About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Still Semaphoring, Even From the Bottom of the Swimming Pool | Main | Arzoxifene: Not the Road to Big Profits? »

August 21, 2009

Obesity Shows Up in the Death Rate? Right?

Email This Entry

Posted by Derek

Here's an interesting post at Chemiotics (a new addition to the blogroll): Something is Wrong With the Model

. . . The Center for Disease Control released new data for 2007 (based on 90% of all USA death certificiates) showing that mortality rates dropped again (by over 2%) to 760/100,000 population. It’s been dropping for the past 8 years, and viewed longer term is half of what it was 60 years ago. Interestingly death rates from heart disease dropped a staggering 5% and even cancer dropped 2%.

But the populace is fat and getting fatter. . .

The heart disease death rate is particularly interesting. One explanation, which we can't rule out, is that these improvements are due to other factors (which the post goes on to elaborate), and that the improvement would be even more impressive if everyone weren't packing on the pounds. Another possibility is that excess weight, up to a point, may not have as big an effect on mortality and morbidity as we've been thinking it does.

That's a real possibility, and it's been looked at in the context of these sorts of public heath figures. The current use of BMI, at the very least, doesn't seem to be that useful in that regard. Only the high end of the BMI envelope (>30) seems to show much of a meaningful health effect. Of course, there are other costs to being obese, but (up to a point) bad health may not be one of the major ones. As for what this means to the current health care proposals, you can go here for the arguing.

Comments (24) + TrackBacks (0) | Category: Diabetes and Obesity | Regulatory Affairs


1. Esteban on August 21, 2009 8:49 AM writes...

My money is on statins -- the anti-inflammatory effect, more so than the lipid lowering effect.

Permalink to Comment

2. Sili on August 21, 2009 9:43 AM writes...

Stupid question, but couldn't it be that the fatties (yes, I'm rude) just aren't old enough yet?

We expect them to die younger, yes, but not in their teens.

That is of course assuming the obesity isn't distributed evenly across all ages.

Permalink to Comment

3. Tok on August 21, 2009 10:29 AM writes...

My money's on BMI being useless. I started working out awhile ago, lost fat but gained a lot of muscle and worked myself into the "obese" category according to BMI. I was in the best shape of my life and I would have been classified as obese. Obviously that's not the general case for BMI going up, but it shows how flawed it is.

Permalink to Comment

4. qetzal on August 21, 2009 10:54 AM writes...

Sure BMI is flawed, but that doesn't mean it's useless. For the vast majority of people, BMI is a very good proxy for percentage of body fat. And since BMI is easily derived from simple height and weight values, it's a very useful initial test to determine whether someone may have excess body fat.

Of course it won't apply to everyone, but no test can. Besides, it's easy to identify the people who have high BMI without high percentage body fat. If Lou Ferrigno walks into his doc's office with a high BMI, I don't think the doc's gonna tell him he's too fat.

They key question here is whether high percentage body fat itself is really as much of a risk factor as it's been purported to be. Only if the answer is no would I agree that BMI is relatively useless.

Permalink to Comment

5. hell to the chief on August 21, 2009 11:21 AM writes...

BMI is a useful measure, but like many other things can't be looked at in isolation. I think waist circumference needs to be part of the equation too.

Permalink to Comment

6. Chrispy on August 21, 2009 2:11 PM writes...

From a public health expense point of view, I wouldn't be surprised if many activities that shorten people's lives actually make them cheaper to support overall. The smoker that dies at 70 is a lot cheaper than the health nut that draws social security until he's 90. Fat people cost more due to diabetes, so we should required that fat people smoke to even things out. (kidding, really)

Permalink to Comment

7. anon on August 21, 2009 2:34 PM writes...

The BMI was developed as a data transform so the height/weight of a population could be analyzed as a normal distribution. It has never been validated for any other use.

Permalink to Comment

8. Esteban on August 21, 2009 4:07 PM writes...

I seem to recall reading somewhere that, as far as easy-to-measure surrogates for obesity go, sex-specific waist-to-hip circumference ratio does a better job than BMI at predicting outcomes. Can't recall where I saw that however. WHR would certainly give fairer treatment to the Lou Ferrigno's of the world.

