December 02, 2010
Body Mass Index Above 25 Raises Mortality Risk

Pooling data from 19 long term studies researchers find shorter life expectancies for body mass indexes above 24.9.

A study looking at deaths from any cause found that a body mass index (BMI) between 20.0 and 24.9 is associated with the lowest risk of death in healthy non-smoking adults. Investigators also provided precise estimates of the increased risk of death among people who are overweight and obese. Previous studies that examined the risks from being overweight were inconclusive, with some reporting only modestly increased risks of death and others showing a reduced risk. Also, the precise risks for different levels of obesity were uncertain. The research team included investigators from the National Cancer Institute (NCI), part of the National Institutes of Health, and collaborators from a dozen other major research institutions worldwide. The results appear in the Dec. 2, 2010, issue of the New England Journal of Medicine.

From 25 to 29.9 BMI all cause mortality went up 13%. Your risk will be higher at 29.9 than at 25.

They found that healthy women who had never smoked and who were overweight were 13 percent more likely to die during the study follow-up period than those with a BMI between 22.5 and 24.9. Women categorized as obese or severely obese had a dramatically higher risk of death. As compared with a BMI of 22.5 to 24.9, the researchers report a 44 percent increase in risk of death for participants with a BMI of 30.0 to 34.9; an 88 percent increase in risk for those with a BMI of 35.0 to 39.9; and a 2.5 times (250 percent) higher risk of death for participants whose BMI was 40.0 to 49.9. Results were broadly similar for men. Overall for men and women combined, for every five unit increase in BMI, the researchers observed a 31 percent increase in risk of death.

You can calculate your BMI here. Then dial back your weight to see how low it has to go to get below a BMI of 25. That's the goal to shot for.

Update If you are a low fat, high muscle guy then you could have a high BMI that does not indicate higher disease risk. But usually BMI is a decent proxy for body fat because so few people are buff, especially among those in middle age and beyond.

Share |      Randall Parker, 2010 December 02 10:30 PM  Aging Weight Studies

arandomperson said at December 3, 2010 5:42 AM:

It would be useful to see a controled study comparing weight-lifters (those with high amounts of muscle thus higher weight thus higher BMI then 'healthy') to non-weightlifters (lower amounts of muscle) who have the same overall weight/weight (and thus BMI). Controls for amount (percent) of body fat would be nice too.

Compare those two group's mortality rates and see if it really is mass or something else.

My current BMI is 29.6. For example, I can deadlift 1 and 3/4ths of my body weight. My waist hovers between 35 and 36 inches. I am 45 year old male. Yet with my BMI of 29.6 I am considered overwieght and unhealthy.

I simply don't believe that I am unhealthy.

WJ said at December 3, 2010 6:07 AM:

I have to agree with arandomperson in questioning the value of BMI data. I entered my present weight on the BMI calculator and it told me I was fat (25.9). I found the weight at which I would be under 25 (about 10 pounds less) and it's a reasonable weight, but as someone who's been weightlifting the last 4 months, losing weight has not been my goal. I then tried to find what the lower limit on my weight would be. It put that at 145 pounds. A wide-shouldered, 6'2", 145 pound man. Sure. I once was somewhere around that, in high school. I was a walking stick. At that BMI you might live longer, but you won't be getting many dates.

As the above commenter said, test (non-steroid using) weightlifters above acceptable BMI and compare them to the random population.

random said at December 3, 2010 7:51 AM:

This may be a more helpful calculator: WebMD BMI Plus Calculator

The BMI calculator is the same, but it also includes a waist-to-height ratio:

"A waist-to-height ratio under .50 is generally considered healthy. This ratio may give a more accurate assessment of health for people who are muscular or for women who have a "pear" rather then an "apple" body shape."

PacRim Jim said at December 3, 2010 8:56 AM:

We're all agreed that mortality is undesirable, right?

kurt9 said at December 3, 2010 3:20 PM:

BMI is a fraudulent metric. The appropriate metric is body fat composition, where the fat is measured as a percentage of your total body weight. BMI does not apply to those of us into resistive weight training (e.g. body building) nor does it work for those who are "big boned".

Aron said at December 3, 2010 5:18 PM:

I tend to believe you don't get much credit towards longevity from weightlifting and exercise. If it was so healthy to be fit, the body wouldn't so avoid it. The most plausible explanation for why the body rapidly deconditions itself is to reduce base calorie consumption. But how plausible is that really? A pound of muscle burns only 4 calories more per day than a pound of fat, and the muscle is much more useful when it comes to appropriating food. And what is the difference in caloric demand between a cardio-fit individual and a non-cardio-fit individual? So why does the body choose to be unfit?

I of course agree that the BMI-mortality science is a sledgehammer, when we want something more precise..

bbartlog said at December 3, 2010 7:37 PM:

BMI is used by researchers (well, epidemiologists...) because the data is dead easy to gather. Huge datasets can be composed. But yeah, don't use it yourself - for personal health evaluation it's retarded.

Anyone have a link to the actual paper? I've come to distrust the for-public-consumption summaries they put together, and when I see weaselly phrases like 'Results were broadly similar for men.' (no numbers, after giving us a complete rundown of women's risk ratios) I really want to see the original dataset.

bbartlog said at December 3, 2010 7:55 PM:

'From 25 to 29.9 BMI all cause mortality went up 13%. Your risk will be higher at 29.9 than at 25.'

No, and not yours either, unless you're a woman. The summary does not give information on male risk ratios. We know from the summary that in aggregate (when they put men's and women's risks together) the risk goes up (from the sub-25 category to the 25 to 29.9 category). And they do say that results for men are 'broadly similar'. This still leaves open the possibility that the all-cause mortality for men is the same (or even lower) from the sub-25 category to the 25 to 29.9 category. Public health academics will often write the summary to try to support the consensus view, rather than give an unshaded view of the data.

Bruce said at December 3, 2010 8:20 PM:


"Not surprisingly, researchers found that death rates were significantly lower for fit women than for unfit women. But surprisingly, fit women with high BMI, body fat percentage, waist circumference and waist-to-height ratio had no greater risk of death than women with normal values.
So, researchers say this suggests that fitness is a stronger predictor than thinness for determining a long and healthy life."

Randall Parker said at December 4, 2010 8:54 AM:

kurt9, WJ, arandomperson,

People who have a lot of muscle mass are sufficiently rare that BMI is a pretty useful metric when dealing with large numbers of people, especially people who are past their 20s. These sorts of studies tend to be done on people who are in their 50s and older.

Look at it this way: If the researchers are not controlling for muscle versus fat then if anything their results understate the harm of being fat.

PacRim Jim,

Depends on whose mortality.

Tom Bri said at December 5, 2010 5:34 PM:

I note the lack of interest in the opposite side of the coin. If the peak is between 20 and 24.9, then those below 20 also have increased risk. How much? Be nice to know. Is being too thin more or less dangerous than too fat?

Also, they had to get to nearly a BMI of 40 before they got to an effect great enough to be of any real concern. Small increases in mortality in a mass study like this don't translate into anything in particular for an individual. It still comes down to how long your grandparents lived, how generally healthy you are, and a lot of chance.

More junk health science.

me said at March 15, 2011 12:53 PM:

What it amounts to are more reasons to "hate" heavy people. Give them a break. They're not doing it to annoy you.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright