Reading a study without a degree

The most useful thing in this course isn’t a protein number. It’s the ability to look at the next confident nutrition claim and work out, in about two minutes, whether it deserves your attention.

Question one: who did they study?

Twelve trained young men is not the same population as you, and rodents are not people. Findings in mice are hypotheses about humans, not conclusions about them. This single question deflates a large share of headlines.

Question two: how many, and for how long?

A study with fifteen participants over four weeks can suggest something. It cannot establish it. Small and short is where noise lives, and noise is what generates surprising results worth writing about.

Question three: what did they compare it to?

“Group that took the supplement improved” is meaningless without “compared to a group that didn’t”. And the comparison has to be fair: comparing a supplement to nothing at all, in people who knew which group they were in, measures enthusiasm as much as biology.

Question four: what did they actually measure?

Watch for swapped outcomes. A study that measures a blood marker is not a study that measures muscle, and a study measuring muscle is not one measuring strength. The further the measured thing is from the thing you care about, the weaker the claim.

Question five: who paid?

Industry funding does not automatically invalidate a study, and pretending otherwise is lazy. But it is a reason to look harder at questions three and four, because that’s where funded studies tend to be generous with themselves.

Putting it together

Run those five questions on the next protein claim you see. Most will fail at question one or two, and you’ll have spent two minutes rather than an afternoon.

You don’t need to be able to design a study. You need to be able to notice when nobody else did either.

What the numbers usually mean