Obesity - A new study and what it means to be a "healthy weight"

ResearchBlogging.orgIn response to the conversation on "Obesity, Evolution and Delayed Gratification" on the main page and Razib's coverage of a fascinating new study on the relationship to the lactase gene and obesity, I thought now would be a good time to write about an important new study that helps define the boundaries of what normal and healthy weights are in humans.

This study, entitled Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies is a whopper of a meta-analysis. That is, a study that increases the power of other similar studies by combining their results so that, in this case, data from hundreds of thousands of patients can be aggregated. Meta-analyses have their flaws, and I criticize them frequently when poorly-done or poor-quality studies end up being averaged-in with the results of better-designed studies, but this one is large enough and thorough enough that its results should not be dismissed.

What this study describes is the mortality, and causes of mortality, one observes when one sorts people by body mass index. Body mass index also has it's flaws but it is a useful, if imperfect method of describing one's relative contribution of body fat to their total mass. It is calculated by taking and individual's body weight in kilograms and dividing by the square of their height in meters. "Normal" is defined between 18.5-25, overweight is 25-30, and obese is greater than 30. These numbers do not describe all people well, and you may be an exception to these predictions. This usually occurs if you have a large amount of muscle mass relative to your height, so Arnold Schwarzenegger would be obese according to these scales. However, most people are not Arnold Schwarzenegger and the scale fits, it's better not to let the perfect spoil the good. One must also remember that it would be unethical to design a study in which we prospectively made people overweight or obese, since we suspect that will cause poor health, so this is necessarily a correlative study of BMI and health. But this information combined with what we know about mechanisms of cardiovascular disease, diabetes, etc., makes a lot of sense, and I believe in the context of the literature we can make a safe assumption the effects we see are causal.

Overall what the study suggests is that the current 18.5-25 recommended BMI is probably about right, BMI of 25-30 marginally increases morbidity and mortality, and BMIs much greater than 30 significantly shorten one's life. The reason I like this study is that they have aggregated such a huge data set, they demonstrate a clear dose-response curve between obesity and mortality, and they've done a better job than most in teasing out the relationship between health, weight, smoking and other co-morbidities at all BMIs.

Let's take a look at some of the data.

Starting with their first figure, it becomes readily apparent what correlates with obesity in these groups.
i-a49db5628393cd695a33c9299dce9ab5-lancetfigure1.jpg
The main contributor to early death one would expect would be the effect obesity has on vascular risk factors such as blood pressure, diabetes, cholesterol etc. These data, which are controlled for age, sex, and baseline smoking status (except where smoking is studied), show that as your BMI increases (the X axis), your risk factors increase. Worse, there is a counter-correlation with other possible causes of disease with decreased smoking and drinking at higher BMI.

The authors do many other analyses to show this is not likely due to other risk factors like drinking or smoking. Instead, when the relationships are teased apart, the risks were mainly additive.

When one looks at what study participants died of, sure enough, vascular complications were what were elevated in the obese population (especially for males)
i-3a50dea9c59070b432cebdf5fdb11de2-lancetfigure5.jpg

When one then looks at all cause mortality of participants in this study, one gets a feel for what risk is taken as BMI increases. The top line is for smokers, the bottom for never smokers.
i-2ca99896f544bdf5c99304df04046b78-lancetfigure6.jpg
While being too skinny puts one at some increased risk, and too fat increases your risk commensurate with the level of obesity, smoking no matter what is bad bad bad. I read this as showing a healthy BMI to be between about 22 and 27, with risk increasing linearly below or over those levels.

And finally my favorite aspect of this study is the dose-response relationship between obesity and all cause mortality.
i-aee028aa4b7acc53f9da5fbe82068ca3-lancetfigure7.jpg
In these figures, a left shift of the curve = bad. It means in general you will die sooner if you are in this category. The BMI ranges of 20-25 and 25-30 aren't too different and are the best curves to be on. But once you go over 30, and especially over 40, the risk of death increases dramatically, just like a dose-response. The risk is kicking in at ages over 50 for males and over 60 for females. Indeed, for morbid obesity the authors compared the risk as being about equivalent to being a life-long smoker, and reducing your life by about 10 years.

We've talked a lot about obesity here at denialism blog, and as it is with all situations in which there is a group of people that desperately wants the truth not to be true, there is some crankery surrounding the health effects obesity. But I think that based on this and other studies we can assert some things as facts.

1. Obesity increases your risk factors for vascular death and some cancers, and in a dose-response relationship increases your risk of death especially over a BMI of 30.
2. We have sufficient evidence for a causal nature of this relationship based on studies such as these, as well as our knowledge of physiology, the relationship between obesity and the risk factors such as diabetes and cholesterol, and the reversal of these risk factors after interventions such as bariatric surgery (both gastric banding and bypass although bypass also has additional effects due to changes in physiology after the surgery and before the weight is lost). When the weight goes on you get the risk factors and the mortality, when you correct the weight the risk factors disappear, and the mortality (aside from surgical mortality) reflects the new weight status. It would be unethical to do the gold standard test of making people gain weight and then increase their risk of death directly, so drop it.
3. "Normal" or "Ideal" weight is cultural and individual, but based on BMI what should be "normal" or "healthy" is between the range of 22 and 27. Currently > 25 is overweight, and we can argue that 25-27 is the "high" range of normal. But there is no point arguing over such small differences when it is clear that after this level obesity becomes increasingly dangerous.
4. It is easier to never gain weight than it is to lose weight. I argue for primary prevention rather than secondary prevention. People with BMI over 30 should try to lose weight. However at BMIs > 40 bariatric surgery should be a consideration, and earlier rather than late to avoid increased risk of surgical complications, and increased time with higher risk factors for vascular disease.

Prospective Studies Collaboration, . (2009). Body-mass index and cause-specific mortality in 900â000 adults: collaborative analyses of 57 prospective studies The Lancet, 373 (9669), 1083-1096 DOI: 10.1016/S0140-6736(09)60318-4

More like this

I think this is a great review, except for this:

"Body mass index also has it's flaws but it is a useful, if imperfect method of describing one's relative contribution of body fat to their total mass."

I'm curious how BMI tells you anything more about body fat than a simple height-weight chart. As far as I can see, the primary advantage to BMI over height-weight is that you can compare a single number between multiple people. But since the only info that goes into it is height and weight, how can it possibly give you information about body fat proportion? Especially since, as you've pointed out, people with very low body fat may score very high on BMI?

I think the rest of this review is great, and convincing. But I hear a lot of people saying things along the lines of "BMI is so much more accurate/precise/measures something different" from height-weight, and I've never understood how that could possibly be the case. To me, that's like saying that calculating the density of iron from its mass and volume tells you something about its atomic structure. But I figure if there is a legitimate explanation for it, then you're the one to ask! Any insight you have would be appreciated.

Dr. Kate, from the first line in the paper:

Although other anthropometric measures (eg, waist circumference, waist-to-hip ratio) could well add extra information to BMI, and BMI to them, BMI is in itself a strong predictor of overall mortality both above and below the apparent optimum of about 22·5â25 kg/m2.

It's just a tool. An imperfect tool, but a good tool. The number of patients that fall off the curve of BMI is very small. We're talking about professional atheletes and body builders.

It doesn't give you direct information about body fat, but it is a measure one can glean rapidly from patients and medical records that correlates closely with obesity status. Since it's a number it makes data analysis and aggregation very simple. Because it's easy to measure it provides consistent data. After all, people know how tall they are. Weighing is easy. But belt size? Ask people that some time. You'll have some middle-aged guy with his belly hanging to his thighs saying he's the same waist size he was in college. There is also currently a debate in the literature as to what the best method is to measure a patients girth or waist size. Should it be at the belly-button? Should it be at the widest point? What about patients with a pannus?

BMI is used because it's a quick, easy measure that correlates nicely with obesity status in the overwhelming majority of humans.

This usually occurs if you have a large amount of muscle mass relative to your height, so Arnold Schwarzenegger would be obese according to these scales.

Or if you're not close to the nominal height, which is about 5'7"

The problem is that BMI scales "ideal" weight as the square of height, and constant-proportion scaling increases as the cube. BMI itself correlates fairly well with mortality risks, but some of that is beyond individual control -- there's nothing that a person can do about being two meters tall, for all that it will shorten life expectancy. Likewise, there are some lucky women who are less than 1.5 meters tall and have outrageous life expectancy despite being round as apples (and their BMIs are low, despite being plump.)

Unfortunately, one of the evils of BMI is that it's so convenient that no studies that I've seen attempt to correct for the effects of height independently.

NB: my favorite BMI example is Shaquille O'Neal: 325 pounds, 7'1" tall, BMI of 31.6: obese. The same BMI in a 4'10" woman would require her to be over 150 pounds -- nearly spherical. There's more here than muscle mass.

By D. C. Sessions (not verified) on 12 May 2009 #permalink

Yes DC,
But was is the variation in human height?

The average height for males is 69inches. The 5th and 95th percentiles are 64 and 74 inches.

The average height for females is 64inches, with 5th and 95th percentiles of 59 and 68.

The variation in human height is not so huge. 90% of men fall between 5'4" and 6'2", and 90% of women fall between 4'11" and 5'8". That's pretty tight. Also when you look at the individual data women do better at a slightly higher BMI than do men, possibly reflecting the height issue.

It's a good tool. It works the overwhelming majority of the time. So what if it doesn't describe Shaq or a spherical dwarf? Nitpicking people, sheesh.

I'm not trying to open a can of worms or dispute anything here or in the study you cite (and I did go back to the study to see if they help clarify for me and they do not), but rather I'm trying to understand something:

I get it that the studies all indicate that the higher your BMI, the more likely you are to be among those who are diabetic, have high blood pressure, and/or have high cholesterol. And that those with diabetes, high blood pressure, and high cholesterol are more likely to be among those who die at younger ages of strokes or heart attacks or whatever.

I'm not clear, however, when it comes to individuals, which factor(s) is/are most useful as predictors. In other words, which patient are you more worried about, A or B?

Pt A: BMI = 35; BP 120/80; chol. ratio = 2.0
Pt B: BMI = 22; BP 140/90; chol. ratio = 4.0

I think it's clear in the data that adding smoking to the mix is, as you say 'bad bad bad' no matter your BMI, but what if A is a nonsmoker and B is a smoker, or vice versa? Will Pt B's smoking put her at greater risk than nonsmoking A's high BMI despite their respective BP/cholesterol readings?

Does my question make sense? I guess what I'm wondering is whether the correlation with BMI -- while clearly there -- is the most useful tool we can use when it comes to assessing *individual* risk, as opposed to predicting an overall population risk, which it appears to do reasonably well?

Or am I missing something and is BMI correlating with early mortality *more* strongly than BP + cholesterol are correlating with early mortality?

Damn these data are clean! The plots almost look like output from computer models, not from real data.

I don't know how much I should believe meta-analyses, because I don't understand the methods. These findings are so clear, though, that it is very important to vet this paper and make sure it has been done right.

I'm going jogging again today...porked back up to BMI=27 or so from a previous BMI=22 from Who Cares Syndrome, and it'd be good to get back into the <25 range.

whoops, wasn't logged in to typekey for that last post

By alphabitch (not verified) on 12 May 2009 #permalink

D.C. Sessions:

The same BMI in a 4'10" woman would require her to be over 150 pounds -- nearly spherical.

That's an exaggeration. 4'10" and 150lbs is substantially overweight, but it is not even close to "spherical".

"The number of patients that fall off the curve of BMI is very small. We're talking about professional atheletes and body builders. "
And pregnant women ;)
The sex disparity is interesting. It also leads me to wonder whether female bodies are better designed for carrying extra weight (obviously a higher body fat % is useful).

I've looked at this post and also at your other cross-linked obesity posts, as I have a lifelong interest in the emotionally-loaded issue of obesity. I have read a lot of the Health at every Size (HAES) material as well. Both sides place a lot of emphasis on an unbiased reading of the evidence - so if there is a disagreement on what the evidence means, I have to assume both sides are at least engaging with it as honestly as possible.

It seems to me the elephant in the room is the single assumption that obesity is a behavioural condition, and therefore preventable, at least at some age. You appear to make this assumption (not so strongly in this post, but in your approach to the subject as a whole), while the experience of the vast majority of obese people begs to differ - including people in my own family.

What interests me is where the data might lead if we temporarily leave this assumption (obesity = preventable behaviour) to one side. Just as a thought experiment, what if the data above were plotted on a height-only chart? What if the resulting curves were exactly the same? What message would we take from it if we started from the assumption that people at the extremes of shortness and tallness are at increased risk of various causes of morbidity and mortality? Since height is generally not seen as something we have much control over, wouldn't we draw somewhat different conclusions as to what must be done?

Also, could we continue to picture the data above as if it derived from a height-only plot, and see if there are further interesting questions. Just for example, supposing extraordinarily tall people had been subject to lifelong shame and humiliation, how would we unpick the health impact of any such social stigma vs their actual height? Suppose, due to the social stigma of excess height, such people were prone to making frequent misguided attempts to modify their height - perhaps by putting heavy weights on their heads, who knows? Again, how would you unpick the health impact of their height-modification attempts from their actual heights? And supposing every time such excessively tall people went to their doctor for advice on a health problem they were told all their symptoms would disappear if they could make themselves shorter? And if constantly receiving such advice made such people afraid to go to their doctors until they were so ill they couldn't avoid it? How could you unpick the health impact of their poor reception/poor attendance for medical care from the actual impact of their height?

If you honestly tried this thought experiment, then several useful avenues for new research on obesity would suddenly have jumped out at you. So far as I know, little research has been done on the effect of social stigma on the health of obese people, nor on the effect of spending long periods of time on very low calorie diets - typical of many, if not most obese people's histories, nor on the effect of poor relationships with their doctors. If you doubt this last, think about it. You know from experience that weight loss is impossible for you, you know your doctor will insist on it. What incentive do you have to consult your doctor until you can no longer avoid it?

Weight problems are all in the mind I think,well mostly I believe,the bmi is very conservative,I know some big boned samoans that would never get close to ideal weight based
on those numbers.

The average height for males is 69inches. The 5th and 95th percentiles are 64 and 74 inches.

