THE OBESITY EPIDEMIC (part 1): What is obesity and what are its consequences?

Obesity.   It seems that nearly every month we hear of yet another potential cause.  It’s due to viruses.  Your gut flora is responsible.  It’s your mother’s fault.  It’s in your genes.  You have a bad thyroid.  Etc., etc., etc.

But in truth, the obesity epidemic that is underway all over the world is probably NOT due to thyroid problems or even a drift in our gene pool toward higher numbers of obesity genes.  It wasn’t a significant problem 100 years ago.  Or even 50 years ago.  And there simply hasn’t been enough time for all of us to develop bad thyroids.  

And to say that obesity is an epidemic is no exaggeration.  It’s not even controversial.  We may argue about whether global warming is caused by human activity, we may debate evolution, and we may disagree about the use of embryonic stem cells.  But I don’t think anyone could disagree that obesity has increased dramatically within the last generation or two.

The data is unequivocal—the world has gotten fatter.  And the United States, ever out in front, is a world leader, along with Samoa, the Solomon Islands, Kuwait, Barbados, the Bahamas, and Micronesia (the fattest of all).
Here in the U.S., the percentage of obese individuals has increased from about 15% in 1976 to 35% in 2004.  

From this graph, you can see that about 70% of the adult population in the U.S. is either overweight or obese.  It is clear that obesity has been increasing for several decades, but surprisingly, the percentage of people who are overweight has not changed.  (I’d like to know more about this phenomenon, but I haven’t been able to find a discussion about it anywhere.  However, it seems to me that in order for people to be classified as obese, they must first have been overweight, so shouldn’t we see an increase in overweight along with—or prior to—the increase in obesity?  Maybe the answer is that large numbers of people who used to be merely overweight have now become obese, and by chance, just enough formerly normal-weight people have moved into the overweight category to keep the size of that group unchanged.)

And what has happened since 2004, the last year covered by the above graph?  According a 2012 report by the National Center for Health Statistics, there has been no significant change in the prevalence of overall obesity in the United States during recent years.  However, the incidence of obesity was previously higher in women than in men, and during the last decade, the men have caught up.

In order to be sure that we understand what we’re talking about, let’s define our terms.  What do “obese” and “overweight” really mean?

Human body fat percentage is commonly estimated by something called the “body mass index” (BMI), which is computed by dividing a person’s weight in kilograms by the square of their height in meters.  It is a far from perfect index—a body builder with lots of muscle may have a high BMI, for example—but it has the benefit of being very simple and straightforward.

Here are the definitions of “obese” and “overweight” as they relate to BMI:

In the United States a BMI of 18.5 to 25 represents “normal” body weight; a BMI between 25 and 29.9 is considered “overweight;” and a BMI above 30 is classified as “obese.”  Since the “above 30” category encompasses a large range of BMI values, it has been further divided into several rough sub-categories:  “severe obesity”  (a BMI of 35–40), “morbid obesity” (a BMI of 40-45 or 40-50), and “super obese” (a BMI greater than 45 or 50, depending on who you ask).  And since BMI is dependent on body type, these categories may change for different ethic groups.   In Asian populations, for example, a BMI of 18.5 to 22.9 is “normal,” and 23 to 27 is “overweight.”

Of course, the most accurate measure of “fatness” involves calculating the percentage of a person’s weight that is due to his or her fat cells.  This is a difficult calculation to make, but it can be done by weighing an individual both in and out of water to obtain a measure of that person’s “density.”  Density can then be related to body fat percentage by a formula first developed in 1953 and then refined in 1963.   

There are also other ways of estimating fatness, including X-ray absorptiometry, near-infrared interactance (using a beam of infrared light directed into the biceps), ultrasound, waist-to-hip ratio, and the “pinch test” (in which a fold of skin is measured by calipers).  But underwater weighing is thought to be the most accurate by far.

Since BMI is so widely used, it is logical to ask how well it is correlated with body fat percentage.  The answer is that it IS correlated—but the variation is substantial.  Here is a graph in which the BMI of 8,550 men is plotted on the horizontal axis and their actual body fat percentage is plotted on the vertical axis.

If each person’s BMI were perfectly correlated with his body fat percentage, this graph would be a straight line because their BMI and their percent body fat would always be the same number.  You can see, however, that the graph is actually shaped more like a jellybean, which means that many of the men tested had an actual body fat percentage that that was higher or lower than suggested by their BMI.  For example, you would expect those in the lower right quadrant to be overweight since they all had a BMI above 25.  But when their body fat was actually measured, it was found to be below 25% (sometimes WAY below).  These men were not fat at all—indeed, they were probably quite muscular.

But the fact that BMI mis-identifies some people as overweight is not the real problem.  The primary purpose of BMI, after all, is to identify people whose body fat percentage is higher than it should be for optimum health.  And just as BMI OVER-estimates fatness in people who are lean and muscular (those in the lower right quadrant), it UNDER-estimates fatness in people who have relatively little lean muscle mass (those in the upper left quadrant).  Therefore, it often fails to sound the alarm for people who actually ARE overweight, which suggests in turn that BMI is underestimating obesity as well.

Obesity is correlated with many diseases, including diabetes type 2, high blood pressure, cardiovascular disease, osteoarthritis, various cancers, sleep apnea, systemic inflammation . . . . the list goes on and on.   At least 50, by my count.  But even though the obese as a class have more health problems than the non-obese, they are not a uniform group.  A very recent (2013) study has shown that approximately 46% of people classified as obese were “metabolically healthy” and otherwise “fit”—and showed no higher percentage of mortality or cardiovascular disease than normal-weight people who were also metabolically healthy.  

As a result, the authors of this study advocate the classification of obese people as either “fit” or “not fit.”  At this point, it is unknown what the differences between these two groups might be, but researchers speculate that that they may actually have different types of fat (more on that in another blog, perhaps).  

It should also be mentioned that obesity is correlated with the so-called “survival paradox.”  First described in 1999, this is a phenomenon in which people with heart failure and a BMI between 30 and 35 (obese) had greater chances of survival than those of a normal weight.  

And a 2006 paper that reviewed 40 studies involving 250,152 patients found that those with a BMI lower than 20 (underweight) had an increased risk for total mortality and cardiovascular mortality, while those with a BMI of 25-29.9 (overweight) had the lowest risk among all the groups for both total mortality and cardiovascular mortality.  Remarkably, the obese patients (BMI 30-35) had NO increased risk for either total mortality or cardiovascular mortality.  And the severely obese (BMI greater than 35) did not have increased total mortality, although they did have greater cardiovascular mortality.

In spite of the results outlined in this paper, it is pretty obvious that the authors did not really believe that obesity has no health consequence—they postulated, correctly it seems, that the results may have something to do with the inaccuracies of the BMI calculation.  

And, yes, according to a 2013 paper, almost ALL research studies reporting the “obesity paradox” used BMI as their measure of obesity.  The authors went on to say that more recent indices of obesity, such as large waist circumference and high waist-to-hip ratio, ARE positively correlated with cardiovascular events and mortality.  

Another aspect of obesity that is of great concern is the increasing financial burden that it places on society.  As reported in 2009, medical expenses for obese individuals in the United States average about $1,429 each year—42% higher than the expenses incurred by those of normal weight.   Further, this same report estimated that the medical costs of obesity, which totaled $78.5 billion in 1998, had risen to $147 billion by 2008, a figure that represents 9.1% of all medical expenses in the U.S.

And that was in 2008.  I wonder what it is today.  

Next time we will begin a discussion of the reasons for obesity.

Useful references