This site publishes numbers. Sometimes numbers are wrong. When ours are, we say so in public rather than fixing them silently, because a site that edits its mistakes away is asking you to trust it on faith.
What we correct
- A trial count that does not match the registry
- A trial attributed to the wrong compound
- An approval status stated incorrectly
- A marketed claim we described inaccurately
- Anything stated with more confidence than the record supports
How to report one
Email hello@whatsactuallyknown.com with the page and what you believe is wrong. A link to the registry record is the fastest way to get it fixed. We would rather hear from you than be wrong in public.
What happens next
- If the data is wrong, the page is fixed and the correction is listed here with the date
- If the data is right, we will explain why — and if our wording caused the confusion, we will change the wording
- Substantive corrections are noted in the next newsletter issue
Known limitations
These are not errors, but they are things a careful reader should know:
- We query specific registries. Studies run outside the systems we check, or never registered at all, will not appear. "Zero trials" means zero in the registries we track.
- Compound names are messy. The same molecule is sold under several names, and fragments are marketed under the name of the full protein. We spot-check that trial records actually name the compound, but the matching is not perfect.
- Registries go stale. A study marked "unknown status" often means nobody updated the record, not that anything specific happened.
- This is automated aggregation. No physician reviews these pages.
Published corrections
None yet. This section will list every correction with its date, what was wrong, and what changed. An empty list here is not a claim of perfection — the site is new.