Skip to content

How dQMGrid meets the CMS opportunity

CMS QualTech 2026 · Opportunity Area 3

CMS invited industry to propose a measure calculator — or showcase an existing one — built on a framework accessible to providers, facilities and CMS, and designed specifically for the approximately 70 dQMs converted from eCQMs. This page takes that request apart, element by element, and shows how dQMGrid answers each one and how far along it is.

Status key: Today demonstrated now · Next the first release · Planned a later phase

What CMS asked for How dQMGrid answers it Status
A measure calculator, or an existing one showcased A working prototype, not a concept: four measures on two engines, reproducing every published test case for those measures. See Prototype results. Today
Built on a framework accessible to providers, facilities and CMS Open source — proposed licensing: code under Apache 2.0, rule artifacts under CC0 1.0, documentation under CC BY 4.0; to be finalized after our upcoming discussions with CMS. Ships as a container that an organization runs inside its own data boundary; a registry or ACO can run it for its members; CMS can run a reference instance. Nextpublic release
Designed specifically for the ~70 dQMs converted from eCQMs Loads the published packages from the eCQI Resource Center at their pinned versions. All 74 published packages — 70 converted eCQMs plus 2 hybrid and 2 NHSN measures — are tracked on the coverage page. Today4Next10Planned~70
As defined and published through the eCQI Resource Center Runs the packages as published: no new measure language, no re-implemented logic. Where a package will not load as published, each change is disclosed in a manifest that travels with every result. Today
What CMS asked for How dQMGrid answers it Status
Align with digital-first quality measurement FHIR-native end to end: FHIR R4 data in, CQL measures evaluated on FHIR Clinical Reasoning and DEQM, standard FHIR MeasureReports out. Today
Advance interoperability Bridges the gap between the US Core profiles certified EHRs export and the QI-Core profiles the measures retrieve by, with rules published as standard FHIR artifacts that any implementer can reuse. TodayrulesNextsigned sets
Lower the barrier to entry for providers and facilities The duplicated part of every ACO’s and registry’s stack — the per-vendor adapter layer — becomes shared, open artifacts. One signed vendor-tier rule reaches every site on that product, including one that could never have written it. Next
Close the gap between measurement, reporting and scoring Two counting modes read the same run: reporting mode counts only what the source coded plus repairs and signed rules; quality-improvement mode also counts unsigned adjustments, marked. The difference between them is published as a bias report. TodaymodesNextbias report
Support true quality improvement The account turns every uncounted patient into a finding — which retrieve was empty, which element was missing, and who owns the fix — so a gap can be closed rather than just measured. TodayaccountNextowner-routed feed
What CMS asked for How dQMGrid answers it Status
A high-performance computing platform dQMGrid is built to scale out: each patient is evaluated independently, so the same container runs on one server or across hundreds of nodes. The project provides the implementation and a scalable deployment model — from one clinic’s server to a national instance — so any operator, including CMS, can run it on its own infrastructure, up to a high-performance environment for national-scale runs. A hosted, multi-tenant instance operated by MDinteractive is possible as a separate service; it is outside the scope of this proposal and would require FedRAMP authorization. See Architecture. Todaysingle-node prototypeNextcontainer imagePlannedscalable deployment model, multi-tenant isolation
Connect directly to EHRs Reads FHIR R4 Bulk Data export output — the interface every EHR certified to ONC criterion 170.315(g)(10) exposes. Run today on a live eClinicalWorks Bulk Data export of 1,339 patients and about 1.8 million resources. Today
…or enterprise data warehouses A warehouse that emits the same NDJSON is indistinguishable to the calculator; no separate connector is needed. Todayby design
Process the structured data requirements inherent to dQMs The Conform stage validates every output against the profile the measure expects using the HL7 FHIR Validator, and declares, repairs or derives elements only through versioned rules. TodayrulesNextvalidator in pipeline
Let providers use their digital data in real time A daily export using the Bulk Data _since parameter re-evaluates only changed patients, so a care gap is visible within one export cycle; a single-patient read re-evaluates one patient on request. Todayby designNextoperational

The opportunity asks for a calculator for the converted measures. These are the limits of what exists today, stated so that nothing on this site is read as more than it is.

  • Four measures run today; about 70 is the target. Ten in the first release; the full converted set in the phase after. We show phases, not dates: the pace depends on funding and partners.
  • Every run to date is ambulatory. The first hospital data source needs inpatient encounter, medication-administration and discharge data that ambulatory exports do not carry, and a facility partner we do not yet have.
  • No rule has yet been clinically signed. Signing is designed; signed rule sets for two ambulatory EHR platforms are an first release deliverable.
  • Some components are designed, not built: the retrieve tracer, the fault-injected test decks and the bias report.
  • A hosted government-facing instance would need FedRAMP authorization, which is not part of the first release.
  • The first release, the full measure set and any reference instance need funding — budget to be scoped with CMS.

See Roadmap for the full plan and Who it is for for how providers, facilities, registries, EHR vendors and CMS would each use it.