Conformance Score & Data Quality Score

Grades a bundle against the official FHIR specification, or against a named Implementation Guide (US Core, mCODE, MedMorph, PH Library, Da Vinci PDex Plan Net) if one applies. The score reflects how much of the bundle came back clean, with a smaller effect from warnings than from errors.

The breakdown is organized into five categories — Structure, Completeness, Terminology, Profiles, and References — so a low score points at what kind of problem it is, not just that one exists.

Every result traces back to a specific, readable rule rather than an opaque judgment call, and the rules themselves were built by running real FHIR bundles through the validator and checking the output against real-world Implementation Guide requirements — not guessed at.

See it on the Validation tab →

A separate question from conformance — a bundle can be perfectly valid FHIR and still contain data that shouldn't be trusted: a lab value copied across dozens of visits, the same reading counted twice, or a date decades out of line with the rest of a patient's record. None of that fails FHIR validation, because none of it is a structural error.

The Data Quality score checks a bundle's clinical data for exactly these kinds of issues — duplication, repetition, internal conflicts, and outlier dates — and reports what it found in plain language, separately from whatever the Conformance Score says.

See it on the Render tab →