Publication Bias at Work: The Case of Parks and Obesity
Sometimes, objective evidence hits the wall against a very popular idea. Bam. Publication bias means that even a careful study might not see the light of day. Consider the case of public parks and obesity.
The Indisputable Value of Public Parks
Who can dispute the value that parks bring to our lives? They might prompt us to be more active. Bring us closer to nature. Give us a place to calm our minds and release our stress.
Community action and public spaces to promote physical activity have a central role in CDC strategies to prevent obesity. The U.S. Surgeon General calls us to “Step It Up” to help prevent obesity. Public parks occupy the core of that call to action.
But Do Public Parks Actually Prevent Any Obesity?
Children who live near built environments – including parks – with more walkability tend to have lower BMI z-scores. In 2012, a systematic literature review told us that 89% of empiric studies report a beneficial relationship between the built environment and physical activity.
Unfortunately, virtually all of those studies were observational. They are inadequate to establish cause and effect. And most of those studies focused on physical activity, not obesity outcomes.
So some of the same researchers embarked on a quasi-experimental study that might answer questions of cause and effect. And they found no evidence for a benefit. As you might guess, it was an unpopular finding. It was difficult to publish. Lecturing this week at UAB’s conference on causality in obesity research, senior author Nir Menachemi explains:
Before this experience, I had never been rejected so many times on a study that was not fatally flawed. Typically, the reviewers told us something that amounted to “I just don’t believe your results.”
This experience made it clear to me. Controversial null findings present a problem for objective publication reviews.
Problem Solving Requires Good Evidence
Multiple levels of factors contribute to obesity. It’s a wickedly complex problem. Perhaps expecting parks to have an effect on obesity rates is asking too much.
But one thing is certain. Without objective facts and solid evidence, real solutions will be hard to find. Publication bias is a serious problem. We can’t afford to dismiss evidence that doesn’t fit with our presumptions.
Click here for Menachemi’s presentation at this week’s conference. For the systematic review of built environment and physical activity, click here. You can find the quasi-experimental study here.
Playground, photograph © Phalinn Ooi / flickr
Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.
July 28, 2017
July 29, 2017 at 11:46 am, Allen Browne said:
Yup!
Excellent!
Thank you!
Allen