How Important Is Breakfast, Really?

F. Perry Wilson, MD, MSCE


April 24, 2019

Welcome to Impact Factor, your weekly bite of commentary on a breaking medical study. I'm Perry Wilson.

This week: Was your mom right all along when she told you that breakfast was the most important meal of the day? The answer to that question may appear in an article in the Journal of the American College of Cardiology[1] linking skipping breakfast with cardiovascular death.

It's OK to be skeptical on this one, folks. We get studies saying that eating X leads to heart disease, or cancer, or dementia, or prevents one of those things every other week, it seems. But let's not discard these studies out of hand. There is some learning to do here.

Researchers used the well-worn National Health and Nutrition Examination Survey (NHANES) to identify 6550 individuals between ages 40 and 75 without a prior history of cancer or cardiovascular disease and who had adequate follow-up information to conduct the analysis.

The first red flag is that sample size. The article states that there were around 40,000 participants in the dataset they were analyzing, but only 6550 met their inclusion criteria.

We are not told which individuals were kicked out of the study and for which reasons, but the fact that only 16% of participants were analyzed raises questions about whether the results can be broadly applicable.

At baseline, the participants were asked how often they eat breakfast. The majority reported eating breakfast every day, but a sizeable number didn't eat breakfast at all, breaking their poor mothers' hearts.


Like all lifestyle measurements, individuals who didn't eat breakfast at baseline differed in substantial ways from those who ate breakfast every day.


They were more likely to be black, less likely to be married, more likely to be a former smoker, more likely to be a heavy drinker, and much more likely to be physically inactive.

Now, of course the authors adjusted for these factors, but I am concerned that their adjustment for family income was inadequate, using three discrete categories instead of a more thorough adjustment for income as a continuous variable.

This is particularly problematic in a study like this one, because income might be a significant predictor of the ability to eat breakfast every day. And, as we've discussed many times, higher income is a strong predictor of better outcomes, for reasons that are fun to debate at dinner parties and political rallies.

On average, 6 out of every 1000 people who ate breakfast every day died yearly from a cardiovascular cause. Seven out of 1000 people who never ate breakfast died each year.


This small difference was statistically significant, given the large number of people in this study, but the lack of a dose-response, as you can see in this graph, diminishes my appetite for the findings a bit.

When we examine a study like this, we have two questions to ask. First, is there causality here? If there is, we can recommend that our patients start eating breakfast to reduce the risk for heart attack. But given these findings, I'm not too excited about prescribing breakfast to our patients.

If there is no causality, the second question is: Can we use breakfast-eating as a new risk predictor? After all, we ask our patients if they smoke, if they drink, and how much they exercise. Should we add "Do you always eat breakfast?" to the list of screening questions?

In short, no. Since the magnitude of risk was so minimally different between the always and never breakfast eaters, there's minimal yield to wasting time discussing it with patients. Better to use that time to, you know, convince them to get some exercise, or complain about the electronic health record.

So what do we tell our moms? Well, of course we tell them that they were right all along. But between you, me, and a bowl of corn flakes, don't worry too much about eating breakfast. There are bigger fish to fry.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.