Corrections and updates
We will get things wrong. Guidance moves, a source gets retracted, or we read a passage more confidently than it deserved. What matters is what happens next, and that's what this page is for.
Tell us
Email support@gyna.co with "correction" in the subject line.
The link to the page and the sentence you're questioning is enough. If you have a source that says something different, send that too, and you'll have saved us the slowest part.
You don't need an account, and you don't need to be a customer. If you're a clinician and you spot a bad statement on one of these pages, you're doing us a favor and we'd like to hear from you.
What happens next
Reports get sorted by how much harm the mistake could do, and that decides how fast we move.
Safety
A statement could lead someone toward harm.
The page comes down, or gets a warning on it, straight away. That happens before we've worked out what the right wording is.
Material
A statement is wrong or unsupported, without being a safety problem.
We mark it on the page while we work out the correction, so you're not reading it as though we still stand behind it.
Minor
A typo, a dead link, or formatting that doesn't change the meaning.
We fix it in place.
A safety report is the one case where we'll change a live page before we've finished thinking. We can take a statement down and put it back properly. We can't unsay it to someone who acted on it in the meantime.
Corrections are shown, not made quietly
When we correct a published statement, the page says so, with the date and what changed. Editing the words and moving on would leave you unable to tell a page we've checked from a page we've never revisited.
If a statement turns out to be wrong badly enough that it shouldn't exist, we withdraw it. The page stays at the same address and says it was withdrawn, rather than becoming a dead link. Anyone who read it, or linked to it, can find out what happened to it.
The same applies when the source moves rather than us. Guidance gets revised, and a statement that was accurate against the 2019 version of a document is not automatically accurate against the 2026 one. Those trigger a re-check, and a re-check can end in a correction.
Two things we won't do
- Change a published statement without a record of the change.
- Describe a statement as reviewed by someone who hasn't reviewed it. Right now that means every statement in the library is labeled as machine-verified with clinical review pending, and none of them is labeled as anything better.
How a statement gets published in the first place is on the methods page.