You might want to update those numbers.

Yes, even with the new numbers I'm on the high side of the 95% mark, but in my continuing battle to get a car I can drive safely I've had to do the research on height of US drivers. The auto manufacturers like to quote figures showing that 95% of drivers are within the design height of their cars, but that only works for drivers over the age of 50. Twenty years younger dramatically increases the median heights for both men and women.

Which means, among other things, that for constant real fitness the BMI of the population is increasing.

By D. C. Sessions (not verified) on 12 May 2009 #permalink

I'm not clear, however, when it comes to individuals, which factor(s) is/are most useful as predictors. In other words, which patient are you more worried about, A or B?

Pt A: BMI = 35; BP 120/80; chol. ratio = 2.0
Pt B: BMI = 22; BP 140/90; chol. ratio = 4.0

I would be concerned about both, because it's probably just a matter of time before B's BP will be elevated and their cholesterol too, and diabetes is almost a matter of course. There certainly are people that are exceptions. Just look at Winston Churchill. Round as a tub, smoker, drinker, and he lived to 94. Then think about Jim Fix; runner, quit smoking, skinny as a rail, dies at 52 while running. These numbers aren't a magic 8-ball. But both patients should get risk factor correction. They both could probably use exercise which lowers these risk factors independently of weight.

I think it's clear in the data that adding smoking to the mix is, as you say 'bad bad bad' no matter your BMI, but what if A is a nonsmoker and B is a smoker, or vice versa? Will Pt B's smoking put her at greater risk than nonsmoking A's high BMI despite their respective BP/cholesterol readings?

As you can see from the third figure I included, smoking is worse at all levels of BMI, even the healthy ones. Only until you get a BMI > 40 is being a smoker about equivalent to morbid obesity.

Does my question make sense? I guess what I'm wondering is whether the correlation with BMI -- while clearly there -- is the most useful tool we can use when it comes to assessing *individual* risk, as opposed to predicting an overall population risk, which it appears to do reasonably well?

Individual risk could of course be imputed from individual risk factors. If you can find an obese person with normal cholesterol, normal blood pressure and no diabetes, that's great. I would still tell them to lose weight, because it's likely that these complications will develop more rapidly in them. Until we have a genomic report card, the risk of developing the risk factors is going to be too high to ignore.

Or am I missing something and is BMI correlating with early mortality *more* strongly than BP + cholesterol are correlating with early mortality?

Obesity appears to be causing the vascular risk factors, and independently increases the risk of some cancers like breast and colon cancer. This is addressed in my post about primary vs secondary prevention. We can address all the risk factors that obesity generates and frequently extend life a great deal. However, primary prevention is better, cheaper, and more pleasant than taking a bunch of pills to correct your BP, your LDL, and your sugars.

One must also remember that obesity carries many other morbidities. Early joint damage, higher risk of infections (bariatric procedures have a rate of infection 5x more than other patients - due to high risk of the population), obese patients report lower quality of life, higher disability rates, and higher utilization of medical services. From my experience in treating obese patients I can tell you, it makes everything more difficult. They tend to be more vasculopathic, worse diabetic control, worse risk for deep vein thrombosis, worse risk for ventilator complications, longer hospital stays etc. You name the illness, getting sick is worse for a morbidly obese patient.

There's just nothing safe about it, I would never ignore obesity as a risk factor even if the other vascular risk indicators were OK.

DC, I used data from the CDC for all adults > 20 from the most recent NHANES survey.

Yer Wan, yes I understand it is difficult for people who are obese to lose weight. However it is behavioral. If you look at the data for the last 40 years, the average male carries 25 more pounds of weight and 1 inch more of height. BMI has gone from about an average of 24 to 27. There has not been a sudden shift in our genetic code. We are eating too much and exercising too little.

You also have to understand that despite what people want to believe, our metabolisms are not so crazily different from each others, and our bodies do not violate the 2nd law of thermodynamics. If energy in = energy out, then you don't gain weight. Worse for the metabolism argument, whenever metabolism in the obese is studied, pound for pound the obese are less metabolically efficient than the thin, so they actually have to eat even more to maintain weight. This has to do with the efficiency of the in-shape vs out of shape body, and the differential efficiency of aerobic vs anaerobic energy generation. We can measure metabolic output pretty easily (CO2 generation basically), and it suggests a higher energy utilization in the obese, pound for pound. The human body only gets so much energy out of food, and the variation in metabolism does not explain the differences, if anything it actively contradicts the metabolism hypothesis.

Finally, when carefully studied, and there are dozens of studies which show this, obese people underestimate food intake and overestimate exercise output when dieting. When you actually carefully study and monitor dieting obese individuals, their estimation of their activities is typically far out of touch with what is actually happening. This is not purposeful deception, they simply misjudge their activities in a fashion consistent with classic personal bias. Smokers do the same thing when they estimate their cigarette intake. Drug users with drugs. You same it. People have a personal bias that blinds them to their negative activities and overaccentuates their positive.

Everyone can lose weight. It's just lots of people need help. Sometimes lots of help. Don't buy into the obesity fatalism of the denialists. It is unhealthy, it is a problem that can be fixed, and the laws of physics are not suspended in the belly of the obese individual.

our bodies do not violate the 2nd law of thermodynamics.

First law, actually. I know the Second has gotten more press lately thanks to the evolution denialists, but this time around we're talking energy, not entropy.

By D. C. Sessions (not verified) on 12 May 2009 #permalink

BMI had always irritated me because I do fall off the curve... high bone density + high muscle mass+ under 6' in height = looks obese on BMI. That doesn't change the fact that I'm carrying 245 lbs at 5'10" which still has to require a bunch of work from my heart, I simply detest the fact that at 12ish% body fat I'm categorized as obese.

All that being said, it does appear to be a useful tool for most people. Well written Mark.

Dammit DC you're right. I'm too used to seeing evolution denialism on the sb site I've forgotten all the other laws. However the statement is still true, just irrelevant. Our bodies violate none of the laws of thermodynamics.

Great post and follow-up discussion.

I'm curious about the skinny end of the spectrum. What makes being a skinny smoker so deadly? Is there a bias here of measuring BMI in unhealthy individuals (cancer patients tend to be a bit skinny)?

I know there are plenty of rat/mouse studies that show extreme calorie deprivation can significantly increase life-span. Any chance of that showing up in these giant meta-studies?

I know there are plenty of rat/mouse studies that show extreme calorie deprivation can significantly increase life-span. Any chance of that showing up in these giant meta-studies?

Only recalling some of what I've read on the "billion heartbeats," but it looks like caloric reduction triggers a number of effects in other species which we seem to have switched on by default. Which, if so, helps explain why our lifespans are so much longer than you'd expect from comparison to other animals.

By D. C. Sessions (not verified) on 12 May 2009 #permalink

MarkH - Thanks for coming back to me. "Everyone can lose weight." I believe this is true - in the short term.

