Subscribe to our mailing list

* indicates required

Monday, March 04, 2013

Body Mass Index vs. Longevity: Latest Findings

The other day, I wrote about the extraordinary amount of evidence (going back almost 80 years) linking calorie restriction (CR) to increased longevity. The effect has been seen in every species of organism tested, from yeast to worms to flies to rodents to primates. Organisms that eat ad libitum (as much as they like) tend to live shorter lives than their peers who are calorie-restricted. The effect is dramatic. The average lifespan of rodents can be nearly doubled by cutting daily calorie intake to a level just above starvation.

Evidence for the CR effect in humans tends to take the form of data showing the life-shortening effect of obesity. Graphs that make the effect clear are available in "Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies" (The Lancet, Volume 373, Issue 9669, 28 March–3 April 2009, Pages 1083–1096). One such graph is shown below.

Kaplan-Meier age versus survival curves for people of various body mass index (BMI) classes, based on data for 900,000 Western Europeans and North American adults. From The Lancet, Volume 373, Issue 9669, 28 March–3 April 2009, Pages 1083–1096.
What this graph shows, first of all, is that survival rates go down as body mass index goes up. (To find your own BMI, consult the chart below.) If you are morbidly obese (BMI of 40 or more, indicating that your actual weight is twice your ideal weight), half of your peers are dead by age 70. If your BMI is 22.5 to 25, only 23% of your peers are dead by age 70.

Interestingly, while there is obviously some lengthening of maximum lifespan with lower BMI, it's only on the order of 10% (for the range of BMIs studied). At a BMI of 35 to 40, you still have a chance of living well into your 90s, though not much of one. The oldest fat person is roughly as old as the oldest thin person, but a really old fat person is a rarity. Certainly that accords with everyday observation. You don't see many fat 90-year-olds in real life.

Click to enlarge.
Still, it's odd how all the curves "bunch up" at ~100 years max lifespan. This is at odds with (for example) the rodent data, which show vastly different endpoints for the CR survival curves. It could simply be that we're looking at two different things: The rodent data reflect calorie restriction, whereas for humans we're looking at body mass index, and a low BMI doesn't necessarily imply CR. In human populations, variations in BMI could just as easily imply variations in the amount of exercise, or it could simply relate to genetics. Some people have the "fat gene," others don't. In all of the rodent studies, individuals were selected from a common gene pool; variations in metabolism were thus slight. With the large, heterogeneous human populations involved in the Lancet study, there were/are undoubtedly huge genetic variations.

It could also be that there's a "brick wall" effect at around 100 years because of some unavoidable buildup of some "toxic something" in the human body that happens after 100 years of living. The possibilities here are many. On the other hand, rather than toxic-something buildup, we could be looking at something as simple as telomere shortening.

The Lancet-study graph doesn't show curves for BMIs lower than 22.5, yet we know that 22.5 is quite far from starvation. (The current consensus is that starvation onset corresponds, roughly, to a BMI of 15.) In one of the most heavily studied Blue Zone populations, namely the native Okinawans, BMIs tend to run from 18 to 22. Oddly, though, even the Okinawans seldom live past 104, and centenarians number only 54 per 100,000 people (about twice the rate of France). So even in Blue Zones, there seems to be a kind of brick-wall effect at around 100 to 105 years of age.

In any case, one thing is certain: If you want to increase your odds of living to a ripe old age, keep your BMI as low as possible (but no lower than 18). You don't have to exercise more; you just have to eat less (and eat smart; avoid animal fat and avoid inflammatory foods).

And if all else fails? Move to Okinawa.


  1. Anonymous6:08 PM

    Your BMI range graph in fig 1 is missing the underweight group. Why? This would support the calorific restriction idea.

    1. Anonymous6:39 AM

      Read the text. Not just looky at pictures.

  2. Cf more recent research 2013 from Monash univ research stating optimal bmi for lifespan iss somewhat overweight

  3. One may wonder how obesity leads to oral issues. Obese people have a higher concentration of adipose tissue which produces cytokines and hormones. This leads to inflammation of the different parts of the body and leads to various diseases like cardiovascular disease and rheumatism. These chemicals and hormones also affect the immune status in a negative way. This inflammation can also affect the gums and a largely inactive immune system can result in periodontitis or infection of the gums. The inflammation can reduce blood flow to the gums and cause the disease to become acute. After smoking, obesity is the greatest risk factor for inflammatory periodontal tissue destruction.!


Add a comment. Registration required because trolls.