MII Implementation Guide Microbiology - Local Development build (v2027.0.0-ballot2) built by the FHIR (HL7® FHIR® Standard) Build Tools. See the Directory of published versions

Interpretation

A susceptibility result carries two different statements: what was measured, and how that measurement is assessed. The module keeps them apart.

Measured value and assessment

Observation.value[x] carries the measurement — a minimum inhibitory concentration in mg/L, or an inhibition zone diameter in mm.

Observation.interpretation carries the assessment: the category the measurement falls into, together with the norm it was derived from.

Not every laboratory measures. Where only the category is available, it is given as a CodeableConcept in value[x] — the European data model settles this explicitly, and an Observation with no value at all would be the more unusual shape. Two invariants keep that from becoming ambiguous:

• empfindlichkeit-kategorie-braucht-interpretation — a category in value[x] requires an interpretation, because the norm hangs there and a category without its norm says nothing.
• empfindlichkeit-kategorie-stimmt-mit-interpretation — where both are given, every code of the value must also appear among the interpretation codes.

So a consumer reads the category from interpretation in every case, and value[x] tells it whether a measurement stands behind it.

The reverse — a measured value without a category — is flagged as a warning by empfindlichkeit-messwert-sollte-bewertet-sein rather than rejected. A bare MIC leaves the assessment to the consumer, who would need the breakpoint tables for it. But a requirement would be wrong: for some organism and agent combinations no breakpoints are defined, and then there is no category to give.

The categories

Code	Meaning
S	Susceptible
I	Intermediate
SDD	Susceptible-dose dependent
R	Resistant
NS	Non-susceptible

EUCAST redefined I in 2019 as susceptible, increased exposure. Where that reading is meant, SNOMED CT 1306583007 states it explicitly rather than leaving I ambiguous.

The norm the assessment rests on

A category is meaningless without the breakpoints it was derived from: the same MIC is susceptible under one norm and resistant under another, and breakpoints change from year to year.

The norm therefore hangs on the interpretation, not on the Observation — Observation.interpretation.extension, mandatory. Because interpretation is 0..*, one measured MIC can carry a EUCAST and a CLSI category side by side, each with its own norm. Hanging the norm on the Observation would make that impossible.

The norm is coded from the module's own CodeSystem: EUCAST, CLSI, Andere.

Predicted susceptibility is a different statement

Susceptibility inferred from a resistance mechanism — a gene, a mutation or a protein found, therefore an expected resistance — is not this profile. It has no measured value and belongs to Predicted Susceptibility, whose result is the expected category itself. The inference is deliberately technology-open: a PBP2a protein detected by lateral flow supports it exactly as a mecA gene found by PCR does. See Profile Selection and Delimitation.

The same boundary applies to the targeted detection of a resistant organism: an MRSA or VRE detection states which organism was found, not how susceptible it is, and belongs to the determination or culture profiles.

Other assessments in this module

interpretation is not limited to susceptibility. Colony counts carry a semiquantitative assessment of the measured value, and the avidity test carries low, equivocal or high — L and H from the value set the parent binds, and for the middle the same SNOMED code this guide uses for a borderline finding elsewhere. The pattern is the same in each case: the value stays measured, the assessment stays separate.