How we verify providers

PsyAccess lists only providers we can trace to an official source. Every record shows which registry it was checked against, a link to that source, and when it was last verified.

Sources of truth, by country

Wherever a public registry or payer list exists for a category, it is the primary source and the record is labelled Registry-verified. The second level, Verified: two sources, exists only for categories with no public registry in their country:

The dual-source rule

Every record must be supported by at least two independent sources: the journalistic standard applied to a directory. Wherever a public state registry exists, one of the two sources must be that registry (an institutional register, an official decree, a state trial registry such as CTIS or ClinicalTrials.gov, or the treating physician's personal licence in a public state register). A record that falls to a single source is not flagged or downgraded: it is removed from the catalog until verification succeeds.

Verification levels

• Registry-verified : two independent sources, at least one of them a public state registry or official public-law register; the registry entry and source links are shown on the record.
• Verified: two sources : used only where no public registry covers the category in that country: two independent non-government sources, or a licence number the clinic itself publishes on its website (cited verbatim, pending the registry cross-check).

There is no lower level. One source: including a provider's own website alone: is never enough to be listed.

A note on Swiss psychedelic-assisted therapy (PAT): individual treatment permits are issued by the FOPH (BAG) per patient (Art. 8 Abs. 5 BetmG) and are not public. PAT listings therefore rest on the SÄPT PAT therapist list, a professional association register, cross-checked against the practitioner's medical licence in MedReg; a listing does not by itself indicate a current FOPH permit.

Update cadence

1. Registry-verified records are re-checked at least every 90 days.
2. Designated-center countries (e.g. Poland, Netherlands) are re-checked whenever the national payer publishes an updated center list.
3. Price, waiting-time and reimbursement notes carry their own "last verified" date and are refreshed on contact with the provider or when a public source changes.
4. Corrections reported by providers or patients are reviewed within 5 working days: see the contact section for how to report one or claim your listing.

The catalog this methodology produces is also published as an open dataset (CC BY 4.0, JSON + CSV, public fields only): PsyAccess Open Dataset.

Quality assurance: the blind audit

On top of the rolling 90-day re-verification, we run a quarterly blind audit of the whole catalog. A seeded pseudorandom sample (about a fifth of all records; the seed is published, so the draw is reproducible) is re-checked field-blind: the checker receives only the record's name, country and its cited sources, never the stored values, and re-collects the facts from the primary sources — registry documents are text-extracted and searched, trial records are checked against the live trial-registry status, provider sites are re-crawled when a cited page has moved.

An error is any material discrepancy: a field contradicting the official source, a dead cited source with no replacement, a service the provider's own sources no longer state, or a closed provider. Sources that block automated checks are disclosed as unverifiable and excluded from the error denominator. Every error found is corrected in the catalog the same day, and the audit protocol with per-error findings is published in the site's open repository. The latest audit (2026-10-01) found an error rate of 4.8% (5 of 104 verifiable records: one spelling discrepancy against a registry, four records stale against a moving market — all corrected or removed same day).

Who checks our work

Frequently asked questions