N of 1 Trial

Wikipedia

An N-of-1 (N=1) trial is a multiple crossover clinical trial, conducted in a single patient...This design, especially if including blinding and wash-out periods, can be effective in confirming causality. 

A LOT of health experiments don't really determine causality. Dietary studies are often self reported and basically anecdotal on a large scale under circumstances where there are many confounding factors.

They frequently boil down to correlation does not prove causation.

But I think you can as an individual come up with a hypothesis and check if it actually works that way or not FOR YOU.

And that can be life changing, even life saving.