Queued to go out tomorrow morning.
The short answer is that it depends on the form, the dose and what you are actually trying to change. The long answer is below, with the trials that support it and, where they exist, the ones that do not.
What the trials measured
Most of the studies here are small - thirty to eighty participants - and run for eight weeks or less. That is enough to detect a large effect and nowhere near enough to rule out a small one, which is worth holding in mind before reading any single result as settled. Where a meta-analysis exists we have cited it in preference to the individual trials, and where the meta-analysis disagrees with the popular summary we have said so rather than quietly picking the flattering one. Several of the most-shared findings in this area come from a single trial that has never been replicated, and a reader deserves to know which ones.
Written for general information. It is not medical advice, and it is not a substitute for speaking to a clinician about your own health.