Permalink to Comment

9. retread on August 21, 2009 4:48 PM writes...

Derek: thanks for the notice. There are a lot of problems with the BMI, but in its favor, measuring height and weight isn't rocket science, and is quite easy to do. Skin fold thickness, waist hip ratio are probably better, but less available. The BMI skeptics might be interested in the following notes I took on an excellent article in Nature this year.

[ Nature vol. 459 pp. 340 - 342 '09 ] There really isn't an epidemic of obesity. Variation in body weight has a Gaussian distribution (e.g. a continuous distribution) and if you define a BMI > 25 as overweight then a small shift in the mean value of weight in the population leads to a disproportionate number of people exceeding the threshold. A 33% increase in the incidence of obesity in the USA from the 80's to the 90's means an average weight gain of 6.6 - 12 pounds in the population as a whole (that's still a lot -- 2 to 4 trillion pounds for the US population). So the secular trend toward obesity is less profound than is generally appreciated.

The variation in weight over a year is minimal, despite the intake of on average 1,000,000 calories -- so weight maintenance is largely unconscious. This is much more than monitoring food intake could insure, so weight maintenance is largely unconscious.

Based on twin studies (identical vs. non-identical) the genetic factor in obesity is variously reported as 70 - 80% -- much more than diabetes, heart disease and cancer. The only trait with consistently higher heritability than obesity is height.

Only 5 - 10% of morbid obesity (BMI > 40) is due to defects in known genes (leptin, leptin receptor, POMC, BDNF, MC4).

When fat people lose weight, they use less energy than lean individuals of that weight who haven't ever been fat. To maintain their reduced weight they must consume fewer calories than their initially lean counterparts. This is one reason that diets don't stick.

Most people who have had bariatric surgery remain fat (BMI > 30) despite a marked reduction in food intake. True of my neice sad to say

Permalink to Comment

10. Morten G on August 22, 2009 8:25 AM writes...

There is a very nice linear relationship between waist circumference (even easier to measure than WHR) and total health care costs.
For men expenses start increasing from 94cm and for women from 80cm. 2% cost increase per cm for men and 1.25% increase for women. White people though.

Terribly journal it was in:

Permalink to Comment

11. Lucifer on August 22, 2009 2:38 PM writes...

Morten G,

Correlation = Causation?

Permalink to Comment

12. pharma guy on August 22, 2009 2:42 PM writes...

Estaban is probably dead in in the effect of statins. At the recent ADA, type 2 diabetes was finally shown to be a bariatric disease for the vast majority, not a glycemic disease.

The tempering metabolic inflammation of statins and PPARs is very important. KO the lymphocytic PPARg receptor and lose a bunch of the TZD effect.

Also, how much effect is due to better CVD treatments? Take the current population and kick it back before current Tx and watch. A great but underapprectiated service from Pharma

Permalink to Comment

13. Anonymous on August 22, 2009 2:46 PM writes...

To get back to Derreks original question:
it might be flawed to equal health with morbidity rates. It is very well possible that the genral health declines, but the morbidity rate drops. It just means more unhealthy years are added....

With the "obesity pandemic" seems to come (mostly in the western world) an increase in diabetics (type 2) and problems with knees wearing out and general mobility issues.
A diabetic can live as long as a healthy person, proficed he takes his medicines. Complications lead to loss of limbs, and possibly loss of vision, but not necesairily death.
Loss of mobility can be partly covered by (electrical) wheelchairs, partially by social support, and does not lead to a shorter life.

In short: due to advances in medicine, greater wealth and more social support our ability to keep unhealthy people alive progresses up till now faster then the added mortality due to obesity.
(And dont forget we dropped some unhealthy habits too, the last decades. Air polution is way down since the sixties, smoking is down, lead is removed from gasoline, heavy working is much reduced, workplaces are much safer, traffic is much safer - and you probably can add a few I know forget).

