TCS Daily

On Obesity, What the Researchers Didn't Find

By Sandy Szwarc - October 7, 2004 12:00 AM

Several studies citing correlations between bad foods or sedentary habits and rising rates of obesity and type 2 diabetes have filled the news lately. The studies seem, at first glance, to confirm what "everyone knows" about why people are fat -- they eat too much "bad" food and exercise too little -- and what to do about it. But correlations can't hold water to sound clinical evidence.

While it's in our best interests to look past correlations for sounder studies and weigh them in light of the entire body of evidence, such studies aren't always easy to find. Like all food and health news, we typically only hear one side. And we almost never hear about the studies that disprove something or find something didn't work or isn't a health concern -- especially if they go against a popular belief.

However, studies disproving a concept are especially important and during the scientific process, ideas that have been disproven are thrown out and scientists move on to find the real explanations. But a classic earmark of pseudoscience ("false" science) is failing to abandon ideas and doggedly continuing to claim something is true or works regardless of overwhelming evidence to the contrary.


On the subject of the evils of sodas and snack foods, for example, here are a few new studies that barely made the news recently. Their negative findings were notably different from the popular axioms and yet they concur with a sizable body of such evidence.

Two different analyses from the Growing Up Today Study (GUTS), based at Brigham and Women's Hospital and Harvard Medical School in Boston, were released last month. GUTS is a databank of questionnaires about diet, lifestyle habits and health that were gathered from more than 16,000 children, 9 to 14 years of age. Their mothers are from the Nurses Health Study, the huge database of questionnaires gathered since 1976 from over 120,000 nurses. The study has the limitations inherent in population studies, but what makes these two studies from GUTS significant is that the researchers couldn't even find a connection between soda or snack (ice cream, candy, chips, sweet baked goods, etc.) consumption and weight among these kids after 3 years. In other words, fat children weren't eating more sweets than thin children.

What the GUTS research, led by Allison Field, did find, however, was that regardless of their overweight status, children who dieted gained significantly more weight compared to children who never dieted. This confirms another study these same researchers released last October which found the BMIs of girls who were frequent dieters versus those who never or rarely dieted were nearly 4 entire BMI points higher. This was after they accounted for all the other factors that could have explained the differences, including physical activity, television watching, etc. The researchers concluded their data suggest that dieting is not only ineffective but in the long-run may actually promote weight gain. And, indeed, clinical studies have confirmed just that.

Perhaps the most significant study to come out this month was the one that got the very least media attention. The results of the DONALD Study (Dortmund Nutritional Anthropometric Longitudinally Designed Study) were released from the Research Institute of Child Nutrition, Dortmund, Germany. This was a small cohort study on 228 nondieting children. The researchers themselves actually weighed the individual children and recorded their diets (the foods, amounts and eating occasions) at least ten times a year and followed them thusly for 17 years. They found that no identifiable dietary patterns during childhood or adolescence could explain their BMIs. While there were great differences in the children's diets, these differences were not related to their weights.

The GUTS and DONALD studies join a profusion of other studies, both clinical and epidemiological, over the past fifty years demonstrating that fat children and adults as a population normally eat exactly the same as thin people. And regardless of their diets, children will still naturally grow up to be a wide range of heights and body weights. "Multiple researchers, using a variety of methodologies, have failed to find any meaningful or replicable differences in the caloric intake or eating patterns of the obese compared to the non-obese to explain obesity," concluded David Garner, Ph.D. and Susan Wooley, Ph.D., for example, in their review of some 500 studies on weight in Clinical Psychology Review.

How Can This Be?

How can this be reconciled with the laws of thermodynamics? The findings seem at odds with what conventional wisdom might suggest to us about eating and weight gain.

One of the country's foremost obesity researchers, Jeffrey M. Friedman, M.D., head of the Laboratory of Molecular Genetics at Rockefeller University explains that the commonly-held simplistic belief that obesity is just a matter of eating too much and/or not exercising enough is "at odds with substantial scientific evidence illuminating a precise and powerful biologic system." According to his research and that of numerous others, obesity is the result of differences in biology and metabolism, not behavior, diet or the environment. Through their own volition, people can control their weight long-term to a very small degree. Even voluntary physical exercise has minimal effect, according to Friedman and Glenn Gaesser, PhD., exercise physiologist and obesity researcher at the University of Virginia. So, while better access to foods can account for some of the increases seen in the average height and weight of all people in developed countries -- 7 to 10 pounds in the U.S. since 1980s -- it's genetics and not the environment that accounts for the largest proportion of the differences in size among people, Friedman explains.

"The propensity to obesity is, to a significant extent, genetically determined," he says. Someone genetically predisposed to obesity "will become obese independent of their caloric intake" even when it's restricted to that of thin counterparts. "The heritability of obesity is equivalent to that of height and greater than that of almost every other condition that has been studied," Friedman states.

Negative studies disproving things "everybody knows" are important for leading us to sounder answers. The strongest research will never find proof of something that doesn't exist. Only pseudoscience can ever do that.

© 2004 Sandy Szwarc. All rights reserved


TCS Daily Archives