3 Competing Obesity Trends: Up, Down, and Sideways
Competing obesity trends fill the headlines and make it tough for people to take health reporting seriously. Here’s a sampling from just the last week.
- Gallup: U.S. Obesity Rate Ticks Up to 27.1% in 2013
- LA Times: Obesity in Young American Children Plummets
- Reuters: Obesity Rates Remain High, but Stable in the U.S.
How can credible sources widely report such seemingly conflicting facts so broadly without reconciling them?
Simply put, despite the complexity of obesity, the public — and therefore the media — see obesity as a very simple result of eating and physical activity. They gloss over the complexities and relatively uninformed people publish their own authoritative explanations of what’s going on.
In the case of these three competing observations, only one is authoritative. Analysis by the smartest obesity epidemiologists around shows that obesity rates overall are high, but stable. That’s it. That’s all you need to know, unless you are an epidemiologist intent on dissecting the cross-currents in obesity trends.
The “plummeting” rate in a subgroup of a subgroup of children — toddlers — is a dubious artifact with no statistical significance. Headline material? No.
The Gallup data tracks self-reported height and weight on a regular basis and they do indeed see a measurable uptick in what people report as their weight compared to their height. They saw that the percent of people who reported height and weight corresponding to a BMI of 30 or more (obesity) went from 26.2% in 2012 to 27.1% in 2013. And they got headlines by saying that the change was greater than their sampling error. The trouble is, Gallup ignores the errors of self-reports. People lie about their weight. They lie about their height, too. And how much they lie varies with the circumstances.
For example, people in the north central U.S. seem to understate their weight more than people in the deep South. We know this by comparing self-reported BMI to actual, measured BMI. And the differences are enough to significantly skew results. Likewise, over time people can change how much they distort self-reports of their height and weight. Given this source of error, don’t bet your fortune on this 0.9% being real. It’s more likely to be an insignificant fluke.
“It is the weight, not numbers of experiments that is to be regarded.” — Isaac Newton
Click here for more on the Gallup study, here for more on the toddler obesity artifact, here for more on the fundamental trend, and here for more about errors in BMI self-reports.
Ups and Downs, photograph © larsomat / flickr
Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.
