Junkfood Science: JFS special report: Obesity statisticulation — When will people get it?

August 03, 2008

JFS special report: Obesity statisticulation — When will people get it?

The latest medical study predicts everyone will soon be obese. It was so unimaginable that anyone would take this study seriously, that I wasn’t even going to comment on it. Even G4TV laughed at the absurdity of researchers predicting that 100% of us would be obese by 2048.

As Ty Colfax piercingly wrote:

In today's edition of doctors can be stupid too, we present an article published in the July issue of Obesity, stating that 86% of Americans could be overweight or obese by 2030. And if you thought that was bad, they expect a 100% obesity rate by 2048... And to that we say, WTF? 100% obesity rate? Are these people serious? Don't they understand arithmetic vs. complex graphing? ... If their graph continued without logical limits, which they seem to have ignored, perhaps 220% of us will be obese by 2061.

It's ridiculous to think that just because someone goes to school for a long time they can publish this kind of b.s. in a medical journal (Obesity) obviously designed to inspire panic and deliver "news" on this epidemic that's fake as it is.

But leave it to mainstream medical journalists [cough] to report it as news. ABC News covered the story, ending with claims that this study is a good wake up call and that “obesity experts agree that certain measures, taken now, can head off major problems.” And who were those experts quoted? Dr. David Katz, director of the Integrative Medicine program at Yale [see here, here, here and here], and Dr. Neal Barnard, head of the vegan organization, Physicians Committee for Responsible Medicine, which had even sent out a press release about the study this week. As ABC reported:

"We are terribly, ominously off-course," says Dr. David Katz, co-founder of the Yale University Prevention Research Center. "To close the gap, we need to fix everything that's broken — from neighborhoods without sidewalks, to the high price of produce, to food marketing to children, to misleading health claims on food packages, to school days devoid of physical activity and school cafeterias devoid of healthful offerings. The list goes on and on."

Others cite individual responsibility for diet and lifestyle habits. Dr. Neal Barnard, founder and president of the Physicians Committee for Responsible Medicine (PCRM) and a staunch supporter of a vegetarian diet, says dietary modification could be a crucial step in solving the problem. "U.S. eating habits are nowhere near where they should be," he says. "The average American eats 50 pounds more meat and 20 pounds more cheese per year, compared to the 1960s... I would strongly encourage Americans to adopt more vegetarian meals. [According to the latest data from the USDA Center for Nutrition Policy and Promotion, and the total food supply of meats, poultry and fish produced in the United States was 18.1% of available food (579 kcal/day) in 1970 and went down to 13.4% (522.6 kcal/day) in 2004. An dairy products also went down, from 11.3% to 8.6%.]

This study in the news, published in the journal of the Obesity Society, was led by Dr. Youfa Wang, Ph.D., M.D., with the Center for Human Nutrition at Johns Hopkins Bloomberg School of Public Health. This is the world’s largest public health school and receives 20% of all U.S. government public health research funding. Partnered with Johnson & Johnson, it also awards grants for community wellness and healthy lifestyle programs addressing obesity, and leads the national Meatless Monday campaign. The Center for Human Nutrition, according to its Director, has had a leading role in IOM committees establishing dietary guidelines, the body of research on food and cancer for the World Cancer Research Fund, and the World Health Organization’s new BMI growth curves for children. [A few World Cancer Fund posts here, here, here, here, here, and here.]

Readers may most remember Dr. Wang from his 2006 study predicting a worldwide pandemic of childhood overweight and obesity, and that rates in the U.S. alone would double to 46% by 2010. [Since rates have been flat for nearly a decade, it’s doubtful his predictions will come true in the next 17 months.]