But over 95% of people cannot keep weight off in the long term. The most consistent result of weight loss efforts over a five-year term is weight gain. I know both these propositions are true of my mother, who is 70 and has been a weight cycler all her life. She has never, ever been happy. When on a "down" curve (which is always initially easy), she is hungry, tired and cranky. When on an "up" curve she simply hates herself - blaming her "lack of self-control". The thing is, she is a busy person who never sits down - she is/was a nurse, walks everywhere or runs for buses, swims every day, and her diet, even on an upcurve (when she can get very heavy, certainly well over the "obesity" line) is ridiculously mean. This is me saying this from careful observation, not her, so it is not wishful thinking!

I don't hear you taking my thought experiment seriously. But wouldn't all of this data on obesity have a very different meaning IF obesity turns out NOT to be a behaviourally controlled condition? Is it too heretical to even try imagining this for five minutes? Even if you disagree, don't you think it is a worthwhile thought experiment?

Let me put it another way. You are an advocate of primary intervention. Ask yourself this. Have you ever known a child that didn't want to run around and play at every opportunity? Children move - they never walk when they can run. It's in their nature. So if a child is disinclined to move, and is hungrier than normal, should that not be taken as a symptom, rather than a chosen behaviour? Mightn't there be an unknown "factor X" that causes both the hunger and the lethargy, and any resulting weight gain? (In an unusually thin person, we would immediately suspect reduced caloric intake or simple hunger). Is it not possible that some of that child's cells are not receiving sufficient nourishment, and those undernourished cells are causing the whole child to act undernourished?

Just as the imports across the border of a country do not guarantee that all the citizens will be fed, is it not at least conceivable that a person's fat cells may act like a country's grabbing classes - sequestering all the good stuff and locking it away from other needy cells? If that were the case, then the behaviours typical of obese people would be seen as a partial malnourishment, and perhaps would then get the investigative research they deserve.

Also, you say our genes have not changed much - but many of our outward characteristics have, not just weight - height has changed more than you grant, and so has longevity. It is seldom mentioned that Americans are about three generations away from population wide food shortages (the Depression), Europeans about two (World War II) and China about one (Cultural Revolution). We don't yet know if these increases in height, weight, and longevity will stabilise once the population as a whole has been exposed to a sufficiency of food for a couple of generations, but there appears to have been little increase in the incidence of obesity in the US over the past few years, so perhaps we are at the end of whatever growth potential is in our genes.

Sorry for the length of the post - it is just that this is an area of medicine where compassion is often lacking, and shame is often seen as being therapeutically useful. And it is close to my heart for personal reasons. I hope this has not come across in any trollish way.

Regarding will power and obesity: It's true most fat folks do not lack will power. But as Mark points out we are fatter than our parents and their parents were. This difference is not genetic. But simply observing that [taking in more calories than are put out leads to obesity] is not very useful in itself. *Why is that happening? When I switched from donuts as in-car snacks to peanut butter on whole grain bread I quickly lost 8 pounds of fat, and my running endurance improved dramatically. (My wife notes that I never run on a full stomach, and as a middle aged-man, my diet change probably improved my glycogen response. I was running in a hypoglycemic state before.)

One problem with folks trying to get leaner is that they try hard things, but not always effective things. The things we do (such as exercise) and the things we eat (such as whole grain bread or blackberries instead of donuts or chocolate bars) makes a difference to the calories we ingest and the calories we use. Grandpa didn't diet, but there's still a reason why he was leaner than the kids.

There's a large pool of confusing data out there, and an ocean of pseudo-data, too much to easily sort thru. But eating less processed foods and moving more will nearly always improve things.

Why use BMI instead of just a weight to height ratio? Our
weight is proportional to our body volume. This has three
dimensions - the toe to head axis (height), the left to
right axis (breadth), and the back to front (belly, chest)
axis (depth or thickness).

When height increases, normally breadth also increases
similarly. Depth rarely does. A 7 footer is 40% taller than
a 5 footer, but rarely 40% thicker from front to back. Note
that we don't even have a good term to describe this
dimension (unlike height and breadth) because it does not
vary much. We hear about tall people and broad people,
but not deep people (in a physical sense).

So when height goes up, breadth also goes up with it,
while depth stays mostly the same. Weight follows
volume which goes up as the square of height. Hence, for a
normal individual, the weight to height-squared should be
constant. This is the BMI metric.

I just wanted to point out that there are rare instances where people simply cannot lose weight, or experience great trouble doing so. One example, of which I know someone with it: Poly-cystic ovarian syndrome. So you should probably modify your statement, "Everyone can lose weight." I understand you mean that it is possible in most instances, but its pretty difficult for many sufferers of PCOS. My friend also wants to add to your statements this: "Before jumping on any plan, people should be encouraged to see a doctor instead, and do not fear second opinions. It took me a few tries to get the right diagnosis, and I was beating myself up over why diet and exercise alone did not work."

When height increases, normally breadth also increases
similarly. Depth rarely does. A 7 footer is 40% taller than
a 5 footer, but rarely 40% thicker from front to back. Note
that we don't even have a good term to describe this
dimension (unlike height and breadth) because it does not
vary much. We hear about tall people and broad people,
but not deep people (in a physical sense).

Nice try, but that's not how it works. Have a look at any collection of skeletal cross-sections (view along the vertical axis) and you'll see that the aspect ratio is pretty consistent.

(And, yes, I have anecdotal data too. Don't take my word for it.)

By D. C. Sessions (not verified) on 12 May 2009 #permalink

Thanks for posting that summary. The data couldn't be clearer, and some of the comments seem to confuse individual risk factors with population data. I speak as a recovered obesity denialist myself -- you can argue with the data all you want, but insulin resistance/pre-diabetes is nothing to play around with. After the diagnosis I got my weight down from a 33 to a 27 BMI (maintained for over a year) and my numbers are normal now, plus I feel a lot better. But it's not easy -- with all the enticements to eat around, plus the few incentives to move, anything less than constant attention results in gaining weight.

By Grubstreet (not verified) on 12 May 2009 #permalink

This is a good paper to see, and I appreciate your analysis of it, Mark. My thoughts on the topic are similar to those of Yer Wan, I believe, with a slightly different twist.

The study shows that obese people face serious health consequences, without doubt. I suspect that some of the obese people in this dataset had tried to lose weight, while others had never given it a moment's thought. There must have been considerable diversity in the other BMI groups as well. Some of the lean folks might have been obese a year earlier.

What is missing, I think, (and I am certainly no expert on this topic) is good evidence that a "strategy" of weight loss can head off those ugly consequences. How would we get such good evidence? How about a prospective study in which obese people are randomized either to undertake weight loss or not to do it. Such a study would (help to) answer the question whether losing weight (or at least trying to do so) can mitigate some of these bad outcomes.

I can think of umpteen reasons why we couldn't do this study. But, absent this information, I think we should continually remind ourselves that lean folks are a combination of those who are perpetually lean and those who were heavy and lost weight. These people who lost weight were self-selected to do so, and that causes a bias in the interpretation of these data.

If someone asks "would it help me to try to lose weight?" I think the absolutely most honest answer is "I think so, but there is no proof."

It seems to me that one problem in prospective studies of weight loss is that subjects are "put on" a specific diet -- and we all know that after one is "on" a diet, one goes "off" it - and often regains. But anecdotal evidence, such as that from the National Registry of Weight Loss, suggests that successful losers often make very individual choices --e.g. I will stop drinking soda, or I'll substitute fish for meat, and I'll take up running -- that simply can't be catered for in a large-scale study.

By Grubstreet (not verified) on 12 May 2009 #permalink

One thing I would like to see is a meta-analysis that looks at health risks as correlated with amount of medical care. Many obese people avoid going to the doctor for preventative care until problems are beyond easy control, and many of those who do have their main symptoms ignored because the physician simply blames their weight for everything without looking for underlying causes. Anecdotal data combined with studies of negative physician and nurse attitudes towards the obese are piling up to indicate that there might really be an institutional problem there, compounded by the reinforcing message that fat=unhealthy. It would be really interesting to see how many of those are really fat+masked problem=unhealthy.

don't know how much I should believe meta-analyses, because I don't understand the methods.

Meta analysis should not be *believed* as such. They are helpful in suggesting areas of reasearch, but bringing together disparate studies (of variable quality) does not improve the quality of data (sort of like the old dictum: the plural of anecdote is not data).

The one indisputable conclusion that can be drawn from this analysis is that regardless of if you smoke or don't, have diabetes or are overweight, statistically you will die before you reach 100. The chances of 200 are zero.

There is no perfect list of things to deny yourself that will get us over the 1.5 billion heartbeat limit. As one of those suffering a "non-smoking" disease I find it offensive that anti-medicine is used to squander health dollars on promoting sin/behaviour causes for disease instead of HPV, spanish flu, epstein-barr and all the other known viral causes of cancers.

"Preventative" medicine has been hijacked by wowsers and has moved instead to blaming the victim. Why should you treat an obese person, someone that ODs, the children of an anti-vax looney or a suicide survivor? The whole medical profession has to seriously rethink it's position - to the core. Anything else sets the bar so low we may as well go back to exorcisms and berating the sick for their sinful ways.

How many skinny wannabe non-smokers have to die from anorexia or suicide from pyschosis induced by drugs to give up smoking before the minority casualties start to count? Taking one for the team sounds good until you are the one and you find that no cures are developed because the greater good is served and the majority of suffers deserve to suffer and die? Pity the poor child awaiting a liver transplant because "gentle" acetaminophen/paracetamol was prescribed to prevent heroine addicts from getting a clean safe fix. Or the cirrhosis sufferer with Hep damage rather than alcohol, the coal miner or fireman suffering emphysema because evil smokers must die. Thanks guys.

Yer Wan, Zeroth, and Tom S.,

I don't think there should be any reason you and MarkH should disagree.

First, as MarkH said, thermodynamics can't be gotten around. If any obese person, with whatever medical condition (like Poly-cystic ovarian syndrome) were put in a prison camp and underfed, that person would lose weight and eventually become emaciated. That can't be in dispute. Can it?

But, as Yer Wan points out, for some people it is *very* difficult to underfeed. Now why this is remains, I think, scientifically an open question. There are probably a dozen or more types of factors that go into it (psychology, portion size, caloric density of food choices, interesting microbes in gut that obese people might have, genetically low rates of metabolism like the Pima Indians, leptin resistance or lack of production, Non-Exercise Associated Thermogenesis, etc., etc.). I think it is important to advocate just *how* hard it is for some people to deny themselves food when hungry, and how that hunger may not correlate well with caloric demand for their body.

But I also think there is some interesting psychology of eating and exercising, too, as MarkH points out. I have seen this so clearly in a few people I know who are obese. They happen to have an extremely strong pleasure response to eating (stronger than thinner people? Maybe some?), do not seek to know how many calories their foods contain, and also have a natural dread of aerobic exercise.

I would never morally or psychologically *blame* a person for being overweight. I have been up to BMI=31.9, and went on a 4 or so month campaign to get fit again and went down to a BMI=21.7 through caloric restriction and lots of jogging. I have since maintained most of the loss for a few years and then within the last year or two put on about half of the loss, bringing me to a BMI=27.7. This may be seen as confirmation for Yer Wan that I "couldn't" keep off the weight. I prefer to think of it as I "didn't" keep it off, and the reason in my case is psychological. I entered a dejecting and boring period of my life, *stopped caring*, stopped jogging, ate every night, didn't move much, and didn't miss any desserts out, and of course I put the weight back on, at least some.

I'd like to get Yer Wan's or others thoughts. Do you feel this is a fair assessment of the state of things?

I should say that by "for some people it is *very* difficult to underfeed" I mean it is very difficult for them to eat just the right amount. I didn't mean the prison camp guards would have trouble underfeeding them--if they were given an 800 calorie/day diet, they'd lose weight quickly (though not like the irritating "12 lbs in 5 days!" nonsense we get at the supermarket checkout, but more like perhaps 3 lbs a week for a person starting obese)

"Have you ever known a child that didn't want to run around and play at every opportunity? "

Me! I learned to read at 3 and afterwards rarely wanted to do anything else, besides eat.

This was a very valuable post. I think a lot of people, especially women, who are only overweight beat themselves up over it, and it's very valuable to see how little the difference is between 22 and 27, especially compared to the difference between 27 and 32. Figure 6 is also one of the most convincing figures I've ever seen about the risks of smoking.

A few have noted that it is not a simple question. Someone close to me has a high BMI but is on a few types of medication - one of which is to treat an under-active thyroid. There has also been considerable employer induced financial stress and a family history that I am sure all contribute.

I find it revealing that whenever someone starts discussing forcing their lifestyle on others the threat of prison camps comes up - for the victim's good of course. And when everyone has been terrorised or taxed out of smoking and drinking, shamed or forcibly compelled to fit a certain size profile all disease will disappear. Cancer will go and cardio vascular disease will be a thing of the past.

The other clear trend - and this stands out like the proverbial canine's gonads - is that these individuals that insist on being male could do a lot to improve their health outcomes if they considered gender reassignment surgery. Or we could ban them. Even a smoking, drinking diabetic woman has more social responsibility than some teetotal testosterone cowboy.

I find it revealing that whenever someone starts discussing forcing their lifestyle on others the threat of prison camps comes up - for the victim's good of course.

Oh? What does it reveal?

Oh? What does it reveal?

how obtuse ppl can be ;)

It is just so much wedge politics. We are seeing it start with both alcohol and obesity in Australia now. I guess the wowsers are embolden from their success with their anti-smoking diatribe and their coffers swollen with the billions that make the tobbacco industry profits look like chicken-feed.

The media is first swamped with horror stories of alcohol, or how deadly the french fries are, then the social responsibility card is played (think of the poor children!!! or the cost to all the wowser taxpayer dollar like the drinker or smokers taxes don't subsidise the wowsers). A case then exists that user should pay (and pay and pay). Then taxes are replaced by fines, incarceration and before you know it we have Al Capone and the crime inherent in the current draconian drug prohibition.

I don't know how medicine works down where you guys are(I'm Canadian FYI), but the evil smoker must die? Do you really honestly believe the doctors are ignoring all evidence to the contrary, and blaming smoking or obesity? Let us consider some statistical thinking. If 90% of the time, a person's problems are because of smoking or obesity, then it makes sense to consider it as a first cause. What you guys keep harping on about are the exceptions... which is not the same point MarkH was making.

