TCS Daily

The BMI Problem

By Radley Balko - January 1, 2005 12:00 AM

Editor's note: The following remarks were delivered at the Risk: Regulation and Reality Conference by Radley Balko, policy analyst at the Cato Institute. The conference was co-hosted by Tech Central Station and was held on October 7, 2004 in Toronto, ON.

James Glassman: We are going to have to move on to our next speaker, but thank you very much, Tim, thank you, Sallie. [Applause]

Pleased to present, Radley Balko, a policy analyst for the Cato Institute. He specializes in the nanny culture, which I referred to in my opening remarks, and consumer choice issues, including alcohol and tobacco control, drug prohibition, obesity, and so forth. He's a columnist for Fox News and he's been published in Time Magazine, The Washington Post, many other publications. Radley will speak today on getting off the scale, weighing in on obesity. Radley Balko. [Applause]

Radley Balko: Thank you. I'd like to thank the Atlantic Institute for Market Studies, Tech Central and The Toronto Insurance Conference for inviting me to speak. I'd like to say that to prepare for my speech on obesity and fat and poor diet, I yesterday sampled a Canadian specialty, I believe it's poutine, which is about the most eloquent name for a bar food I think I've ever heard. Earlier we talked a little bit with Dr. Kunze about caffeine. He mentioned, skeptically, about how caffeine might be the next war. Actually, it's on the way. I ran up to my room really quickly over the break and found a few studies that had been released just in the last year on caffeine. There's a term in debate called reductio ad absurdum, which means you take the reasoning behind your opponent's argument and you apply it in its most extreme form and it sort of exposes the fallacy of that reason.

About ten years ago when the tobacco suits, lawsuits, started coming up, and they were justified under this notion of public health, a lot of people wrote, well, what's next? Are we going to start suing fast food companies because they are feeding us fattening food? It was presented as a sort of satirical, absurd argument that this should defeat the whole notion of public health and eroding personal responsibility. Well, as we all know, that's happened now.

So I'd submit that we sort of live now in a post-reductio society, where there is really nothing about the nanny stake that is so absurd that the nanny status aren't willing to try it, and caffeine is a great example. There is a study that just came out yesterday, actually two days ago, basically saying that caffeine has all the addictive properties of nicotine. There are lots of problems with the study. I won't get into them here. But just to give you an example, I'm also the Center for Science of the Public Interest, which is a U.S. nanny state organization, and puts out a lot of really alarmist hysterical studies, and on its website notes that according to studies, caffeinated coffee drinkers, actually, I'm sorry, decaffeinated coffee drinkers take more vitamins, exercise more faithfully, eat more cruciferous vegetables like broccoli, and are more likely to wear their seatbelts than caffeinated coffee drinkers. So, bring on the war on big java. It's probably just down the road.

Because I work in Washington, my talk today is going to focus mostly on the obesity crisis in America, but I hope that if I take a critical look at what's going on in America, it'll inspire you to take a more skeptical look at what's going on here in Canada.

Well, that said, let's jump right in and let me see if I can't give you a reason to take a second trip to the buffet line without guilt. The first question is, are we really that fat? I'm originally from Indiana, from the Midwest, and every time I go home the anecdotal evidence seems pretty cut and dry that we are. It's hard to go to a Wal-Mart or a mall or a county fair without noticing people do seem to be quite a bit bigger. But, at the same time, it's also hard not to notice the proliferation of health clubs, joggers and bike paths. In fact, there actually are 40 percent more health clubs in 2001 than there were in 1992, and health club memberships are actually up 60 percent over the same period. So, what exactly is going on?

Let me show you a map that we see pretty frequently in the United States. This is what the obesity situation allegedly looked like in 1988 and the anti-fat warriors trot this map out, and the one I'm going to show you in a second, to show just how pervasive the obesity crisis is. Here's the "after" map. So it definitely seems like America is getting thicker, but here are a few things you want to keep in mind. A lot of that thickening came about not because Americans actually got fatter and put on weight, but because in 1997 the American government actually redefined what it means to be fat. Thanks to the NIH, the National Institute for Health in America, one night, magical night in 1997, 29 million Americans went to bed with average figures and actually woke up overweight.

So, before that decision, we have about 68 million Americans who were overweight or obese, they went to bed and woke up, and America had 97 million fat people. And, of course, nobody gained a pound.

Here's a quote from The Washington Post on the NIH decision.

I think this next quote is actually even better, not only did these people wake up fatter, they woke up much, much less healthy, without ever gaining a pound. All of a sudden, 12 hours later, they are at an increased risk of diabetes, elevated blood cholesterol, heart disease, stroke and high blood pressure.

All these statistics are taken from the BMI, which is called the Body Mass Index, which is the way the government measures overweight. This is the formula that they use. There are lots of problems with the formula, which I'm going to get into here. As you can see, the only criteria that is used in calculating BMI is weight and height. So, it doesn't allow for any discrepancies between sex, race, age or ethnicity. It doesn't distinguish between weight and fitness, or the fact that you can be overweight and very active, and exercise regularly. Most importantly, it doesn't distinguish fat and muscle tissue.

This last distinction is pretty important because if you have a room full of modestly fit people, or slightly overweight people, who then take the NIH's recommendations and go to a health club and start lifting weights three or four times a week, what's going to happen is you are going to start replacing body fat with more dense muscle tissue, which means your actual weight is going to go up. Well, when the NIH then takes its statistics, your BMI is going to go up, too. So, according to the government, you're following the government's advice, you're doing exactly what you're supposed to do, but you're getting heavier, and you're contributing to the obesity epidemic.

So, to demonstrate just sort of how absurd this is, I'm going to give you a few examples of people most of us probably know, celebrities, and show you how the BMI applies to them.

Matt LeBlanc, Joey, from Friends, and I think he has his own show now, he's 5'11, 218 pounds. That gives him a BMI of 30, which, according to the U.S. makes him obese.

Johnny Depp, People Magazine's Sexiest Man Alive, 5'7", 190 pounds, he's doing a little better, but, he's still overweight.

Yao Ming, looks like of lanky, tall, skinny, 7'6", 310 pounds, overweight.

Tom Cruise, obese.

And, finally, the President of the United States, he is, according to his last physical, he was 6'0", 194 pounds, makes him overweight. Just for a reference point, President Bush apparently jogs three miles three times per week, trains on an elliptical machine more than one hour each week and lifts weights twice per week. His own doctor at his last physical declared him in superior fitness for a man his age, but, according to his own government, he needs to take drastic steps to get his weight under control.

Here's a better way to illustrate the BMI problem. This comes from the Center for Disease Control's own website. We have our in fit muscle man, 6'3", 240 pounds. We have his couch potato counterpart, 6'3", 240 pounds. Both of these people, according to the U.S. government, are obese and there's really no difference between them when calculating how healthy the country is in terms of weight.

Actually, let me elaborate just a little bit. The celebrity BMI productions are fun, but there's an important point here too. When a reporter, a television anchor, a government official starts citing these obscene percent obese statistics people notice. In the last few years, these numbers have been cited thousands of times and invariably they are called by calls for action and almost always government action. In a special last December, ABC news anchor, Peter Jennings, likened fatty foods to tobacco and pled, "when will government get involved?" Activists and politicians have cited these numbers to call for restrictions on the marketing of high calorie foods, fat taxes on high calorie foods and various other measures aimed at restricting consumer choice. And, actually, here in Canada, the liberal party, last year, proposed an 8 percent tax on restaurant meals under $4.00. Apparently trying to go after the fast food industry. But when people like Tom Cruise, George W. Bush and over half of the National Basketball Association are counted among the overweight or obese, and are then healthy by government standards, what exactly do these numbers mean? I would suggest they mean very little.

This is a quote from a Rockefeller University professor, Jeffery Friedman, who he took an extensive look at the data and determined that what's happening is the very, very obese are getting more, very obese, if you can follow that, whereas the majority of us have put on five to seven pounds over the last 20 years. So what's happening is the mean weight has gone up quite a bit, but we're not really all that less healthy than we were before. In fact, there's lots of evidence that we are more healthy, which I'll get to in a minute.

So, is fat really killing us? The conventional wisdom is that it is, and that if we don't act now we'll be asked to pay a fortune later in the added healthcare costs and illnesses associated with obesity.

Since I'm speaking to an insurance group, one idea that I'm really trying to push in the U.S. is to free health insurers to start offering a carrot and stick approach to fitness and lifestyle habits. What I'd like to see is, and there's actually, Tommy Thompson, the Secretary of Health and Human Services of the U.S., supports this idea, the restrictions are at the state level, but there are lots of restrictions in the U.S. where if you are a health insurance company, and you're providing a group plan to a corporation, everyone has to pay the same premium, you can't vary rates according to risk, like car insurance and life insurance companies do.

So what ends up happening is you have overweight people or less healthy people subsidizing, I'm sorry, it's the other way around, you find fit people, people who are living healthy lifestyles, subsidizing people who aren't fit. What I'd like to see is the insurance companies freed to start experimenting, to unleash the power of the market to find the best lifestyle. And, currently, that isn't happening, and it's something that I'd like to see.

So, let's look at how dangerous fat really is, and you see the editorial cartoon here, and it may be a little amusing to you, or maybe you think it's a little over the top to suggest that the government would compare obesity to terrorism. But, in fact, they have. The U.S. Surgeon General, speaking last June at an obesity conference, said that excess weight is every bit as threatening to the United States as terrorism.

Here's another popular notion making its way through the American media -- that obesity will soon overtake smoking as the number one cause of preventable death.

Here's more hyperbole. This is from a study, a very widely sided studied saying that 400,000 Americans will die each year due to obesity. I'll get to that number in a minute as well.

So let's look at the data a little bit. Here's a quick snapshot of the alleged obesity epidemic. Let's, for the sake of argument, let's put our quarrels with BMI aside and go ahead and accept how the government defines obesity. These are the number of the states who 10 percent of their population or more was defined as obese. You see it went from 8 in 1985 to 33 in 1990, to all 50 states in 2001.

Keep that in mind when you look at this chart. These are life expectancies over those exact same periods. As you can see, life expectancy increased pretty much across the board, even as we got collectively fatter. If you look, break it down by race, actually, this is a pretty important quote also, Paul Ernsberger, Professor of Medicine, says that, notes that, if obesity does kill 400,000 people a year, we should see huge decreases in life expectancy and we've just shown, life expectancy has increased. In fact, in the illnesses most associated with obesity, we've actually seen the biggest drop in mortality rates.

Here's another example, if you compare black women to white women, since 1988 black women have actually, by, I think that's more than 50 percent, increased their obesity rate as compared to white women, but they've actually added to their life expectancy by a year. Similar with black men and white men, black men got thicker over that period, but they added quite a bit more to their lifespan.

Let's look at obesity and cancer. Here are a few more scare quotes I pulled from a Lexis search. Actually, I'm sorry, I skipped over one part. The reason why I broke it down by race, which seems to be a fascination in America, is because the media has made a big deal about race and they pointed out that obesity race have risen higher in African American communities. A couple of quotes, Good Morning America parenting editor, Ann Pleshette Murphy said, at a recent obesity conference, "the whole issue of weight to me is inherent racism." And, Sally Squires of The Washington Post actually compared the obesity crisis to a "genocide" on the black race. But what they don't take account, in fact, is that African Americans are actually adding to their life expectancy at a bigger rate than white Americans, and, if you look at mortality rates and epidemiological tables, the ideal weight for black Americans is actually what the government defines as "overweight." So, in other words, it's actually healthier for African Americans to have excess weight than it is for Caucasian Americans.

So let's look at cancer. The anti-fat people have always hinted at a link between fat and cancer, and they got their talking points when The New England Journal of Medicine published a study called, "Overweight, Obesity and Mortality for Cancer" in 2003. The authors of that study said we could prevent 90,000 cancer deaths each year if Americans would start losing weight and maintain a BMI in the "ideal" zone. But, as the author, Paul Campos, points out in his book, The Obesity Myth, the study's own data doesn't match its conclusions. The very obese did increase the risk of cancer, but not by much, generally by a rate between 1 in every 1,000 people to 1 in every 2,000 people. But the study's authors and the media overlooked another very interesting finding that their data revealed, and that's that overweight people, the people the government defines as overweight, actually had a decreased risk of cancer over people in the "ideal" weight group. But that wasn't reported. The study's authors drew a line between overweight and the very obese, lumped the numbers together, and averaged them out. So actually being a little overweight by government standards in terms of cancer prevention is better than being ideal.

Here, according the NCHS is just sort of a quick snapshot at overweight or obesity trends since 1975-1976.

And then here, this next study, or this next chart, looks at changes in cancer mortality rates in the ten cancers that are most associated with obesity. As you can see, in nine of the ten cancers most associated with obesity since 1992, mortality rates have dropped, and they've dropped pretty significantly. There's a little bit of discrepancy between the data, but the reason I did that is because if we started getting bigger in the late '70s, early 1980s, you would think that we'd start to see the effects of that by the early to mid 1990s, in other words it should start producing more, at least the front end of more cancer cases, because they can see the cancer rates have gone down, mortality rates, incidence rates have also gone down.

Now, this isn't to suggest that being obese is healthy, it's not to suggest that these bars wouldn't extend much, much further to the left if people were at an ideal weight. I'm not in any way saying that obesity aids in preventing cancer. What I am saying is that these people are predicting this health catastrophe that we're going to see because of obesity, and if that's true, we should at least be seeing the front end of that. The whole public health argument is that we need to start fighting obesity now or we're going to have this huge healthcare crisis down the road. And, at least in terms of cancer and in terms of mortality rates and life expectancy, it just hasn't happened.

Just a few more stats on cancer. New incidents of cancer have fallen an average of a half percent a year since 1993. Deaths have fallen 1.1 percent since 1993. Overall cancer death rate has dropped 10 percent between 1991 and 2001. The cancer survivor rate has jumped 43 percent since the mid '70s.

Let's look quickly at heart disease, frequently linked to obesity. Here are some more statistics. Both of these statistics assert that heart disease is on the rise as a result of obesity. The problem is, heart disease isn't on the rise at all, in fact, it's declining by pretty dramatic numbers. Total cardiovascular disease is down 16 percent since 1990. Coronary heart disease is down by an amazing 25 percent. Stroke is down by 10 percent. This is across the board. Every single state in the U.S. had a decrease in overall rates of cardiovascular disease by no less than 7.5 percent.

So every state in the country saw a significant decrease in overall cases. Every state but the District of Columbia saw a decrease in coronary heart disease by at least 8.5 percent over the same period. And 47 states, 94 percent, saw an overall decrease in stroke cases over that period. Decrease in heart disease is rampant. If we've been getting bigger over the same period, again, we shouldn't be seeing numbers like this.

This is, again, a quote from Campos, who says that the link between obesity and high blood pressure and high rates of bad cholesterol is probably more a result of inactivity than it is actual weight. This is an important distinction that research just hasn't addressed yet. It's only been in the last three to five years that nutrition researchers have actually separated fitness from fatness. And what they're finding is pretty much every study done in the last few years that has examined this has found that activity levels are far, far more important to health than weight. When people lead sedentary lifestyles, they are more prone to disease. I'll show you a graph here in a minute that illustrates that point pretty dramatically.

About diabetes, here are the rates of diabetes over the last 25 years. As you can see, it's a pretty dramatic increase, and it corresponds pretty nicely to the obesity curves we looked at earlier. It looks like there's probably a pretty strong correlation there, but there are a few explanations for it. First, we should acknowledge that Americans are not eating nearly as well as they used to be. Diabetes, of course, is an insulin disease, it's related to diet more than anything else, and, so, I would, Americans are eating much more sugar, much more starch than they were 20-25 years ago. So, certainly, we have to acknowledge that that probably does contribute somewhat to the rise in diabetes cases. But there are other factors that play here too.

An aging population. The elderly are more susceptible to diabetes, and America is definitely getting older. Forty percent of current diabetes cases are of people age 65 or older, and, we should also point out that that's the one demographic group in America that's least likely to be obese.

