I’m still shaking my head. Well worth the read, thank you!

(Bob M) #8

@SomeGuy I first saw the idea that “overweight” might be “better” than “normal” weight, at least as far as death rates go, on Malcolm Kendrick’s blog. I also have all of his books, except for the last one.

As for the article, I was shocked at the extent of what they did to attack this. Particularly because who really cares whether there’s a study saying “overweight” might be better than “normal weight”? What does it matter? To me, nothing. To them, apparently a lot.

(Joey) #9

With a BMI of just over 20, it now makes me wish I were overweight :roll_eyes:

(Todd Allen) #10

It’s mind boggling to me that people fight over observational studies. They only produce statistical correlations but can be completely meaningless with respect to causation. In this case it may very well be true that those who are overweight live longer on average than those of normal weight. But the overweight includes heavily muscled people of good fitness and normal weight includes people who are fat but not heavy due to things such as smoking, sarcopenia, cachexia and countless other conditions. I see similar issues with epidemiology being used to assert things like plant foods are healthier or cholesterol is the enemy.

These studies should be the basis for hypothesizing not for jumping to conclusions.


Having lived in a retirement community for over 13 years, I do believe the elderly and infirm are better off being slightly overweight.

Anecdotally, the normal weight, and especially the underweight, residents seem to go downhill a lot faster when they have what should be a short-term medical problem. They have no reserves to draw on and aren’t interested in eating or drinking, so they just get dehydrated and more infirm. It becomes a vicious cycle.

(Bob M) #12

@SomeGuy These are only observational, so it’s unclear what’s going on anyway. It’s like when I read studies where people with high LDL or total cholesterol live longer, and lower not as long…and I’ve had lower cholesterol my entire life. What do I do with that info?

@brownfat I agree with everything you say.

@OgreZed On the other hand, I can see where some amount of extra fat could be helpful in certain situations. And being very thin is correlated with typically the worst outcomes. Chicken or egg though? Maybe you got super thin because you were sick?

And I listened to a podcast with Ben Bikman where he said he was in Singapore and he realized that people of Asian descent who gain weight are much worse off physically than people of European descent. In other words, people of European descent can gain weight, and this can tend to protect them (myself included) from the effects of insulin. Whereas people of Asian descent who gain a lot less are a lot sicker.

Dang genetics!

(Michael - When reality fails to meet expectations, the problem is not reality.) #13

Re epidemological studies and correlations, I posted this a couple weeks ago in Breakfast with Bubba: