Never mind the numbers, look at the neuroscience.
Q: Have you seen the headline: “Skipping breakfast makes you gain weight: study”?
A: If that’s the one with the chocolate cupcake photo, yes.
Q: Was this just another mouse study, or did they look at weight gain in people?
A: People, yes, but they didn’t measure weight gain.
Q: But doesn’t the headline say “makes you gain weight”?
Q: So what did they do?
A: They measured brain waves, and how much pasta lunch people ate. The people who skipped breakfast ate more.
Q: So it was a lab experiment.
A: You can’t really tell from the Herald story, which makes it sound as though the participants just chose whether or not to have breakfast, but yes. If you look at the BBC version, it says that the same people were measured twice, once when they had breakfast and once when they didn’t.
Q: And how much more lunch did they eat when they didn’t eat breakfast?
A: An average of 250 calories more.
Q: How does that compare to how much they would have eaten at breakfast?
A: There were brain waves, as well.
Q: How many calories would the participants have eaten at breakfast?
A: The part of the brain thought to be involved in “food appeal”, the orbitofrontal cortex, became more active on an empty stomach.
Q: Are you avoiding the question about breakfast?
A: Why would you think that? The breakfast was 730 calories. But the MRI imaging showed that fasting made people hungrier
Q: Isn’t 730 more than 250?
A: Comments like that are why people hate statisticians.
Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »