How to read a study about a pepper

Five questions that say a lot

1. Who was it tested on? Cells in a dish, lab animals or people? An effect in a test tube says nothing yet about what happens in a person.
2. How many people took part? With twenty participants, chance can play a big role. The bigger the group, the more reliable the outcome.
3. Was there a placebo group? In a well designed study, some people get a dummy, and nobody knows who gets what. That's called double blind and placebo controlled.
4. What amount, and for how long? A study with a high dose over two weeks says little about a different amount over a longer period.
5. Who paid for it? Money from a manufacturer doesn't automatically make a study wrong. But it's good to know.

One study isn't proof

Science works with piles of research. A systematic review or meta-analysis lines up all the good studies on one question. That says more than a single study, however exciting the outcome sounds.

Also watch the difference between relative and absolute. "50% lower risk" sounds big, but going from 2 in 100 to 1 in 100 is also 50% lower.

Press release or publication?

News stories are often based on a press release. The actual study sits in a scientific journal, with an abstract anyone can read. That usually says exactly how it was set up.

And what does it say about you?

A study is about a group, not about you. What you personally need is something for your GP or a dietitian. That's also why we don't give advice, only factual information.