In describing their method for this latest prediction, Dr. Wang and colleagues wrote that they based their projection analysis on NHANES data collected between 1970s and 2004. According to the authors, “the prevalence of overweight and obesity has increased steadily among all U.S. population groups over the past two to three decades.” And:

We estimated the average annual increase in the prevalence of overweight and obesity and predicted the future prevalence among U.S. adults and children assuming the trends would be similar of the past three decades. Most of the past prevalence estimates based on NHANES data used in our analysis were based on previously published estimates, all of which were based on analyses done with consideration of survey design effects and sampling weights. [They referenced two previous papers led by Dr. Wang.]

They went on to use linear projection models to predict what the future will be if their past trend estimates continue. They said that “by the year 2030, about 90% of all American adults would be overweight and 51.5% would be obese.” And by the year 2048, “the prevalence would reach 100%.” See the mostly linear line straight up since 1970:

This is what a lot of the public thinks the obesity epidemic looks like. Before we examine this chartsmanship, the authors not surprisingly concluded by calling for more research and broad, comprehensive public interventions to address “the pervasive and serious problem” of that projected obesity epidemic:

As articulated by the World Health Organization for the global situation and by the Institute of Medicine with respect to childhood obesity in the United States, dramatic and effective population-based programs and related policies need to be developed and implemented to address the epidemic... More recently, there has been increasing recognition of the major role that the “obesogenic” environment plays in perpetuating the epidemic. What is needed now are creative initiatives to actually effect environmental changes, and this will require a strong and sustained collaboration among the public and private sectors, educators, food producers, urban planners, transportation experts, parents, and the general public. [A few IOM posts here, here, here, here, here, here and here.]


Convincing us of an obesity epidemic

We’ll try this one more time. It is frighteningly easy to lie with statistics, charts, maps and graphs and even fool healthcare professionals into believing there's a runaway obesity epidemic. Darrell Huff’s classic book should be required reading for every student, healthcare professional, journalist, politician and anyone else who thinks they know how to fix things for the rest of us.

Illustrating the nonsense inherent in mathematical extrapolations, for example, between 1947 and 1952, the number of television sets in American homes increased around 10,000%. Projected for the next five years would predict 40 sets per family! And if you want to be even sillier, begin with a year before television sets were available, and you can “prove” that each family will have 40,000 sets! Now, that’s an epidemic of television sets.

Popular perceptions of an obesity epidemic come from nonstop marketing about skyrocketing increases in the number of ‘obese’ Americans and even the CDC saying that ‘obesity rates’ have doubled between 1980 and 2000. Most people have an image of nearly half the population now as fat as those headless images shown on television, and believe that our weights, themselves, have been skyrocketing just like that graph created by Dr. Wang and colleagues. As Mr. Huff would say, we’ve been misled by “statisticulation.”

Sadly, the populace has been led to confuse a contrived obesity epidemic with actual obesity.

Please read that sentence again. The two are not the same thing. The factors behind each one are very different, too. And not one of the popularly proposed “solutions to the obesity crisis” understand the difference and they will continue to be ineffective, waste vast amounts of limited public health resources, and put people at risks for harm. But when you understand the difference, then the fallacies of logic [see here, here, here, here and here] and prejudices behind the war on obesity will become much more readily apparent.

Throughout time, humans have exhibited a natural diversity of physical characteristics. People have always come in a wide range of shapes and sizes — some thin, some heavy and the largest numbers falling somewhere in the middle. When the heights and weights of a population are plotted on a graph, a rough bell curve is seen. There’s a small number normally at the low-weight end, and small number at the high-weight end, with the largest in between.

Those whose natural, genetic sizes are at the most extreme high end have been called ‘obese.’ This obesity is not a new characteristic and has been documented since the earliest history of the human species. But cultures didn’t used to see it as a fashion negative. As Dr. Jeffrey M. Friedman, M.D., Ph.D., head of the Laboratory of Molecular Genetics at Rockefeller University and a renowned molecular geneticist and obesity researcher, explained several years ago during an NPR interview:

Historically, being obese was the desirable body habit as so. If you go to museums… all the rich people in Egypt would pay extra money to have extra chins put onto their sculpture. Rubenesque figures were the vogue in the 1700s. Renoir’s characters were all heavy. In aboriginal societies the chieftains were all quite obese. For reasons that — you all have as good an idea about as I do I guess — things changed here about what our views of what was attractive in the 60s and it set up an idealized view of what people should weigh and who they should be that just isn’t matched by our genetic endowment.