He acknowledges there are exceptions, and it is the doctor's duty to know how to notice them. For example knowing someone is a miner would lead you to a different conclusion about their emphysema. If the patient knows there is a problem with the doctor's diagnosis, they should go get a second opinion. Its like the story with my friend. Even though she was one of those exceptions you guys feel so noble about defending, she points out that for the vast majority of people, MarkH is right.

Get off your high horse, understand the difference between population statistics and individual cases. Learn a little bit about differential diagnosis, or even just how doctor's diagnose illnesses. Many diseases are very, very similar. It is a very tough job, and you add on people with unrealistic expectations for the treatment. The doctor is not, and has never been, your enemy.

The plain and simple fact is, despite the exceptions, everyone should lose weight, become more active, and stop smoking. Just because people promote that, does not mean they ignore the other causes for diseases. It is just that taking those preventive measures, can, for millions upon millions of people, improve their lives drastically. A study last year showed that major positive lifestyle changes, like losing weight, turned on disease fighting genes that were formerly turned off, and contributed to the health of patients afflicted with cancer, but that had refused treatment.

To assume that doctor's will ignore all other causes to espouse that obesity and smoking is the only cause of disease is an assumption that medicine is politicized. That is a wrong assumption. In all honesty, a doctor that politicizes medicine should be barred from treating patients.

"If you look at the data for the last 40 years, the average male carries 25 more pounds of weight and 1 inch more of height. "
Fact. (I'll assume)

BMI has gone from about an average of 24 to 27.
Fact.

"There has not been a sudden shift in our genetic code."
Fact.

"We are eating too much and exercising too little."
Almost certainly true, but it is conjecture to think it is necessarily the sole cause of the above facts.

As a microbiologist, I feel compelled to point out that we are not mostly human cells. We are little ecosystems of microbes, and we are not all the same. And the bacterial genes in our ecosystems could easily change on this timescale. If you don't think so, talk to someone prescribing antibiotics 40 years ago vs. today. And certain microflora are associated with obesity (although correlation != casuation, and we wouldn't be sure about the direction of casuation anyway).

The research on mousies and what appears to be epigenetic regulation of metabolism programmed during embryonic development is also neat stuff. If I'm remembering correctly, a period of starvation during pregnancy led to offspring that survive starvation better, but are prone to obesity on high-fat diets. If the calorie consumption of the mother is high-fat and in excess during gestation, the offpsring survive high-fat diets well, but are more vulnerable in starvation conditions.

To my mind, this means there are plausible "non-genetic" and "non-behavioral"* mechanisms that may influence weight.
*Technically, it's still behavioral, but it's your parents or grandparents behavior that is up for scrutiny.

Also, invoking the first-law-of-thermodynamics is a bit disingenuous. What that law tells us is that we can never get more energy than the food we take it. It doesn't tell us the lower limit of usable energy we get out of it (Everybody Poops).
If somebody told you that, due to quirks in your metabolism, you had to swim at a Michael Phelps training schedule and eat no more than 200 calories a day to *maintain* your weight (never mind loosing it), could you do it?
Granted, baring some alien tapeworm parasite, I don't really see that happening. Thank goodness.

Just to be clear, I think overweight status and obesity are preventable with diet and exercise, at least on the population scale. I just don't believe in false dicotomies between Nature and Nurture (or Genes and Behavior) or that misinterperting physical laws is informative.

*note- a parasitic alien tapeworm is actually much more likely to cause reverse problem. an alien tapeworm in a mutualistic relationship that gets >38 ATP/glucose would be another story; but there would still be a thermodynamic limit at some point.

BMI was developed to measure populations, not individuals. Comments that most of us aren't Arnold Schwarzenegger are cavalier and do not take into account high levels of individual variation. This meta-analysis is the right use for BMI, and gives us useful information about larger trends, but people (especially on the edge of what is considered 'normal' - 25-27) should not get hung up on their own number.

And if we want to see obesity decrease in this country, the primary way to do it would be to get rid of cars for all but long trips.

It(He,She) is lamentable these things, because a time ago behind wise that the medical services were a problem for many persons and up to the moment they neither find they do not even give any solution, apparently the government forgot what promised and it is now where it is that to there be remembered(reminded), before that is very late, the medical assurance is important for many people, like that they indicate it in findrxonline, the web page that delivers a lot of information about this debate

This makes me feel better. I used to be BMI 44.7, now down to about a BMI 33 and holding there for over a year now.

Still have some to lose, but I'm thinking now I might target a BMI of 26-27 instead of 25 as my final goal (difference of about 9 pounds).

Anecdote: I retired from a stressful job in Dec '04; 5'3" 215#.
Over the next 3 years I lost 1/2# per week (~80#).
I didn't deliberately do anything at first, I just stopped 'stress eating' without even realizing it.
Maintained for a year; then in Dec '08 I had a serious run in with an icy patio and put on 5# in 2 weeks.

How much of obesity is related to stress?

If you were looking for issues, you must not have looked very hard. All of the issues that you raise have one (fairly unusual) feature in common -- they all require behavior modification. I have periodontal disease (probably genetic in origin), which as far as I know I can't opt out of, yet it leads to premature death as much as smoking does. Any suggestions?

I loathe the BMI as a measure (and the increasing favoritism of policymakers to use it as a litmus test for things like earning the right to have health insurance).

Using the BMI, I am obese. No, I am just in EXCELLENT shape, with a RHR of 60, and RBP of 90/50, good cholesterol etc etc etc. And I AM NOT a professional athlete or bodybuilder (your claim that those are the outliers with high BMIs but low body fat). I am a very normal 42 year old 5'4" married woman with kids who works a sedentary office job and who likes to be active when I can get away from the computer.

Well, normal except for the fact that if you challenge me, I can take you down. Heh.

Pink Pig:

Likely you know more about periodontal disease (PD) and risk of premature death than I do, and I know nothing about how genetic factors influence periodontal disease, but I have suspicion that any correlation between decreased life expectancy and PD might be due to a common factors that cause both, namely, lack of health-promoting behaviors like good diet, sleep, etc. Maybe the studies have tried to parse all that out, but I could imagine that people who eat diets that are not nutritious enough could predispose themselves to PD. Also, people with worse dental hygiene might generally be more lax with medical care, exercise, etc. (There is also a story about how moth bacteria in gum lesions can get into the bloodstream and damage the heart, I think?) I have been on both sides of this, in that a few years ago I "got religion" in terms of flossing, SonicCare, and mouthwash, and now my gums are *way* better than before. My point is, if your PD is 100% genetic, it may be no cause for worry, in that as long as you take care of yourself well enough to prevent PD if you *didn't* have the genetics for it, you'd be fine. If that makes sense?

epd:
If you are 5'4", you have to be at least 175 lbs to begin to qualify as obese. Maybe you have an unusually dense or large bones, or are genetically very muscled, but for most people 175 lbs on a 5'4" frame seems like it is overweight (whether most people's expectations are right...that's another issue). Whether it should be called "obese" is a reasonable question. When I hit BMI=30 I thought, "What? I'm obese now?" and it struck me as odd, even kind of funny. People seeing me on the street at that time would not use the word obese, I'm sure. So maybe the names need to get more finely cut.

But the point of the study is that--call it what you will--BMI seems to be correlated with health outcomes. If you are "obese", BMI-wise, and have wonderful health markers otherwise, it is possible you will be one of the people that doesn't fit the curve well. It is just a description of the population data. For myself, I *did* have worse health readings when "obese", and it was important for me to bring it back down to a regular range.

Interesting study, useful for the layperson. (Like me.)
I know BMI is useful, but I always found it a funny measure, as I get heavier when I work out. I'm 5'6" and when I drop below 175 I know I'm getting flabby. To determine this correlation I use the reflective albedo test. If I look in the mirror and my belly stares back at me like a full moon, I am in need of the gym. I am hardly Arnold, but stocky little people like me fall off the curve pretty easily.
I think it would be interesting to see how the data from all these studies matches up with ethnic/regional groups and body types. I would not be surprised at all to find that some ethnic groups have different relationships to BMI and morbidity.Someone mentioned Samoans, but I would be interested in climate adaptations as well. Do tropical people such as Kenyans have a different ideal BMI than Nordic and Slavic folks? This might be very useful data for populations that have greater risks. I think everyone is familiar with the Navajo response to the modern diet, but I think there may be more variation in our history dependent adaptations than we know at present. This sort of study might also reveal a cause (not likely the only one) of the much discussed variation in difficulty that people encounter in reducing BMI.

By dennymack (not verified) on 13 May 2009 #permalink

Mark, great post!
I have a question for you (and all the commenters). In the second figure, there's a marked decrease in annual deaths due to GI cancers. It's around BMI 24 or so, but already evident at 22 and stays flat even at max BMI. Does anybody know what that's caused by? Ditto the respiratory deaths, really.

Is it perhaps vice versa, that GI cancer/resp. patients lose weight before dying?

Torben- I know some physicians working with predominantly geriatric populations (in assisted living facilities and the like) view weight loss as more often a bad sign. If you're already old and light, getting lighter isn't a good thing.
I also know that lung cancer definitely has an association with low weight; and it's been attributed to the disease making people underweight at late stages.

Thanks for an interesting take on this. I'm 5'11", about 200, BMI about 27. My MD doesn't worry too much about it because I don't smoke and have normal blood pressure. A phsyician friend jokes that I'm what he calls "undertall" - a little pudgy but nothing to get excited about, and would be the right weight if I was a bit taller.

I suppose I am one of those people you would consider a denialist. I'm fat (you'd probably say morbidly obese, but fat is quicker and I prefer it). I don't deny that.

I've tried losing weight...and succeeded moderately. And gained it back as many people do. It was really not pleasant to become so obsessed. I did not like it. I became clinically depressed when I was slimmer. I probably qualify for weight-loss surgery, but I refuse (especially because I have celiac disease) to understand why damaging a perfectly good system is a good thing. We are experimenting and there have, for many, been disasterous consequences.

So where does that leave me when people confront me with the "fat people die earlier" stuff? I'm left feeling like my right to make decisions about my health, even one's that other's may not approve of, is on the verge of being taken away.

Someone above mentioned that even if everyone got thin, we would still have disease. Even if everyone looked a like, there would still be catastrophic events like strokes, heart attacks, etc. in the population.

I choose to be the size that I am, regardless of population risks, and I ask myself in this body that I love how do I become the healthiest that I can be? But most people never get to hear the idea that they can be just fine the way that they are...and maybe if they heard that, they actually might be more active in their lives, or spend less time in self-destructive behaviors because they feel shamed by the general public for being "unhealthy," "lazy," and "ignorant."

My rant is a bit disorganized, but I'm rushing and I'm angry (again) at the way that our society loves to moralize about me and my body.

Mark,
Can you explain how the final figure is derived? I can't tell if old obese people are rare because they die or because they lose weight and change categories. Is the BMI shown there the BMI at that age, or the integrated lifetime BMI?

Looking at data on the weight and height of the US population from the latest NHANES study, I came up with a formula to replace BMI. I call it cubic BMI, or CBMI. Take your height in feet (so that 5'5" = 5.5 feet) and cube it. That's your ideal weight in pounds, if you're an American.

Ummm. 5'5" does not equal 5.5 feet. It equals 5.4166667 feet.

And for people over 6 foot, this formula recommends being a fat-ass. There is no way I would be healthier if I gained 50 pounds.

Lab lemming, it's called sarcasm.

spinsterwitch,
you said:

My rant is a bit disorganized, but I'm rushing and I'm angry (again) at the way that our society loves to moralize about me and my body.

Who is moralizing here? If there is a study that shows that people of a certain BMI have different health outcomes, that is not moralizing. If MarkH points out that metabolism must follow the laws of thermodynamics, that is not moralizing. No one is moralizing. It is a discussion of health matters.

I have struggled with weight, too, and I have been in the obese range, which gave me a higher probability of having later negative health outcomes. Papers like this goad me to get in better shape--I NEVER see them as moralizing. (I am also a scientist, so that might predispose me to see things this way).

No one should judge a person morally for their weight, though people often do, but that is to their discredit.

I wonder whether obesity is the other side of eating disorders? I volunteered around ED sufferers for years and the arguments of anorectics are almost mirror images to what obesity denialist say. Oh, I cannot gain weight, it's my metabolism. I would like to gain weight but it's sooo hard. It cannot be that bad, Audrey Hepburn (or whoever else) starved herself and look how long she lived. I would answer something along the lines of Well, despite the existence of long-lived and emaciated individuals, you're very likely to die slowly, painfully and miserably. And, when someone mentioned concentration camps and how everybody lost weight there, well, anorexic patients are in the severe cases locked up, restricted in movement so that they cannot work out and forced to eat. Sure, it's a voluntary therapy but there are people who opt for it because they cannot manage themselves. But.... well, you may get my point.

By kultakutri (not verified) on 17 May 2009 #permalink

When the laws of thermodynamics are brought into the argument, it strikes me that almost never is there mention of the actual processes involved, but that a calorie in the mouth is assumed to be a calorie in the blood, a supposition which conveniently skirts the fact that calories come to the blood by diffusion, a process which is always far from 100% efficient and which is dependent on material properties one might expect to have some genetic variation. Do we all have the exact same intestinal topography after all? Or the same conversion efficiencies from chemical energy potential to mechanical power in muscles?

CM - you are right...the data never moralizes, just the people who use the data. To imagine that this data is not going to be used to fuel fat hatred is short-sighted or completely blind.

The term "health" is increasingly being used as a moral yardstick to judge a person's lifestyle (and their relative worth in society), and I would argue that this is wrong.

Avery Ray Colter:
I believe the processes whereby food calories are used in the body is well (though not completely) understood. And of course, efficiency of the process varies by animal and across people. The Pima Indians are one group that seems to have a genetic propensity for obesity, possible because they lived on a low-calorie diet for enough generations due to living in the desert (and now that calories are plentiful their bodies store them as fat). But the point of thermodynamics is that NO ONE can beat thermodynamics, no matter how efficient a user of calories one might be. This is to say, at a sufficiently reduced caloric intake, any person MUST lose weight, or at very least stop moving and generating heat (that is, they would be dead), because if not they are producing heat and kinetic energy out of nowhere. I suppose it is possible there are people with mutations in which their fat cells are unable to liberate fat for fuel, and so will not burn stored fats as fuel, and so will pass out and ultimately die still overweight, but if so I am not aware of this. I can't imagine this could be common, though--the selective pressure against it would be huge unless the genes evolved in an environment in which calorie sources were always guaranteed to be around.

spinsterwitch:
the data never moralizes, just the people who use the data. To imagine that this data is not going to be used to fuel fat hatred is short-sighted or completely blind.

