CDC Researchers Find Lower Mortality Rates Among Overweight People Except, their using BMI to do their calculations. BMI is utterly flawed, and tends to classify a large number of very athletic people as 'Overweight'. For an example, lets take MMA fighter, Junior Dos Santos, 6'4", 239 lbs. His BMI is 29.1, he's considered, 'Overweight', hell, he's almost classified as Obese!
I'm very excited to learn that my high BMI might mean I'm not a fat slob but actually a professional athlete.
The study doesn't say people with a higher BMI are more healthy, it says they tend not to die early. It does not try to say the cause of the findings, just that they found this data after doing a study. That the article you linked wants to say more than what the study says does not make the study moronic, it makes the writer of the article moronic. I'd read elsewhere that the cause was probably that overweight people tend to make more trips to the doctor and thus diseases are caught earlier. Again, that doesn't mean that overweight people are healthier. Why can't journalists ever report on studies correctly?
But doing a study on BMI, which is already flawed, is like taking historical fiction, and then writing a history report on it.
Also healthier people are more likely to engage in outdoor/more dangerous activities, and so on. BMI alone is indeed a flawed metric because "overweight" is clearly meant to imply a negative state. Beyond that, into obese and especially morbidly obese, it regains its worth.
No, it's not. You're drastically overestimating the flaws of BMI (which are more correctly described as limitations). You're also assuming the authors of the original study aren't familiar with the limitations of BMI. argh wait no what am I doing eject eject eject.
I don't really agree that basing a study on BMI is worthless. I mean, why would probably the most famous peer reviewed medical journal publish a study on what you relate to historical fiction? I'm not saying JAMA can't get stuff wrong, but I think you're overstating your case, as jeffd said. Fuck the article. Let's go to the source!
Citing an MMA fighter to illustrate why the whole system is akin to historical fiction is pretty tenuous. A 6'-4", 240 pound dude is an outlier in a whole bunch of ways. Sure there are going to be exceptions that don't fit the categories. That 1% (or whatever small number) of people are incorrectly categorized doesn't necessarily make the categories "flawed." Without looking at any numbers, I would guess that that "large number of very athletic people" being designated as overweight isn't actually that large, and in a properly-designed study with a large enough sample size (2.88 million, here), they don't carry much....weight.
There are other possible reasons why having a slightly high BMI could reduce mortality. Once you get old, being very skinny is not all that healthy. Under some circumstances, keeping the weight on during illness becomes a high priority; in that case, fat stores are your friend. I'm not saying that the brain circuits that drive us towards obesity are well-calibrated at all (they aren't), but that a lot of the stigma of being overweight is based on social and economic factors, not the health effects. But I imagine I'm preaching to the choir here. I think most posters here get that there's a difference between overweight as judged by BMI, overweight as judged by local social norms, and overweight to the point that poses a genuine danger. Edit: wait this is the sanctum Bluh bluh skinny bird people gonna die young bluh bluh bluh
It helps that to be considered underweight by BMI, you are incredibly underweight., a guy who is 5' 11 and weighs 133 lbs is considered 'Normal' by BMI standards.
To make my point better, the problem is that the BMI categories don't really correlate well with public perception of what the categories are. Underweight is not Skinny. What folks would consider Skinny is actually in the Normal band. (5' 11 and 140 is also skinny, and I'm still told I'm skinny at 155...) And the overlap between what society considers Athletic and Overweight is also annoying, and why mortality rates for Overweight are a bit jumbled. It's not just professional athletes, if I can walk into my local gym, and by BMI standards the majority of people there are considered 'Overweight regardless of athleticism, that's going to muck with the results. So about the only thing that BMI study can tell you is that being its version of Underweight or Obese is really bad, which is a bit of a no-brainer.
Oh, don't think I am down with the BMI or anything. I'm aware of its issues. ;) I mostly just went to look up how "normal" it considered that, as when I was 5'11" and 135, I ... did not look or feel healthy by a longshot. I was (irrationally) relieved to see it was just barely in the normal range. <3
A few points regarding the article: 1. This is a meta-analysis, which is basically a mathematically rigorous review of previous studies. One advantage of a meta-analysis is that it allows you to combine small studies into larger populations, in this case ~ 2.9 million. A disadvantage is that you can only look at measurements that other people have already made. Since past medical research generally used BMI to quantify obesity, this meta-analysis is forced to do the same. 2. You might think it's a "no-brainer" that morbid obesity is associated with increased mortality. But does morbid obesity increase mortality by 500%? 50%? 5%? That question is hardly a "no-brainer". In fact, it's quite difficult to answer, but it's critical to any cost-benefit analysis. And that's the point of this study. 3. There are plenty of possible explanations for why mildly increased BMI was associated with reduced mortality, which have been discussed already. Another possibility, of course, is that mildly increased BMI actually is good for you. Don't be so quick to wave away the possibility, particularly because other studies have suggested the same. Clearly, more research will be necessary to sort this out.