We have definition problems here too. In 1997, just as it did with obesity, the CDC redefined what it means to be diabetic, changing the diagnostic cutoff from a fasting blood sugar level of 140 to 126. So, once again, hundreds of thousands of Americans became diabetic overnight, without any significant change in their condition.

Our methodology changes, in 1997, it seems to be a magical year for this, NIH changed the way it sampled the population to get its diabetes numbers in phone surveys. This new sampling, and the CDC actually recognizes this on its website, resulted in an aberrant 20 percent increase in diabetes cases from 1996 to 1997. That methodology change is rarely mentioned when media outlets compare recent figure with pre-1997 figures. So you want to keep that in mind as well.

And, finally, awareness. Coverage of obesity has made people much more aware of diabetes, it's made doctors much more likely to diagnose it. And of course, this is a great, I mean, this is a good thing. We want people to know and we want diabetes to be treated because it's a very treatable condition. But it also means the dramatic rise we saw on the graph earlier probably isn't so dramatic after all.

There is also the problem that lots of doctors tell their patients in an effort to kind of scare them, that they are borderline diabetics. This isn't necessarily an accepted medical condition, but, what it might do is create enough confusion for people to report to phone surveyors that they are diabetics because their doctor mentioned it to them.

What about children and diabetes? We hear a lot about the fact that so many kids are being diagnosed with diabetes because they are overweight now that what used to be called adult onset diabetes is now called Type II. That's true, there are a lot more cases of childhood diabetes than there were before. The problem is the solutions that people are recommending. The state of Arkansas recently enacted a program, they are going to start weighing kids in public schools and sending fitness report cards home to their parents. Other schools are going to start monitoring kids diets, their lifestyles, they are also going to weigh them. The problem with this is that we have a huge problem with eating disorders in the country. If you look, if you compare, and we're talking about risk here, if you compare the number of cases of anorexia or bulimia to the number of cases of Type II diabetes, asking kids to step on a scale in front of their peers once a week, or even not in front of their peers, just calling attention to weight and body image, may not be the best solution to the problem.

I'd also point out that a lot of the rise in the childhood diabetes cases can be explained by the things we talked about in the adult cases, changes in definition, changes in awareness, changes in diet, the aging probably doesn't apply here, but most of the other factors can be applied to the childhood cases as well.

Let's look at mortality rates. And here's where I'd like to get into this 300,000, 400,000 number. The 300,000 number comes from a study that was published in the early '90s in The New England Journal of Medicine, and there was a follow-up study published just last year that raised the number to 400,000. Well, here are some problems with the number. First of all, the study that was published in The New England Journal of Medicine was actually a meta study, which is a survey of other studies.

A lot of those other studies actually went back pretty far into the past, in fact, a few of the studies dated back as far as the late 1940s. Well, when you're looking at BMIs, when you're taking height and weight and calculating BMIs from mortality stats that date back to the '40s, you're not accounting for advances in medical technology. So a lot of those deaths could probably be prevented on the way to that 400,000 number.

About 17 percent of the deaths in the study weren't from obese people, but overweight people. That's a problem because the study's own data show that there is no correlation between overweight and premature death. In fact, lots of epidemiological studies show that being overweight, as opposed to obese, keep in mind, actually has a protective effect against premature death.

Which brings me to my next point, the study's conclusion differs with this data, and, again, we go back to Professor Paul Campos, who points out that, according to the study, people with a BMI of 20, which, according to the government is an ideal, run the same risk of premature death as people with a BMI of 30, which the government considers obese. It gets even weirder. Both of those people actually have a higher risk of premature death than someone right in the middle with a BMI of 25, which is overweight by government standards, which basically means this is all about line drawing. They laid out the data and they drew the lines in places where they could separate the data that would support conclusions they had before they went into the study.

This also leads me to the fourth point, which is absolutely the most egregious error in the study. Basically the study assumes that all premature death among obese people is caused by obesity, and they admit this. It's the authors of the study. So basically what this means is that if an obese man goes out to get the newspaper and is struck by lightening, his death is a death attributable to obesity. It's pretty remarkable. It doesn't concede car accidents, household accidents, diseases and conditions that are wholly unrelated to excess weight. Basically anyone who is obese and died before the average life expectancy, was considered to have died because they are obese. It's a complete falsehood and it really blew up the numbers, blew the numbers out of proportion.

Given all these shortcomings with the author's methodology, it's probably safe to say that the 400,000 number is a gross overestimation, but it's incredibly pervasive. I did a Lexis search just before coming out here and the studies released last year, and there were already 1,077 media mentions of obesity and the number 400,000 in the last year.

And, again, we go back to the Ernsberger quote about the ridiculousness of the number. Again, particularly if the number went from 300,000 to 400,000 in the last year, we should see this, we should see it in mortality rates from diseases connected to obesity, we should see it in mortality rates overall, and, again, the numbers just don't pan out.

I guess we are a little short on time, I'm going to kind of cut the last part short just a little bit, but, basically what the last part deals with is the connection between fitness and health and fatness and health. And, again, as I said, as new data is emerging and as research is starting to separate the two, they're finding that people who are active tend to have lower mortality rates, whereas weight plays a far less important factor. If you look at this graph, this is, one represents just a normal mortality rate. A normal weight, physically fit person, is a one. You can see a normal weight, unfit person is twice as likely to die prematurely as an obese person who exercises or is active three to four times a week. I think this is pretty compelling data. And, so, when the fat warriors and the nanny status are out there, they're trying to limit access to food, they're trying to blame food manufacturers and the food industry for obesity, they're trying to restrict consumer choice, you can see what's happening is we are going about this exactly the wrong way. What also happens is, when we encourage people who are overweight to lose weight, studies have shown that people who try to diet, not only do diets fail, but people end up putting on more weight than they did before when they fail.

So, basically what we're doing is, people are starting to exercise, they're getting healthier, but they're not seeing the results, so they get discouraged and they stop exercising and so they put on more weight. So, basically, we see, as we've seen in a lot of other speakers before, this effect of unintended consequences, this massive government invention is actually causing people to become less healthy, it's actually causing them to put on weight, and it's a typical, complete, sort of overreaction that's not based on evidence and is based more on hysteria and scare tactics.

I think I'll cut it off right there and see if we have questions. [Applause]

James Glassman: Thank you, Radley, for that excellent presentation. I know we are over, and, but does anyone have a question they want to? Yes?

Female Audience Member: One of the questions I have is there is a lot of science out there, what's labeled as science, on different studies that examine health problems, obesity, consumption of certain foods and beverages. For instance, you have the University of North Carolina study that says, childhood obesity is actually not, is not related due to increase of caloric intake, the caloric intake remains steady while physical activity diminished. However, then you just have a recent study that was published in the Journal of the American Medical Association that talks about how consumption of sugar-sweetened drinks automatically causes diabetes. How do you sort through this? How do you sift through the information that's thrown at people, especially communicated by the media?

Radley Balko: That's the problem, is you get these studies, and you get studies saying the other thing, and people don't know what to believe. I mean, the only thing I can tell you is to really look closely at the methodology of the studies and try to figure out, or find, seek out, always seek out contrary opinions when these studies come out. The media almost always takes them at face value and provides no skepticism whatsoever. The study you mentioned on soda, there have been lots of debunkings of that study. Steve Malloy [ph.], I think, did one, the proprietor of the Junk Science website. So, just, I would look for contrary opinions and then sort of make up your own mind on it. The problem is, the people who want to restrict our choices and restrict our behavior are really, really well-funded, and, so, the number of studies funded, and, keep in mind also, scary headlines generate publicity, publicity makes the underwriters of these studies happy.

So there is this incentive for researchers to come up with these outrageous statistics, it's not saying that everyone is dishonest, it's just saying look at the incentive system out there as well. So, just, be skeptical and I would say look for contrary opinions. That's about the only thing you can do.

James Glassman: I think that's a good suggestion. Almost all, certainly all of the risks we discussed today have been exaggerated and the media of which I guess I'm a member have an incentive to exaggerate, even, forget about an ideological bias, they are in the business of scaring people.

Time for lunch, which is very apt. Thank you very much, Radley. [Applause]


TCS Daily Archives