Subscribe to Decision Science News by Email (one email per week, easy unsubscribe)
OBSERVATIONAL STUDIES: “CONVINCING PEOPLE OF THINGS WITHOUT ACTUALLY GENERATING EVIDENCE”
The New York Times had an article on the widespread confusion about whether skipping breakfast causes you to gain or lose weight. It seems there’s no substantial evidence that skipping breakfast has any effect on obesity.
What we found most amusing was this passage with a quote from Dr. David Allison:
Dr. Allison said that the true relationship between eating breakfast and body weight, if there is one, was still an open question. But observational studies that tout an association between the two are churned out “just about every week,” despite doing nothing to actually test or prove the claim.
“At some point, this becomes absurd,” he said. “We’re doing studies that have little or no value. We’re wasting time, intellect and resources, and we’re convincing people of things without actually generating evidence.”
It reminds us of the recent paper in which the authors decided to pick 50 common cookbook ingredients at random and found that 80% of them were claimed to cause or prevent cancer. Little evidence was found for any of the claims.
Here’s an idea: If the reason behind having IRB approval for studies is to prevent their doing any harm, why isn’t there an IRB approval process for publishing (as opposed to conducting) observational health and nutrition studies? They seem to do a lot of harm in making people change their diets and behavior for no good reason, not to mention the needless worry they cause.
Large scale randomized trials would be exempt from this pre-publication IRB process. Wink.