My interest in writing a blog right now is pretty much nil, but I cannot let today’s misleading boston.com article entitled “Obesity Cuts Life Expectancy by Up to 14 Years, Study Shows” go by without reacting, for I know the damage that pieces like this do to people, including some of my patients.
Long story short: The researchers who authored the primary source article did not adequately control for behaviors. They screened out potential participants who had ever smoked and/or had a history of certain diseases, but the lifestyle behavior information they collected from participants was limited to alcohol use and physical activity level. Researchers collected no information about other lifestyle factors, like stress management and eating and sleeping habits, all of which can impact health. The behavioral data they did collect was self reported, which introduces all sorts of error. Other research has shown that when behaviors are controlled for, body weight does not seem to matter, but the study design that these authors used prohibited any opportunity from being able to confirm or refute those findings.
The boston.com piece discusses a second article as well that examined the relationship between obesity and exercise. In reference to this latter article, the boston.com piece’s subheading concludes with, “And it’s under-exercise, not overeating, that’s causing America’s [obesity] epidemic.” That eye-catching text will certainly garner many clicks, which is unfortunate because it is not true. The actual research piece reads, “The research highlights the correlation between obesity and sedentary lifestyles, but because it is an observational study, it does not address the possible causal link between inactivity and weight gain.”
I cannot stress it enough: Correlation is not causation. They are entirely different. I know, I know, we each know somebody who has put on weight after they stopped working out. Sure, that does happen sometimes, but on the macroscopic level that is the population, the picture is much more complex than that with many other factors in play.
The boston.com article’s final paragraph begins with, “Losing weight is proven to significantly reverse the health effects of obesity.” Wrong. When we adapt healthier lifestyle behaviors, our body weight might change as well, but if we credit the weight change instead of the behavior change then we have it backwards.
The harm in all of this is that it reinforces a weight-centered model of eating and physical activity that ultimately fails nearly everybody who uses it. If we take a weight-centered approach and do not maintain the weight we want, we risk losing motivation and reverting to old behaviors because the goal was unattainable.
There is a better way. In the health-centered model that we advocate, the behaviors in and of themselves matter independent of weight. Whether weight goes up, down, or stays the same is irrelevant because the behaviors themselves are what count. Better-designed research seems to support this model: When we control for behaviors, health and weight look to be independent.