Data coverage and limitations

Voxsanity is built on public government registries. Those registries are incomplete, and so, therefore, is this site. This page says plainly where the gaps are, so that you are not left assuming a completeness that does not exist.

Last updated: July 2026

If you are making a decision about your own care, treat this site as a starting point for a conversation with your doctor, never as the full picture.

Plain English summaries are written by AI, and we do not read every one

The plain English trial eligibility summaries and results summaries on this site are generated by an AI model from the official registry text. This is the part of the site you should treat with the most care.

No person reviews every summary before it is published. We spot check them on a sample basis, and we have automated checks that block certain kinds of unsafe phrasing. But the honest position is that a summary can go live on this site without any human having read it. AI models make mistakes, including confident ones.

Every AI generated summary is labelled as such, directly above the text, so you always know what you are reading. The original registry wording is always linked. If a summary matters to a decision you are making, read the original. If you find a summary that is wrong, use the "Report an issue" link on the page and we will correct it. More detail is in our terms of use.

Conditions: 103, not all of them

Voxsanity currently covers 103 conditions. There are thousands of recognised medical conditions. If yours is not among the 103, that is not a signal that no research exists. It means we have not built a page for it yet.

We intend to expand this list. We are not going to promise a date, because we would only be guessing.

Trial registries: two of them, not all of them

Trial data comes from ClinicalTrials.gov, the United States registry and the largest in the world, and from ISRCTN, a UK based international registry. Together these are the overwhelming majority of what we hold.

We do not currently include:

The WHO International Clinical Trials Registry Platform (ICTRP). This is a deliberate, permanent exclusion, not an oversight.
The Australian New Zealand Clinical Trials Registry (ANZCTR). We have not confirmed in writing that ANZCTR's terms permit the reuse we would need, so we do not sync it. For an Australian reader this is the most important gap on this page: an Australian trial registered only with ANZCTR, and not also with ClinicalTrials.gov, does not appear on this site at all. Many Australian trials register with both. Some do not.

If you are looking for a trial in Australia, search the ANZCTR directly as well as using this site. Do not treat our list as the complete set of Australian trials.

Australian access data: largely not there yet

This is where the site is furthest from where we want it to be.

PBS listing status is not shown for any medicine. The field exists in our database and is empty for every medicine we track. We are not going to show a "PBS listed" or "not on the PBS" indicator until it is real, because a wrong answer here is worse than no answer.
The lag between FDA approval and PBS listing is not shown, for the same reason. It is one of the things we most want to publish. It is not ready.
TGA approval dates are not shown. The public data source we use for medicine approvals does not return them.
TGA medicine shortage information is not shown. We collect it, but TGA material is restricted for commercial reuse and we do not yet have written permission from the TGA to republish it. Until we do, it stays unpublished. For current shortages, use the TGA's own Medicine Shortage Reports Database.

Research funding: United States only

Research funding figures come from the US National Institutes of Health. They tell you what the United States is spending, which is a real signal of where global research attention is going. It is not Australian funding.

Australian NHMRC funding is not included. Our database has a field for it, and that field is empty for all 103 conditions. Nothing on this site currently reflects Australian research investment, and you should not read the funding figures as though it did.

Rare disease and orphan drug data: minimal

Orphan drug designation is effectively not covered. We check it for a single catch all "rare disease" category, and for no individual condition. Do not read the absence of an orphan drug indicator on a condition page as meaning anything at all.
Orphanet reference data is matched for 17 of the 103 conditions, and even where it is matched we do not yet hold usable prevalence figures.

Two things to hold in mind about all of the above

Registry data is only as current as the people who file it. A trial's status changes when the sponsor updates the registry, which can be long after the reality changed. A trial listed as recruiting may have stopped recruiting. Verify directly with the trial site before acting.

Absence of data is not evidence of absence. If a condition page shows no trials, no approvals, or no funding, the honest reading is "the registries we use do not show us any", not "there are none".

If you find something on this site that is wrong, tell us. Every trial, condition, and medicine page has a "Report an issue" link, and we aim to respond within 48 hours.

Related: our data sources, our methodology, and our editorial policy.