Permalink to Comment

14. Lucifer on August 22, 2009 3:29 PM writes...


Maybe you should read a bit about the causes of excess mortality in people with Type 2 DM.

The majority of this excess mortality is related to an increased risk of MI, something that current hypoglycemic drugs have not been shown to reduce. Also note that people with insulin resistance, but without clinical type 2 DM, have almost the same increase in risk of dying from MI.

Permalink to Comment

15. Retread on August 22, 2009 4:41 PM writes...

Apropos to Lucifer

There have been a bunch of intriguing suggestions to explain the data -- appearing here, on my blog, and on McArdle's.

One of most interesting (to me) was the decline in pollutants. The problem with this sort of thinking, is that you have to accept the bad along with good. Just plot the drop in ambient lead levels over the past 30 – 40 years against time. Now plot the drop in College Board scores the same way (before the Board normed them up so they wouldn’t look so bad — this really happened). The two curves look the same.

Correlation is not causation

Permalink to Comment

16. Ed Mooring on August 23, 2009 2:54 AM writes...


I'm very afraid this is one of the few places on the planet where the audience can properly appreciate your subtle sense of humor.

Permalink to Comment

17. Morten G on August 23, 2009 4:59 AM writes...


No correlation does not automatically mean causation. Usually though it is polite to suggest an alternate hypothesis. What did you get out of reading the article?
Derek's post was questioning whether the increases in adipose tissue was so bad when mortality was dropping so much. My argument is that there is a very strong correlation between being a fatty and having poor health. Poor health is usually associated with poor quality of life. And to me that is more important than what exact date I croak. And 94cm waist does leave room for a fair level of comfort. No washboard required.

Disclosure: my waist is 83cm, my BMI is 22, and I have no co-morbidity.

Permalink to Comment

18. Matt on August 25, 2009 6:13 PM writes...


There is no time-independent component of College Board scores, no time-independent test creators who can dispassionately make an exactly equivalent test 30-40 years apart. The norming does exactly what the test is supposed to do, compare applicants with their peers. Not their predecessors or successors outside a narrow (5 years max) time frame, or some universal measure of intelligence.

On the other hand, and perhaps this was what your joke was slicing, there is data showing similar trends in atmospheric lead levels in cities vs. crime.

Permalink to Comment

19. Duh on August 26, 2009 11:16 PM writes...

It is well established that within birth cohorts and cross-culturally BMI shows a J-shaped relation to all-causes mortality with a nadir ~24-26 in both sexes. The decline in population-wide mortality is due to independent factors. It would certainly be a greater decrease if BMI was not steadily increasing.

Permalink to Comment

20. pharma guy on August 29, 2009 12:47 PM writes...

Duh mid is dead on.
If I remember correctly, there is a time dependent part to the J curve.

If you can make it to your mid-seventies with a high BMI, the high BMI no longer is associated with higher morbundity. Those who are sensitive to the increased weight are selected out earlier?

There should be little selection for this trait since women are post-menopausal or you're lucky like Hugh Hefner or William O Douglas

Permalink to Comment

21. pharma guy on August 29, 2009 12:47 PM writes...

Duh mid is dead on.
If I remember correctly, there is a time dependent part to the J curve.

If you can make it to your mid-seventies with a high BMI, the high BMI no longer is associated with higher morbundity. Those who are sensitive to the increased weight are selected out earlier?

There should be little selection for this trait since women are post-menopausal or you're lucky like Hugh Hefner or William O Douglas

Permalink to Comment

22. Retread on August 31, 2009 1:19 PM writes...


"The decline in population-wide mortality is due to independent factors. It would certainly be a greater decrease if BMI was not steadily increasing."

The model is also quite certain about the effects of obesity. You may be correct in what you say, but I don't think you can be certain.

Permalink to Comment

23. Paul Norman on August 31, 2009 10:27 PM writes...

Just want to comment that there's a missing end italics tag in the entry.

Permalink to Comment

24. e27 led light bulb on January 12, 2013 12:48 AM writes...

Certainly compare the foreign car diagnostic the domestic also was not lazy.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry