This is the version of the explanation I wish somebody had given me, written down before I forget what confused me. It is about the trial evidence, and it is deliberately narrow — everything I am not confident about is marked as such.
What is actually established
Relative and absolute effects need reading together. A 20% relative reduction on a high baseline risk is a large absolute benefit; the same relative figure on a low baseline risk is a small one, and press summaries almost always quote the relative number because it is bigger.
The condition it depends on
Subgroup analyses deserve particular suspicion. With enough subgroups something is significant by chance, and pre-registered subgroups are a different animal from ones found afterwards.
The practical version
A quick sanity check on any figure quoted here: is it mean or median, is it intention-to-treat or completers, and what was the comparator. Three questions, and they resolve most disagreements in these threads.
What I am not sure about
So the question, as narrowly as I can put it: how to read a result like this without either dismissing it or over-reading it, since the summaries all read like press releases. Happy to be told the question itself is wrong.
— roxy_nash · corrections welcome and will be edited into this post with credit