A federal policy and industry roadmap for trusted, scalable digital quality measurement for ACOs
Accountable Care Organizations (ACOs) are responsible for improving care and demonstrating value across populations that span multiple providers, practices, hospitals, electronic health record (EHR) systems, and care settings. Yet quality reporting remains heavily dependent on EHR-centered processes that can be labor-intensive, costly, difficult to scale, and disconnected from the infrastructure ACOs already use to manage population health.
This new report from Leavitt Partners Supporting Real-Time Quality Insights through HL7® FHIR® recommends that the Centers for Medicare & Medicaid Services (CMS) allow qualified ACOs to use standards-based HL7 Fast Healthcare Interoperability Resources (FHIR) infrastructure for Medicare quality reporting. While CMS recognizes this need and is moving towards this approach, regulatory changes described here are necessary to make this architecture possible without requiring it
Under this approach, EHRs would remain essential sources of clinical information, while ACOs could use population-level FHIR infrastructure to aggregate, validate, calculate, and report quality measures. The same infrastructure could also support care coordination, population health management, quality improvement, clinical decision support, and other authorized uses.
Key findings
- Quality reporting costs are concentrated in data collection, validation, integration, and maintenance, not simply in calculating measures.
- CMS has positioned electronic clinical quality measures (eCQMs) as the pathway to digital quality measurement. The flaw in the eCQM approach is the reliance on quality-specific data architecture and submission from EHRs.
- ACOs often manage information from multiple EHRs, hospitals, health information exchanges, laboratories, claims systems, registries, and population health platforms.
- One large ACO reported connecting to 171 unique EHR instances to obtain quality data.
- While eCQM costs grow with each EHR instance and every reporting program, costs for FHIR-based quality reporting do not. One ACO estimated annual reporting costs $950,000 for eCQM reporting compared with $405,000 for FHIR-based digital quality measure reporting. This is because the same underlying infrastructure used for MSSP reporting meets the ACO’s quality reporting obligations under value-based contracts with multiple Medicare Advantage plans.
- FHIR-based reporting can create a reusable data and analytics foundation for quality improvement and value-based care contracts.
- FHIR conformance alone does not guarantee that data are complete, accurate, timely, or usable for quality measurement.
- Population-level APIs must be tested for reliability and performance across large patient panels, concurrent requests, authentication requirements, and sustained operational demand.
The proposed three-part framework
The report recommends that FHIR endpoints used for ACO quality reporting be evaluated across three core dimensions.
1. FHIR conformance
Endpoints should demonstrate that they implement the required FHIR resources, profiles, interactions, authorization mechanisms, implementation guides, and other technical requirements.
The ONC Inferno testing framework provides an existing foundation for evaluating FHIR implementations, including standardized API capabilities such as SMART App Launch, US Core, Bulk Data, single-patient APIs, and multi-patient APIs.
2. Data quality
FHIR conformance does not establish whether the underlying information is fit for a specific quality-measurement purpose. Data may conform technically while still being incomplete, stale, inconsistent, or coded in ways that prevent reliable measurement.
The Patient Information Quality Improvement (PIQI) Framework could help CMS and ACOs evaluate whether data are complete, accurate, timely, consistent, and reliable enough to support digital quality reporting.
3. API performance
An endpoint that works for a single patient record may not perform adequately for population health. Qualification should therefore include measurable requirements for response time, throughput, availability, error rates, retry behavior, concurrent requests, and sustained performance.
These benchmarks should initially be tested and validated in real-world ACO environments before being considered for regulation.
Recommended policy actions for CMS
The report recommends that CMS:
- Redefine certified EHR technology (CEHRT) to include any certified FHIR-based API
- Create a new Medicare digital quality measures (Medicare DQMs) data collection type that uses standardized FHIR-based APIs
- Allow ACOs to meet CEHRT use requirements by reporting conformance, data quality, and performance results for the FHIR APIs they use for quality reporting
- Relieve the cost of cloud computing necessary for Medicare DQMs
Why this matters
ACOs are already accountable for managing populations and reporting quality results. The missing piece is regulatory alignment that allows those responsibilities to be supported by modern, standards-based infrastructure.
Modern, standards-based data architecture could preserve the EHR’s role as the source of clinical information while giving ACOs the infrastructure needed to aggregate data from multiple systems, validate its quality, calculate measures, submit auditable results, and use quality insights to improve care.
Frequently asked questions
What is the report’s main recommendation?
The report recommends that CMS allow qualified ACOs to use standards-based FHIR infrastructure for Medicare quality reporting rather than requiring reporting exclusively through individual EHR systems.
What are the three requirements for a qualified FHIR endpoint?
The endpoint should demonstrate FHIR conformance, meet data-quality requirements, and perform reliably at population scale.
What role should EHRs play?
EHRs should remain essential sources of clinical information. ACOs should also be able to use population-level infrastructure to aggregate, validate, calculate, and report quality information.
How could FHIR-based reporting reduce costs?
In one ACO estimate, annual FHIR-based digital quality measure reporting was projected to cost $405,000 compared with $950,000 for eCQM reporting. The report presents this as an illustrative estimate, not a universal cost guarantee.
Does FHIR conformance guarantee high-quality data?
No. An API can conform to FHIR standards while returning data that are incomplete, inaccurate, stale, inconsistently coded, or otherwise insufficient for reliable quality measurement.
What additional issues remain?
The report also identifies unresolved issues involving measure specifications, unstructured clinical documentation, measure-calculation costs, ongoing connectivity operations, cross-program measure alignment, and the suitability of current measure sets for specialty practices.
Featured Authors
This report was also written by Wakely Consulting Group’s Daniela Simpson.