CMS Hospital Quality Reporting FHIR Implementation Guide
0.1.0 - ci-build
CMS Hospital Quality Reporting FHIR Implementation Guide
0.1.0 - ci-build
CMS Hospital Quality Reporting FHIR Implementation Guide - Local Development build (v0.1.0) built by the FHIR (HL7® FHIR® Standard) Build Tools. See the Directory of published versions
This guide is a deliberately unfinished 0.1.0 concept review, not a near-final specification. The September 2026 HL7 FHIR Connectathon is IG validation, not a product demonstration: the goal is pressure-testing these concepts before they harden, ahead of a more complete guide and fuller testing in a subsequent pass.
| # | Criterion | Pass looks like |
|---|---|---|
| 1 | Submission | Submitter successfully sends a MeasureReport for one of this Connectathon's three either/or test measures (CMS71, CMS506, or EXMConnectathonSept2026Simple — synthetic patient) to the Receiver via $submit-data |
| 2 | Acceptance | Receiver accepts a conformant submission and returns a clear status response |
| 3 | Results retrieval | Submitter can retrieve submission status and results after the fact, without relying on the original synchronous response |
| 4 | Rejection and correction | An intentionally invalid submission receives an error specific enough for a Submitter to plausibly correct it |
| 5 | Scale | Runs cleanly across one to three test cases, drawn from any of the three either/or test measures, under 100 synthetic patients |
| 6 | Participation | At least three active participants exercise this capability in some role |
Three measures are available for this Connectathon's testing, presented as either/or options
with no primary or stretch tiering — participants may exercise any one (or more) of them: CMS71
(STK-3, Anticoagulation Therapy for Atrial Fibrillation/Flutter), CMS506 (Safe Use of Opioids –
Concurrent Prescribing), and EXMConnectathonSept2026Simple
(https://madie.cms.gov/Measure/EXMConnectathonSept2026Simple). None is a structural dependency of
this guide, and HQRSubmissionMeasureReport itself is measure-independent — it does not encode
anything specific to any one measure. This section explains why these three, specifically for this
Connectathon.
A cross-measure analysis assessed seventeen candidate eCQMs against US Core 6.1.0 through 9.0.0
for data-element coverage. CMS71 and CMS506 were two of only three candidates found with zero
unprofiled elements across those US Core versions. Neither depends on
MedicationAdministration, the Present-on-Admission extension, AdverseEvent, or any unprofiled
screening or assessment instrument — the gap categories that make most of the other candidates
harder to exercise using only US Core-profiled data. Both are process measures, which keeps
population logic simpler — no risk-adjustment variables and no negation-pattern forks. One caveat
applies to both: the Patient race, ethnicity, and sex SDE extensions are "Additional USCDI," not
Mandatory or Must Support, for either of them — worth noting, not a blocker.
EXMConnectathonSept2026Simple is not a real CMS eCQM and has no CMS number — it is a MADiE-authored
measure built specifically for this Connectathon (draft/experimental, proportion scoring), so
participants can exercise DEQM exchange mechanics without the noise of real measure-calculation
complexity. Unlike CMS71/CMS506, its own CQL criteria are deliberately trivial (Initial
Population/Denominator both true, Numerator checking only Patient.sex) rather than a full
eCQM's worth of data elements — see the Data Element Testing Guide
for its own single-element checklist alongside CMS71/CMS506's.
CMS832 (HH-AKI, Hospital Harm – Acute Kidney Injury) is not part of this guide's current test set. Unlike CMS71/CMS506, its data-element coverage against QI-Core/US Core has not been independently verified, and it should not be assumed to pass the same gap check without going through it — see Open Questions.
Invited-pull ($bulk-submit-data ping + Bulk $export) is presented as a concept for discussion
— see Data Flow and Open Questions — not as a
live-tested commitment for this event. A strong objection to the concept is a successful outcome,
not a failure, since the goal is identifying where it breaks before it hardens further.
This event succeeds if push submission and results retrieval run cleanly for at least one participant pair, the 0.1.0 concepts get real engagement and pushback, and the resulting feedback is specific enough to turn into concrete revisions — not if every criterion above passes.