Now, to create perceptions of a contemporary obesity epidemic, two key things have transpired. And neither was that the numbers of truly obese people have gone through the roof as is widely believed. The bell curve still looks roughly the same as it always has.

First, an arbitrary line through the bell curve was drawn and everyone to the right of the line was called ‘overweight’ or ‘obese’. The line used to be drawn to the right of the main peak on the curve (BMI 27), but that changed in 1998, when the line was moved to the left, to cut directly through what is average for most of the population (BMI 25). Instantly, 30.5 million more people found themselves added to the rosters of the ‘overweight’ and the arbitrary change increased the numbers who fell into that category by nearly 50% — without a single person gaining a single pound. [The image, from Dr. Wang et.al., plots BMIs since 1976 from NHANES surveys.]

The second thing that has transpired over the decades, as you can see, is that the entire bell curve has scooted over to the right a bit. This means that the average weights and heights of the population have increased. But as you can see, it’s not all that impressive. There’s a reason you hear about obesity rates (the number of people crossing that arbitrary BMI threshold) as doubling and never hear how much our heights and weights have actually increased: the change is far less dramatic-sounding and wouldn’t illicit the hysterical reaction needed to create a crisis. Going back forty years, by the way, we’ve also gained an inch in height (the other part of the BMI equation).

And how much weight have we, as a population, gained on average?

While describing the difference between genes and the “misconception about the role of environment because of the misuse of statistics,” Dr. Friedman explained:

So let me give you an analogy... Imagine that 40 years ago the average IQ was 100 and there was a bell-shaped curve. Imagine now that our educational system improves and the bell-shaped curve shifts a little and the average IQ is now 105. With that you could imagine that the number of people who have an IQ greater than 140, so-called geniuses, might have doubled. Now is it more useful to think about how our education is doing by saying, “average IQ increased 5 points” or “number of geniuses is doubled.” I think probably both are of interest but the former seems to me more informative.

Ok. So how does that analogize to weight? Over the time period that you’ve heard that the obesity rates have quote “doubled” or gone up by 70 percent, the average weight gain is 7 to 10 pounds... think about the fact that 7 to 10 pounds is almost nothing compared to the hundreds of pounds of difference in weight that you might see in any two people walking around the street today, both of whom essentially have unlimited access to calories.

A crisis declared over 7-10 pounds! Failing to understand statistics also means a lot of people don’t understand why moving the entire bell curve on a graph by such a nominal amount can mathematically increase exponentially the numbers of people falling within a category and enhance perceptions of an epidemic.

Remember that 2003 Rand report which had announced that the prevalence of the fattest people has increased the most? Once again, it’s not that their numbers have gone through the roof or that they’re responsible for the ‘obesity epidemic’. Over the past two decades, the largest people (those common television images of obesity), with BMIs ≥50 have increased 5-fold (to a mere 0.5%), and those with BMIs ≥40 have quadrupled (to only 2%). The exponential increases are a mathematical phenomenon of that entire bell curve shift.

What has continued to exasperate those trying to create a crisis of this shift in average weights is that the population keeps getting healthier and living longer than at any time in our history. There’s good reason that the health crisis of the obesity epidemic has failed to materialize.

Some of this can also be explained by the factors that have contributed to that shift in the bell curve — mostly neutral or positive changes as a result of our modern life and how good we have things.

