Slashdot Mirror
Causes of Death Linked To Weight
An anonymous reader writes to mention that while a couple of years ago researchers found that overweight people have a lower death rate than people with a normal weight, it may be more complicated than that. "Now, investigating further, they found out which diseases are more likely to lead to death in each weight group. Linking, for the first time, causes of death to specific weights, they report that overweight people have a lower death rate because they are much less likely to die from a grab bag of diseases that includes Alzheimer's and Parkinson's, infections and lung disease. And that lower risk is not counteracted by increased risks of dying from any other disease, including cancer, diabetes or heart disease."
Personally, with my current health state, I don't want to live forever. And yes, I live in what most believe to be the most technologically advanced society on the planet, however, medical technology ain't cheap. What good is top-notch health care if you can't afford it?
Understanding the scope of the problem is the first step on the path to true panic.
Clearly, NOBODY EVER thought to try to control for other health factors in the study.
Obviously, you, and only you, have noticed this awful, systematic flaw in this study that obviously didn't have to pass an kind of rigorous review process to get published in JAMA.
Praise be, we've found a new Einstein!
Just remember that the average height in the Western world has gone up significantly in the last century due to better nutrition. A lot of the really old people who are really short also didn't get the best quality, variety, or quantity of food when they were growing up, which is a contributing factor to their shortness.
I bet you the average height of men 80 or older has gone up at least 3 inches in the last 30-40 years. By the time you'd be 80, who knows where it will be?
Start a happiness pandemic
You're technically correct, but I disagreed.
BMI, as a measurement of fatness, gives a really good combination of coverage, it's accurate for the majority of westerners, it's very easy to check and it's easy to explain to people.
Sure, it breaks down when applied to fit people, but works a charm when applied to fatties or anorexics. I honestly can't believe that anyone who was fit enough to be over BMI 25 with muscle would ever think the scale has any application to them.
So, it's technically wrong but practically good.
If they weighed a person suffering from lukemia - by the time the disease had devastated the body - they wouldn't be fat anymore! Therefore... skinny people die young! Stupid.
Why is there a category missing completely? They've got underweight, overweight and obese. Where's the "target weight" category?
More
BMI combined with a shred of common sense is a perfectly fine approximation of obesity.
Maybe it's common sense that needs to be taught in schools then.
Seriously, now we're going to have to deal with a bunch of obese people pointing to this study as evidence of why they don't feel a need to lose weight.
Something's obviously missing in this study, because there is a positive correlation between average lifespan and obesity rates, both when comparing countries around the world and when comparing historical rates within this country. The simple fact is that all else being equal, the fatter a population is, the shorter its average lifespan. The United States, for example, ranks 42nd in world life expectancy - Japan, with much lower rates obesity and average weight, ranks #2. (Behind Andorra.)
Not to even mention other studies (like this one, for example) that show that being even moderately overweight can increase your risk of heart disease by more than 30% - and that's our nation's #1 killer. That's to say nothing of diabetes.
I'll take my chances on being thin, thanks. One study that appears to contradict all scientific knowledge we've accumulated to this point isn't going to change my mind.
So, less than 100%?
Yes - assuming you define the death rate as no. deaths/no. births then it will be less than 100% and will remain so while the population is above zero. To define it otherwise is to assume that everyone alive now will die which, while likely, is not yet proven. Afterall I think I speak for most of the human race when I say that we all want to live forever, or at least die trying!
Maybe I am a rube, but don't all people or groups of people have the same death rate? 100%?
Yes, of course. My family line are all a bunch of meat eating farmers who were tall and lived to old age. You really believe all that crap?
Please sign petition to restore sanity to our banking system!!!
http://financialpetition.org/
You say (and bold), "all else being equal," and then compare the US to Japan.
Exam 4/C again. Maybe I'll do better this time.
First of all, I'd rather be physically fit even if doing so shortened my life a bit. The increased quality of life would still be worth it to me.
But you're very correct that BMI is a useless indicator of fitness.
I'm a tad over 40 years old. 4 months ago I was diagnosed as pre-diabetic. I drank 4-5 liters of Pepsi per day. I looked like a big blob of partially hydrogenated whale blubber, marinated in high fructose corn syrup made from genetically modified corn. I couldn't climb 2 flights of stairs, bench-press half my weight, or even take what I now consider to be a deep breath.
Knowing either it was this or insulin, I decided to hit the gym, and to give up the Pepsi. It's worked out very well for me. I've more than doubled my strength, and probably improved my cardiovascular endurance tenfold (I couldn't walk half a mile before; now I can power-walk for over an hour, although I still can't jog or run because the weight is too hard on my legs).
I feel infinitely better, and even look somewhat better.
