Reviewed terminology & panel equivalence

The source decision contract solia.clinical.equivalence.v1 describes reviewed terminology, not order or release authority. Compare laboratory codes and dictionary LOINC/UCUM identifiers with source/target fingerprints, kind, specimen, method and the complete component multiset. Synonyms and names may suggest candidates but cannot grant equivalence or published coverage.

• exact_reviewed — current review over nonempty known components, identical test/panel kind and complete component multiset including code, unit, specimen, method and ROLE.
• reviewed_qualified_alternative — current reviewer decision with explicit nonblank constraints and safe required coverage; it cannot bypass disqualifying tenant, specimen/method, missing-code or stale-revision refusals.
• partial_coverage — overlap is recorded, not sufficient to grant equivalence.
• specimen_method_mismatch — specimen or analytical method differs; refuse automatic substitution.
• unreviewed_suggestion — names/codes suggest a draft candidate only.
• unsupported — unknown, empty, duplicate or structurally incompatible components; no equivalence.

• Same required analytes but different optional/conditional/reflex members are not exact. Required-only set equality loses clinically meaningful composition.
• A single test and a panel with the same display name are not exact. Empty panels and duplicate required components refuse.
• Altered units or component roles change the multiset. Output units cannot discriminate an ambiguous source code; do not guess conversions.
• Cross-tenant comparison, changed source/target revision and missing core codes refuse. Overlapping biomarkers alone do not justify substitution.

1. Bind the decision to the current source and target laboratory dictionary versions and fingerprints.
2. Record authorized reviewer identity, review version, source provenance, decision and any justified constraints. No synthetic test supplies a real reviewer.
3. After composition, specimen, method or code changes, invalidate stale evidence and review a new version; retain the earlier reviewed snapshot.
4. Verify the full positive/negative matrix before using the decision as documentation of clinical coverage. Existing canonical Program/order/release checks still independently apply.