I disagree. I see no evidence on this blog post of anything like moralizing (and therefore one example of using data and no moralizing), nor do I have any reason at all to believe that the scientists who did these analyses are moralizing. I certainly do not believe either of these groups will do anything remotely resembling intentionally fueling "fat hatred". I have seen doctors treat an obese person (BMI>40) very kindly and gently advised small modifications of diet and I have seen obese friends treated very respectfully for close to 20 years.

Does this mean there are not mean-spirited people who mock people for being overweight? Obviously there are, and very many of them. Those people are ethically wrong, callous, unsympathetic jerks, or they are children. But I would bet that the overlap in that population and the population who carries out, writes, reads, or blogs about studies such as this one is vanishingly small.

I would welcome examples you might have of actual scientists or doctors "moralizing" (that is, saying that being overweight is a moral wrong as opposed to an unhealthy choice) in regards to this kind of data.

The term "health" is increasingly being used as a moral yardstick to judge a person's lifestyle (and their relative worth in society), and I would argue that this is wrong.

I live in this society, and I am just not seeing that. What I see that I find objectionable--and very common--is that whenever the 10 o'clock local news does a segment on obesity they show appreciably overweight people walking down a public street, but shot from the neck down. They often feature the person shot from the lower back and down, or they focus on the abdominal area. I find this silly and in poor taste, but not a moralizing.

So where are the cases in which health (why do you put it in quotation marks as "health"?) is used as a moral yardstick?

Maybe you mean something by "moral" than I do? That is not sarcastic.

I really think there are many many doctors and scientists who want nothing more than to either understand the mechanisms of metabolism/anorexia/obesity/etc. and/or to help people whose lives are somehow compromised by weight-related issues. By painting all of society with the same very broad brush, I think you are bringing on anger where it need not be. Yes, there are some really poorly socialized people out there who are cruel and mean-spirited to people with dwarfism or giantism, very scrawny people, obese people, or those with any condition that is not the norm. But one should try to not concern oneself with their opinions. I know sometimes it is difficult, though.

Best wishes.

There is lots of moralism directed at fat people: that they are lazy, that they choose to be fat, that they are self-indulgent and killing themselves, etc. The "poorly socialized people" are just the tip of the iceberg. The above beliefs are widely, politely, accepted, even if they are not voiced in a rude manner. But it's not true. There is nothing morally wrong with fat people.

But it's also true that lots of excess fat is not good for your health. AND, it's also true that losing weight and keeping it off is very hard. I know that a lot of fat people are angry at the bad news about fat, but it's there, and it won't go away. I know you didn't write this posting to hurt people, just to tell the truth.

I haven't read all the comments, but I wanted to say:

Thank you for breaking down a great big study that tells us that there's a positive correlation between being in the middle of the BMI range and living a longer life.

But that's all the study tells us. Those of you who are saying "people need to lose weight to live longer" aren't being fair or using the data properly. The study doesn't say the fat (or lack thereof) is what's killing people. It just says very fat (or very thin) people (assuming the BMI is being "accurate" with the majority of them) are less likely to live long lives, and thus the "why" behind that could probably warrant more study. Even if this study "teases out" some of the big co-morbidities and breaks things down to BMI, there's no way it pulled out all other variables other than weight. Someone already mentioned that fat people may be less likely to go to doctors, and the same can apply to thin people, since there's the pervasive message in this society that you don't have to worry about your health if you're already thin (which is insanity). What about mental health, job choice, the company one keeps? Do you think these things aren't affected by weight, and that THEY can't affect length of life? And what about all those cases where weight gain/loss is the RESULT of the shortening of life? People with chronic, terminal conditions probably have weight as the LAST thing on their minds, so they may stop eating or eat all the time to cope, throwing them on either end of the BMI scale. Positive correlation between too thin/too fat and shorter life, right there!

I'm not a statistician, but I am a researcher (biochemical with a clinical slant, to be exact). And when it comes to human beings, there's always a million variables coming from a million angles, ESPECIALLY when you're looking at a huge population. Nobody's metabolism is the same as everyone else's, and those who are citing the first law of thermodynamics should go back to the lab and work within a controlled, closed system if they want to see anything resembling a neat answer like that. People don't live in bubbles and their bodies don't all work the same.

This study doesn't say losing weight or gaining weight will necessarily lead to a longer life. Those of you who think that are putting your own biases onto the data. It's saying the WHAT about BMI and life length, not the WHY. And BMI and health should be taken with a grain of salt to begin with.

This is a great post, and I love the discussion afterward. I love studies that show correlations between things. Our society, as a whole, earns more than it did 50 years ago. Our society, as a whole, is fatter than it was 50 years ago. So, that leads me to believe that a society that earns more gets fatter. Maybe wealth=fat (I swear I read something about that somewhere.... Oh, that's right, it was in my history books). Also, society, as a whole lives longer now than 50 years ago. Maybe fat=live longer. People here want to discount individual stories because this is supposed to be about populations. To look at BMI alone and ignore all other factors is either crazy or ignorant. Or both. I don't care what a chart created by an 18th century mathematician has to say. The chart they use wasn't even originally intended as a measure of health, anyway. This study just shows that they can come up with a lot of numbers that don't actually mean a lot.

Now, do I believe that everyone should just sit around eating and not exercise? No. Get at least half an hour of exercise a day. Walk more. Use the stairs more. Don't be lazy. Eat good, wholesome foods. Like chicken, and spinach, and wheat. Get some more milk and less soda.

Stop obsessing over what a stupid chart says. If you feel healthy, most likely, you are. If you feel like you can hop out of bed in the morning, take on a full days work, and come home to enjoy the family, then you are probably healthy. I can point to a million studies that say a lot of things (most of them made up by me - some of them taken as science). Don't believe them. Evey time you purchase some weight loss product, you are funding the same jerks that are making you feel bad about your weight in the first place. I can honestly say that the biggest thing you can do for your health is run for 20 - 30 minutes a day. Try to enjoy it. Go to a peaceful park or nature trail, and run. Take in the scenery and breathe the fresh air (if you can). Then take a shower, and be amazed by the energy increase.

I am NOT trying to play both sides. I AM emotionally connected to the issue. This BMI junk drives people insane. People make poor decisions in pursuit of this and some people actually die (remember phen-phen?). Just be smart and eat decent - not good (no one really knows what good is) - and exercise. You don't need that new drug, and you certainly don't need that ab-a-whatzit. You also don't need that extra piece of cake. Don't live lavishly; but don't live like a monk, either. Everything in moderation, and you will be fine. If all of that still makes you obese, then so be it. At least you will be a healthy obese.

Weight problems are all in the mind I think,well mostly I believe,the bmi is very conservative,I know some big boned samoans that would never get close to ideal weight based
on those numbers