Yet my weight only dropped slightly during this whole time, by about 10 pounds (from 215 to 205). Thus my BMI also changed only slightly. I've gained significant muscle mass, and thus lost significantly more than 10 pounds of fat. The increased muscle mass should, with any luck, help increase my metabolism, making it easier to burn the remaining excess fat. The BMI reflects none of this. According to the BMI, I was obese before, and only slightly less obese now. But I sure as hell do feel better, and it seems pretty certain that I will have a far better quality and quantity of life, assuming I keep up the present level of exercise (and don't get hit by a bus), than if I did not.
Nonaggression works!
The implication in all these kinds of stories is "If you don't die from fill-in-the-blank disease, then you will never die." Of course, the real case is you will die from something else.
If I have to choose, I choose to die from being a lazy, gluttonous pig. If you are working hard to not be a lazy gluttonous pig, guess what? You'll die too. At least I had fun.
The best method available (without special equipment) is the ratio of waist size to height.
There's my problem, I'm too short!
-- Alastair
It's not a controlled study. It's a correlation study. We need to be careful about mixing up the idea of controlling for a factor in an experimental design, and controlling for a factor in a correlation or meta study. In the latter case, it's important if you want to come up with interesting correlations, but I doubt you can ever be that sure you've really disaggregated the confounding data.
For example, note the huge fraction of reduced deaths among the overweight that is attributed to lung and respiratory illnesses. Now if they've successfully removed smokers from this analysis, it's a verrry interesting correlation. It might mean there is some mechanism heretofore unknown by which body fat protects against lung cancer and acute respiratory infections.
On the other hand, it isn't entirely unreasonable to suppose it's possible they might have mixed in a few smokers into their non-smoking overweight group. Smoking has a huge impact on both body weight and respiratory disease, and failing mixing up smoking with body weight would be expected to produce something very like we're looking at here. Isn't it possible their datasets, which rely on self-reporting, under-represent the rate of smoking? I bet the life insurance company datasets under-represent smoking, and I wouldn't be surprised if the same was true for the NHANES data too. When Mr. Government worker asks, could enough smokers be untruthful with him to skew the results?
If your life depended on getting the answer right, which hypothesis would you choose: an unknown mechanism that protects fat people from lung cancer, or a failure of a self-report data set to count all the smokers?
I'm not saying this is a bad study. It is probably a good one. But these kinds of studies aren't supposed to give you answers; they're supposed to raise productive questions. Anybody who uses this kind of mainstream news outlet report of statistical or meta studies to guide his personal health choices is a fool.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
I wonder how much constantly being reminded that you're going to die contributes to your death?
Unless I'm misunderstanding you, you're confusing causality regarding weight with health. If a person is overweight or obese because they don't get any exercise, they're going to have a piss-poor, plaque-laced cardiovascular system. If a person who is naturally overweight, or just eats a bit too much, exercises regularly they are going to be in better internal shape than a thin or normal-weight person who doesn't exercise. This is, of course, assuming equal diets between the two. An improper diet can also destroy your life expectancy.
Numerous studies have already shown that regular exercise does more for your health than simply losing or gaining a little fat. Obviously, there are still studies to be done, but I think this reason, combined with the flaws in BMI, explains the results fairly well. I still haven't read the actual text of this study, so it's possible that they took these into account, but I seriously doubt they did. It seems like they analyzed larger NIH studies instead of taking their own data.
Another factor to consider is social pressure. People don't want to be fat, and will spend billions to try to avoid it. And, despite popular beliefs, many of them work very hard at it and are still unable to change their weight/appearance. Being fat means you are automatically considered lazy, dumb, stupid, and any number of other negative stereotypes. I don't have any data to back this up, but I'm willing to bet that fat/overweight Americans are actually more likely to work out regularly. Since regular exercise does more for your health than a moderate layer of fat hurts it, this would, in an odd way, actually mean that being overweight indirectly leads to better health.
BMI is, of course, heavily flawed. It doesn't measure resting heart rates or take body fat into account. Every body builder I've ever met is obese according to BMI. Brad Pitt is overweight. This means that a society composed entirely of body builders would have 90%+ obesity rates using BMI. BMI is still a decent statistical tool if you use it properly and consider trends in society that can effect the body composition of a significant portion of the population.
BMI should never be used to evaluate individual health. I, for example, would need to have less than 5% body fat to be in the "normal" BMI range. If a doctor wants a good excuse to stop practicing medicine, they can start advising people like me that we need to reach that "normal" weight.
But seriously, to sum it all up, "It's ok to be a little overweight. No matter how much you weigh, exercise regularly if you want to live a longer life. Severe/morbid/chronic obesity is bad for your health."
SWM seeks new sig for a brief fling