When the ImOnFHIR validator flags something, it also tries to tell you — in plain language — what it means, why a payer cares, and what to change. Those explanations are a real convenience, but they come with a rule attached: they show their provenance, and they never overstate it.
There's an interactive index of these explanations, and they also appear next to findings in the validator itself.
Two kinds of finding
Not every finding is the same kind of claim:
- Structural / profile conformance — "this doesn't match the FHIR profile." That is the standard talking, and the authoritative check is the Inferno / HL7 validator that ImOnFHIR runs underneath.
- Semantic / operational — "a payer probably can't process this as-is." That is our product's interpretation of how prior auth works operationally. It is useful, and it is not the PAS specification.
The plain-language explanations attach to the second kind. That's why they carry a badge.
The provenance badge
Every explanation shows a badge:
- Unverified — product interpretation. The honest default. It means a human hasn't yet reviewed this specific explanation, so you should treat it as orientation and verify against the actual rule or your payer's guidance. Most explanations sit here on purpose.
- Reviewed. A person has checked this explanation. Even then, it's an interpretation, not the standard.
We default to unverified deliberately. It is easy to build software that quietly claims "verified" for machine-generated prose — and that is exactly the wrong move on a feature whose whole value is trust. The rule we hold ourselves to is sell provenance, never authority: tell you what we think and how sure we are, and let the standard and your counsel be the authority.
How to use them
Read the explanation to understand the finding fast. Then, before you act on anything that matters, check the cited rule, the profile, or your payer's companion guide. An explanation is a good map; it is not the territory.