It is hard to read any mainstream media content these days without being told that food X is either death on a stick or the key to longevity.

These kinds of articles are often written by “science” journalists, an oxymoron if there ever was one.

They get a press release from a company who have paid for some “science” to be done and, would you believe it, their product leads to miraculous changes in X,Y or Z.

The journalist then dutifully repeats the headline and gets some serious click bait.

How easy is it to do?.

Turns out very.

John Bohannon PhD set up to see if he could convince the world you can lose weight by eating chocolate.

They paid 16 people 150 euros to go on a diet for three weeks and ran a battery of fifteen tests plus weighing them daily.

They were then divided into three groups:

  • Low carb diet
  • Low carb diet plus 45g chocolate
  • Control group (do what you want)

At the end of the three weeks, they were weighed again and tests were re done.

Both low carb groups lost 5 pounds in weight, while the control group showed no changes, but…..the low carb chocolate group lost weight 10% faster.

This was a “statistically significant” difference.

Now you might be thinking, hang on, the study did show a difference, so how is it a con?

Because the only difference was 10% faster, which while it may be statically significant, the reality is it’s clinically meaningless.

Also, more importantly, if you measure enough stuff, eventually you get a random chance that will show a statistically significant change, but it is pure luck.

Remember a P value of 0.05 or less means the chance of the result being chance is 5% or less, but they measured 15 things, so they had a 60% chance of finding a change that was “significant” but done to pure luck and nothing to do with chocolate.

They also can simply ignore other measurements that showed nothing or showed the opposite.

This is all known as “P hacking”.

(Remember David Sackett the Father of EBM from last week. This kind of crap is why he wrote the spoof article).

Measure as much as you can, repeat the experiment if needed, then cherry pick the bits you want that are “significant” and boom you get a headline.

Plus, we should also note the tiny number of people in each group, just 5.

A woman’s weight can fluctuate by five pounds during her menstrual cycle, hence why you need a larger group to iron out those fluctuations.

So he submitted his paper, entitled “Chocolate with high cocoa content as a weight-loss accelerator”, to 20 journals and waited.

Remember last week we talked about the sham of peer review?

The review can take months if done properly, and it does need to be done. We are not saying it’s worthless, just not a guarantee that the paper is “fact” or “true”.

Within 24 hours, they had received multiple acceptance offers.

They ultimately went with the International Archives of Medicine, sounds pretty impressive, doesn’t it?

Two weeks later, it was published with zero changes to the paper, none.

They then went out to hook some journalists.

They distilled down the results, so the journalists didn’t even have to read the paper. Who has time for that?!?

In the summary, they neglected to mention the tiny number of subjects in the study or the meaningless difference between groups, or the fact they didn’t check what the groups were eating at all, no calories counting, only 3 week study I could go on, but you get the idea.

The results?


I could go on but you get the idea.

Hook, line and lazy sinker.

Even the so-called “fact checkers” got it wrong.

Shape magazine employed one of these faithful and ever honest, truth seekers, who double checked a few sentences from the press release and the author’s names and job done.

The facts were confirmed!

Next time you see something labeled as “misinformation” or “partly false” (could that also be labelled partly true?), remember these people have no idea.

When it comes to nutrition, just go back to basics.

Eat as much real/whole food as you can.

Aim for a nutrient dense diet with foods that your immune system tolerates.

Think mediterranean, think paleo-esq.

Diets are not a religion, they are a means to an end.

If one diet is not working, change it.

You are unique, but we are all descended from the same genetics, so there are common themes to stick to.

I love the book “deep nutrition”. Check it out.

It is based on the work of Weston Price DDS who travelled to remote tribes in the early 1900’s and found common themes of healthy groups.

As soon as they introduced modern processed foods, things always slowly went downhill.

Note, wherever meat/offal is available it is ALWAYS eaten, same with raw dairy.

They are high density nutrition and are prized in these tribes.

The Maasai warriors love a blood and raw milk, milkshake!

Do not try to deny evolution, your biological system does not give a flying f**k about your opinion or theory.

At the end of the day, you are a series of pathways inside a collection of cells.

You need building blocks – fat/protein.

And co-factors to turn on the enzymes – vitamin/minerals.

If you do not have these because you don’t eat them or cannot absorb them, you are going to develop symptoms as your system starts to malfunction.

If you have too much bad stuff.

These poison you and create inflammation, they also slow down the enzymes (that is how chemical warfare works).

Too much bad stuff, not enough good stuff = illness.


Do to make nutrition harder than it needs to be.