…a piece of wisdom that gets repeated often and gladly, and whose origin is surprisingly well documented.

There is also a royal version: “Eat breakfast like a king, lunch like a prince and dinner like a pauper.”

Spoiler: With many pieces of wisdom, nobody remembers where they came from. For “breakfast is the most important meal of the day” we even know the parents: a sanatorium and a PR man. Whether the two happened to be right anyway is another question. As we’ll see: the reason it spread so widely is not some broadly accepted finding of nutrition science, but good PR and marketing.

First references: 1917

First references appear in 1917, in an article by Lenna Frances Cooper(1), a dietitian at John Harvey Kellogg’s sanatorium (yes, the cornflakes Kellogg), co-founder of what is now the Academy of Nutrition and Dietetics(2) and a Seventh-day Adventist(3). She used the phrase in “Good Health”, a publication edited by John H. Kellogg.

This article, however, is probably not the reason why the saying stuck so firmly.

PR and marketing

In the 1920s, the pioneer of modern public relations and nephew of Sigmund Freud, Edward Bernays(4), was hired by the Beech-Nut Packing Company, which was struggling to sell its bacon. Bernays launched a PR campaign in which he asked doctors whether a heartier breakfast (than coffee and toast) might be beneficial - and then published the opinion of the doctors who answered as “Breakfast is the most important meal of the day” in magazines. From there it settled into the public mind. And bacon sales really did shoot up(5).

In the 1940s, the slogan was picked up again by Kellogg’s, who used it in their advertising to position cornflakes as the ideal start to the day.

So, if you like, you can use this old “wisdom” - with a source to cite - to justify a breakfast of bacon and eggs just as well as to argue for cereal.

Both have the same scientific basis: none!

Fine. There is science

There are plenty of observational studies on breakfast - but just like in other articles on this site (I link some below), they either show insignificant associations (as we have noted for observational studies several times before) or are plagued by “healthy user bias”, so I’ll spare myself the references here.

Because: anyone who skips breakfast and replaces it with a cigarette can - perhaps not entirely surprisingly - expect a less healthy outcome (to name just one example of this bias).

But there are also randomised studies, and they carry more weight than observational studies.

Betts et al.(6) split lean adults into two groups for six weeks: one ate breakfast daily, the other skipped it and fasted until late morning. With breakfast, participants took in more energy overall, but their resting metabolic rate stayed the same, and blood glucose in the afternoon and evening was more stable.

For a meaningful comparison, though, they actually had to introduce a specific definition of “breakfast” - more on that in the section “Breaking the fast”.

Dhurandhar et al.(7) followed around 300 adults with overweight for 16 weeks. One group was told to eat breakfast, one to skip it, and the third received only general nutrition information. In the end, weight loss did not differ significantly between the groups, even though the recommendation did change the participants’ self-reported breakfast habits.

In another paper by Betts et al.(8), very fittingly titled “Is breakfast the most important meal of the day?”, they write:

These trials were categorically not designed to answer whether or not breakfast is the most important meal of the day. However, this review will philosophise about the meaning of that question and about what questions we should be asking to better understand the effects of breakfast, before summarising how individual components of energy balance and health respond to breakfast v. fasting in lean and obese adults.

and they find no evidence that breakfast moves weight or energy expenditure in the direction the saying promises.

Questionable relevance criteria in observational studies

Scrabble tiles spelling the word 'Health'
Is Meat Carcinogenic?

Meat is considered both extremely nutrient-dense and carcinogenic - how can that be, and what’s behind it?

A steak with a twig of rosemary, surrounded by plenty of salt, which in the context of this article may just as happily be read as sugar
Does Meat Cause Diabetes?

The ‘fact everybody knows’ - that meat causes diabetes - keeps crossing my path. So I went and read the studies, and ended up with the longest reference list of any article here.

Breaking the fast

The answer, in a way, has been hiding in plain sight in the English word itself. “Breakfast” is literally the meal that breaks the fast. It says nothing about the clock, only that you haven’t eaten for a while. German doesn’t have that luxury: “FrΓΌhstΓΌck” means roughly “early piece” (of bread, originally), so the German word is all about the time of day and says nothing about fasting.

Intermittent fasting has its own breaking of the fast, too - but depending on where you set your “eating window”, it happens earlier or later in the day. Not unimportant, that meal, but no more important than any other.

Betts et al. noticed this as well. Back in 2011, as part of the Bath Breakfast Project, they introduced a definition of breakfast that takes the time of day into account(9). Their trial distinguished between:

  • 0 kcal until noon
  • 350 kcal within 2 hours of waking and 700 kcal before 11:00
  • no change in habits (control group)

The king-prince-pauper proverb, by the way, thinks only in portion sizes across the day. Fasting doesn’t come into it at all.

Biology (or biochemistry)

One of the reasons often given in favour of breakfast is the energy it supplies for the day. Widespread, after all, is the myth of sugar as the preferred fuel - in this case I don’t even want to argue against it, but rather address one part of the argument:

Sugar (or rather: carbohydrate) is at least the usual fuel in today’s society - because it is consumed in absolute excess. I have already written about the rest:

Why "usual" doesn't mean "preferred"

Fuel gauge with an overly large 'reserve' area. The by far smallest part makes up the actual fuel meter

The Preferred Fuel

The body’s preference: Which fuel does it like to burn most? (The answer will surprise you πŸ˜‰)

Read more β†’

Let’s look at the metabolism simulator. Before waking, cortisol levels rise, which causes the liver to release more glucose. Every day simulation shows this, without any meal having been eaten beforehand (default wake-up time: 6 am, the “dawn” effect starts at 4:30 am). Here is an example of a fasting day with no meals at all, which makes the chart easier to read.

A fasting day with the πŸŒ… morning rise in cortisol and blood glucose (red line).

If we added activity - or even simulated ketones during the fast - it would be clear: there is always enough fuel for the organism, even without eating any meals, and therefore entirely without carbohydrates.

So the body has already served breakfast in bed. The real question is whether you want a second one.

Conclusion

What remains: the breaking of the fast that is built into the English word is something the body takes care of entirely by itself, on time and without an advertising budget. Whatever lands on the table afterwards - whenever that may be - is a meal like any other, just with better PR.