For example, remember the bell curve is based on population surveys — and not only is our population very different from earlier decades, but the NHANES surveys themselves have been redesigned to oversample certain segments of the population — all of which would tend to amplify average weights — lending caution to seeing trends by piecing surveys together. Our population has become dramatically more diverse just since 1970, when ethnic/racial minorities were 16% of the population, compared to a third today. The bell curve (and NHANES surveys)also isn’t corrected for age and the population is aging, significantly so since the 1950s and 1970s, especially in the babyboomer age group. Animals and the human species naturally gain weight with age, especially during middle age. As Dr. Friedman said, aging could notably contribute to the perception that weights in our population are increasing.

And then there are all of those modern advances that have helped children and adults to live longer, healthier lives, while contributing to healthier average weights and heights among the entire population. Modern food production has improved, enabling more affordable food to feed more people. Compared to past generations and previous war times, fewer people are going hungry or facing food rationing or severe nutritional shortfalls, which had left more people underweight for their genes and skewed the population averages slightly downward. Our food and water are considerably safer today and have reduced by 100-fold the number of children and adults sick (and dying) with foodborne illnesses. Far fewer children spend their childhoods sick and fighting childhood illnesses, that have been eradicated just in recent decades. Most of us remember measles, mumps, chicken pox and polio among our classmates and siblings when we were younger. As recently reported, according to the CDC, before the introduction of the measles vaccine, alone, in 1963, “approximately 3 to 4 million persons had measles annually in the United States; approximately 400-500 died, 48,000 were hospitalized, and 1,000 developed chronic disability from measles encephalitis.”

And, of course, there are other possible iatrogenic factors, such as stress and prescriptions for anti-depressants and anti-anxiety medications that tripled between 1988 and 1999, most of which have been shown to increase weight, age-related diabetes, and auto-immune disorders. Smoking, which many used to help control weights, is down. Dieting, which since the 1970s has grown into a $61 billion/year industry in the U.S., can help to ratchet up weights if diets are repeated in rapid succession. And so on.

So, for that ‘obesity epidemic’, the entire bell curve has scooted over, not because it’s the fat people eating more or being sloths compared to the thinner people. Fat and thin have been clinically shown, by researchers at the University of Rockefeller and others, to eat and behave no differently to explain the diversities of sizes. And, as you see from the bell curve, despite the media portrayals, the variation of our sizes isn’t nearly as dramatically different today than it’s always been.

Through a lifetime of diligently controlling eating and behaviors, most people can have a long-term control over their weight within a 10-15 pound range, according to Dr. Friedman and colleagues, not enough to change someone’s body type. Just as true obesity is primarily genetic and not caused by or changed by how much people eat or how active they are, so will interventions focused on calories in-out — and all the fruits and vegetables and sidewalks in the world — do nothing to eradicate actual obesity.

As Dr. Friedman added about the misguided focus on calories and foods:

[T]he focus is wrong and there are many levels at which to answer the question. The first is that the idea that you can change your weight voluntarily is one that the diet industry has a huge financial interest in. And so anybody can tune into infomercials telling you what you need to do to lose weight is fork over some money to their diet plan, eat it or not eat it and you’ll lose weight. And I don’t think that message and that marketing muscle can be easily counteracted by what scientists have to say about it. Part of the problem is that that notion fits in with people’s intuition. And this gets very complicated. It’s a control issue; people want to feel like they’re in control of what they eat and what they weigh... the available evidence would suggest that the vast majority of people who find themselves at a particular weight — be it thin, medium, or heavy — that’s pretty much their weight. Now actually one of my neighbors heard me talking about this at one point and he said, “Oh, but that’s so sad.” Well, I don’t know...

I mean it’s also sad if you don’t like your height and you don’t like your eye color and you don’t like lots of things about yourself. It’s that somehow people think this is something they ought to be able to control. And they accept all these other things you can’t control that are just who you are. But people are very loathe to believe that what they weigh is [what] they are.

And others to accept them as they are, too.


© 2008 Sandy Szwarc. All rights reserved.

Bookmark and Share