Looking for a Pony in the Muck of Uncontrolled Dietary Nudges
“There must be a pony somewhere.” So goes the punchline of a joke about a manure pile that has circulated in various forms for more than a century. A new report in BMC Nutrition brings this joke to mind. From a great muck of uncontrolled data about dietary nudges in a single hospital convenience store, researchers find justification for claiming “nudge tactics can improve staff’s food intake and lower Na/K.”
To say the least, there are methodological problems that make it inappropriate to use this study for supporting for such a bold claim.
Pre-Post Design
Perhaps the most glaring problem is the pre-post design this study employs. In 2018, the researchers surveyed the dietary habits of 222 employees in a single hospital. They also collected data on sodium and potassium excretion from a medical checkup then. They implemented nudges in the hospital convenience store a year later. The intent was to promote healthier food purchases. Then in April and May of 2020, researchers collected another round of dietary self-reports and data on sodium and potassium excretion.
There’s no control group – just a single cluster of people exposed to the same intervention a year after the baseline data was collected. We have no idea what changes might have happened even before the intervention started. For this reason alone, these results have very little meaning.
The problems with pre-post studies are numerous. Spontaneous change in the variables under study are always a possibility. Other changes might be in play. For instance, the post intervention measures were taken when the COVID pandemic was accelerating at terrifying pace. This alone might have an overwhelming effect on dietary behaviors.
Post Hoc Protocol
We also note that the protocol for this study came after the study was complete. This is precisely the opposite of how clinical trial registrations are supposed to work. The registration of the protocol should come first. But this study started in 2018 and finished in 2020. The protocol was not registered until 2022.
That’s a problem.
Where Is That Pony?
We really would like to believe nudges to the food environment would be sufficient to prompt meaningful dietary changes – but this uncontrolled study offers no evidence for that. Wishful thinking will bring us neither a pony nor a suite of dietary nudges that work to improve health.
Click here for this study and here for further perspective on the need for rigor and credibility in nutrition research. For more on the problems of pre-post study designs, click here. Thanks to Dean David Allison for inspiring this post.
The Pony, painting by Pierre Bonnard / WikiArt
Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.
August 19, 2024
