Digging Through Data on Obesity to Find a Prize
Confirmation bias is a potent motivator. Pollsters get fired when their polls don’t tell the right story. Industry scientists may do better when they find results to advance an employer’s interests. Academics can prosper when they produce evidence to support their hypotheses. And government researchers may feel more job security when their work aligns with government policies. So digging through data to find a prize is not so unusual.
This week, digging in obesity data has yielded some pleasing but misleading numbers.
Bring Out the Megaphone
Obesity Is Down (in a Subset of Children)
This is a script we’ve read before. Reversing the upward trend in childhood obesity has been a priority for decades. And from time to time, data on a subset of children pops up to suggest a downward blip.
Thus it’s completely unsurprising to see researchers pull out a megaphone when they find data to suggest that an obesity prevention program is working. It feels good and it grabs headlines. Yesterday, the Associated Press ran this headline:
US Preschoolers Less Pudgy
In Latest Sign of Falling Obesity
This is a recurring, misleading story. Obesity rates are not falling. Preschoolers are not becoming “less pudgy.” Indeed if you look at the definitive data on obesity prevalence, NHANES, the number went up between 2010 to 2016 for all preschoolers.
The number fueling the current headlines is for children in the WIC program for food assistance. That’s a population that changes constantly. During the years of this study, it was shrinking because the economy was improving. On the right, you can see the trendline for all children 2-5, compared to the trendline for children 2-4 in WIC.
Back in 2014, the CDC used numbers from the 2012 NHANES data to promote headlines about obesity “plummeting” for preschoolers. But with new data, it became clear: that number for 2012 was merely a blip. The overall rate is still going up through 2016.
The Fine Print
In the JAMA publication that triggered headlines this week, the authors from CDC concede that these trends may reflect changes in the WIC population. Or it may be that changes in WIC are having a positive effect. The true explanation is “undetermined.”
But that’s fine print. The headlines are all about “falling obesity.” The quotes from the authors of the study suggest that “this is a real change.”
It’s a great narrative about fixing a problem. Unfortunately, it doesn’t square with reality. If we are to solve this problem of high childhood obesity rates, we will need a passion for objectivity that’s lacking. We need genuine curiosity about what’s really going on. And above all, we need more caring for the children and families who are affected.
Click here for the study in JAMA, here for reporting from the AP, and here for more on WIC and childhood obesity.
The Diggers, painting by Jean-Francois Millet / WikiArt
Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.
June 20, 2019
June 20, 2019 at 6:45 am, Mary-Jo said:
It was encouraging to read the WIC data. But, for headlines to come out of it suggesting ‘success’ is plain irresponsible. Why can’t headlines say something like “encouraging results in childhood obesity, but much more still to do “, “good news re: childhood obesity, so much more must happen”, etc. just to add a small blurb of reality to the ongoing story. This will incite/create continued curiosity, and push for more objectivity, care, and attention to this complex, far from adequately addressed situation.