And, here we have another example of the federal government making recommendations on next to no scientific evidence, built mostly on a foundation of speculation and a deep desire to tell people what to do, that ends up probably having hurt the people they were trying to help.

We’ve seen this before, with trans-fats, eggs, and salt, and good ol’ fat.

Now, breakfast. The most important meal of the day might not be so important.

Researchers at a New York City hospital several years ago conducted a test of the widely accepted notion that skipping breakfast can make you fat.

For some nutritionists, this idea is an article of faith. Indeed, it is enshrined in the U.S. Dietary Guidelines, the federal government’s advice book, which recommends having breakfast every day because “not eating breakfast has been associated with excess body weight.”

As with many nutrition tips, though, including some offered by the U.S. Dietary Guidelines, the tidbit about skipping breakfast is based on scientific speculation, not certainty, and indeed, it may be completely unfounded, as the experiment in New York indicated.

At 8:30 in the morning for four weeks, one group of subjects got oatmeal, another got frosted corn flakes and a third got nothing. And the only group to lose weight was … the group that skipped breakfast. Other trials, too, have similarly contradicted the federal advice, showing that skipping breakfast led to lower weight or no change at all.

“In overweight individuals, skipping breakfast daily for 4 weeks leads to a reduction in body weight,” the researchers from Columbia University concluded in a paper published last year.

Oops.

In defense of breakfast, I’ll add that if you have high-protein meals in the morning for a couple weeks instead of high-carb, you’d probably stay flat or lose weight, but it seems likely that skipping breakfast is not the danger it was once thought to be.

The recommendation on breakfast is rather recent, unlike the longtime vilification of eggs and fat and salt. Perhaps it’s had less time to do damage before being reexamined and changed during the review process of dietary guidelines that happens every five years? But how did the recommendation get there in the first place? The culprit is a reliance on observational studies instead of randomized controlled trials, which are the gold standard of such studies.

In an observational study, scientists are unable to eliminate other factors that might influence weight disposition, and adjust for such discrepancies statistically after the fact:

In analyzing the results of observational studies, scientists make statistical adjustments to adjust for the potential confounding factors that they can measure — age, alcohol consumption, exercise, employment, and the like. Breakfast skippers in the health professionals study, for example, tended to drink more, smoke more, and exercise less. The scientists adjusted their statistics accordingly. But the adjustments are imprecise, and there is no guarantee that the groups are not different in some other unmeasured way.

Randomized controlled trials on breakfast show the exact opposite influence, but it was the observational study conclusions that made it into dietary guidelines, into federal programs, and into media coverage of nutrition from 2010 forward. Only one randomized controlled trial was included in the 2010 guideline discussion and it was inconclusive on breakfast’s influence. Oops.

David Allison, of the University of Alabama-Birmingham, has become one of the leading critics of what he sees as the misuse of nutritional research. He recently compiled a list of the randomized controlled trials that investigated links between breakfast and obesity.

He found five, and none offered clear evidence that skipping breakfast leads to weight gain. (The New York research was funded by Quaker Oats, a unit of Pepsi Co, though the results could hardly have been what the breakfast food company would have hoped for.) Mostly, it seemed, skipping breakfast made no difference. A sixth study, published this month in Obesity, also showed no differences in weight loss between those who ate a breakfast and those who skipped, though subjects who had a high-protein breakfast gained less body fat.)

Allison attributes the widespread adoption of the breakfast hypothesis at least in part on researchers who read too much into observational studies, and wrongly ignore the stronger evidence from the randomized controlled trials.

I understand that science is a process of discovery, and not every nutritional recommendation is going to stand the test of time. It’s fine if they evolve and change as new evidence emerges. The problem with federal recommendations is they are given disproportionate weight by media and citizens. They dictate food choices and subsidies in all kinds of federal programs, funded by us. They are repeatedly shown to be based on a lot of speculation and extrapolation and very little reliable data. And, they are very, very slow to change when real evidence refutes their conclusions. Their effects, therefore, can be widespread, longterm, and damaging to Americans. Even with the evidence mounting that saturated fats, salt, eggs, and skipping breakfast might not be nearly as damaging as the federal government’s advice on them, there’s some question as to whether the board making dietary recommendations in the government’s five-year review process will actually change any recommendations based on this evidence.

But by all means, let’s have them ban more things categorically. They’ve been so reliable up until now.