June 28, 2021
Chiquita Brooks-LaSure, Administrator
Centers for Medicare & Medicaid Services
7500 Security Blvd.
Baltimore MD 21244
Dear Administrator Brooks-LaSure:
Thank you for the opportunity to comment on the CMS proposed rule for fiscal year 2022 Medicare Hospital Inpatient Prospective Payment System and Long Term Care Hospital. The National Committee for Quality Assurance (NCQA) will comment specifically on the RFI to improve the nation’s quality measurement infrastructure through a move to all digital quality measures (dQM) by 2025 (Section IX, pp. 480–485).
As an independent, not-for-profit organization dedicated to improving health care quality through measurement, transparency and accountability, NCQA is a national leader in quality oversight and a pioneer in digital quality measurement. As the market leader in healthcare quality, we often work with a wide variety of stakeholders to drive alignment across the health care system—and we are eager to do so in the move to dQMs. Leveraging our strengths as a trusted third party, we are committed to helping organizations to navigate the challenges associated with transitioning to a digital future. Our mission to improve the quality of health for all Americans, with an intentional focus on health equity and support for meaningful value-based payment models, propels our daily work.
NCQA is proud to share the following steps we have taken (and the steps we plan to take) to drive quality measurement toward a digital future—we believe many of these steps align closely with the vision described in the RFI and should contribute to CMS’s goal of fully digital reporting by 2025:
- We are evolving our measures portfolio (including the HEDIS® measures that underpin the majority of value-based payment initiatives throughout health care) to be “digital first” and to provide the ability to measure and aggregate quality data across all levels of the delivery system.
- We are working to deliver the next generation of digital measures as configurable, modular software applications, accessed and integrated via the Fast Healthcare Interoperability Resources (FHIR) and related standards-based application programming interfaces (API), in line with the CMS definition of dQMs. We will support multiple measures in addition to HEDIS and use cases (not only reporting) with these capabilities.
- We are evolving our role in measures development, consensus mechanisms, measurement program administration, data quality and validation, and quality improvement to support the capabilities and ecosystem referenced in the RFI.
- We continue to develop patient-reported outcome measures that can be leveraged to allow programs such as Medicare to more easily measure, incorporate and learn from a patient’s progress toward achieving personal goals.
- We will build measures to interact with clinical support guidelines and return real-time insights.
NCQA supports CMS’s proposed definition of dQMs as “software that processes digital data to produce a measure or measure scores.” We believe the definition should include the potential for dQMs to be developed in a way that allows components of the software package to be used for a variety of use cases.
dQMs should be viewed and defined like the internet protocols (IP) that are the backbone of the internet. IPs enable computers to communicate even though they may use vastly different software and hardware. dQMs need to be standardized in a way that allows end users to interact with quality measures in a consistent, reliable and interoperable format.
We believe deploying dQMs to interface with FHIR-based APIs, and standardizing and collecting data from FHIR-API endpoints, will revolutionize quality measurement in the following ways:
- Reduce burden and provide more actionable, timely feedback to the provider community.
- Atomize health data and enable customization of measure results for quality improvement and reporting.
- Enable a service-oriented architecture for measure development, design and implementation.
- Enhance the ability to assess data quality and integrity for dQMs before, or simultaneously with, execution.
- Promote clinical intervention at evaluation, even if more information will be aggregated or accumulated for measure reporting.
- Enable more seamless data exchange.
We agree that deploying dQMs to interface with FHIR-based APIs is promising for CMS quality reporting programs and that moving to a universal API is the appropriate long-term goal. In the meantime, we need to tackle the challenges of data standardization to support all dQMs now, even as we await development of more APIs. For example, other transmission formats (e.g., flat files) and mechanisms can support the FHIR USCDI standardized data model without having to obtain data through a FHIR API. This approach would allow careful consideration, consensus building and alignment necessary if FHIR APIs are to be the exclusive means for data collection and testing and validating quality measures in the future.
Similarly, to the industry at large, we are excited about the promise of FHIR standards–but there are currently only limited examples of using the standards to share large amounts of data at scale. CMS should proactively fund efforts to test the ability of FHIR-based queries to support the extensive data needed for all dQMs by 2025.
Quality reporting and value-based payment are high-value business cases that support adoption of interoperability and data-sharing standards. With strong leadership from CMS and other federal agencies, this should drive interoperability and accelerate the move to dQMs while eliminating burdensome and costly manual processes.
Use of FHIR for Current eCQMs
NCQA strongly supports the proposal to transition from the Quality Data Model to the FHIR standard for all CMS quality measure reporting. We have already created 22 FHIR-CQL digital HEDIS measures and we intend to advance our digital measures portfolio. NCQA welcomes the opportunity to support CMS in defining and building the future of dQMs based on our experience converting existing HEDIS measures to digital specifications and creating de novo dQMs. The expertise we developed during our journey will benefit CMS programs during the transition to fully digital measurement (e.g., existing claims-based measures may not be good candidates for conversion).
Moving forward, and to support individual and population health interventions, NCQA intends to develop measures that work with clinical decision support systems based on data collected during the course of care. We believe this will enhance the value to practitioners by providing real-time insights rather than retrospective quality measurement results.
Transition to dQMs by 2025
NCQA fully supports and encourages CMS to implement the proposed transition to dQMs by 2025, as outlined in IX.A.4.b. NCQA believes that approach is the most promising to fully realize a learning health system, as we outlined in our recommendations to the new Administration. Digitizing and automating the processes related to quality reporting, management and improvement can result in better measures, better measurement systems and better data—while dramatically reducing burden. By building data collection into clinician workflows, dQMs greatly reduce what is currently additional and separate work to collect quality data, freeing clinicians to focus on patient care.
Great gains have been made in the collection and sharing of health care data since passage of the HITECH provisions in the 2009 stimulus bill, yet billions of dollars in performance-based payments are made each year based on data that is inadequately validated or merely attested to. Strengthening standards and requirements for validation of quality data and the platforms through which it flows will make measurement and identification of high-value care more accurate and will continuously improve the efficacy of future initiatives.
Although we believe interoperability requirements should facilitate exchange of data needed to support a comprehensive dQM strategy, they should extend beyond EHRs, given the necessity for dQMs to be populated by data from multiple sources that might not conform to existing interoperability standards. And while interoperability requirements support data exchange, they do not designate protocols for accuracy, completeness or usability of data for any use case. CMS and its partner agencies will need to effectuate policies to address these issues, to create trust in the solutions described in this section of the RFI.
Redesign quality measures to be self-contained tools
We support CMS’s vision for an end-to-end measure calculation solution that retrieves data from primarily FHIR-based endpoints. It is imperative that a trusted, impartial and independent entity maintains a leading role in operating and validating implementation of a measurement system/solution as described. A key challenge to the proposed modernization is overcoming inertia and facilitating alignment among diverse stakeholders to build, test and implement an end-to-end measures calculation solution. Much of the groundwork for the vision CMS outlined can be achieved through technological advancements, standards maturity and successful demonstration projects—but variability in implementation locks the health care system into an inefficient and costly model.
Specifically, we need a universal, end-to-end solution that invites collaboration of multiple stakeholders across the quality spectrum; that includes accessible, connected tools and resources; that enhances the utility of shareable, computable knowledge to foster alignment across quality measurement processes and deliver real-time, evidence-based decision support informed by recent data. This capability is critically important in situations such as the COVID-19 pandemic, when rapid dissemination of new knowledge, guidance and feedback, based on trustworthy data, can inform care and save lives. It is also a critical link in identifying and closing the gaps in health equity.
NCQA believes a public-private partnership provides the best chance for a sustainable model to accelerate and maintain systemwide adoption, and to validate to regulators, developers and end-users how a uniform set of tools in a secure, trusted environment can reduce burden and health care costs.
CMS can realize the following benefits from a pilot completed by the end of 2022 and provide the evidence for a national model for all digital quality measures by 2025:
- Prove the value of an end-to-end measure calculation solution.
- Prove the distribution of digital measures from the calculation solution to end-users and facilitate seamless data exchange to support measurement results.
- Strengthen and test the quality of data obtained through FHIR-based APIs for quality measurement, validation and reporting.
- Demonstrate a SMART on FHIR approach to digital data collection and measure execution with provisions for clinical decision support systems and real-time quality insights.
Building a Pathway to Data Aggregation in Support of Quality Measurement
We encourage CMS to establish policies for data aggregators (e.g., HIEs, vendors), as well as for EHRs, clinical registries and similar data sources, to support CMS quality reporting. The health care industry has invested significant resources to digitize health data, but much of that data is inconsistent or unstructured, and confidence in its integrity is low. Data aggregators can play an essential role in ensuring that data is accurate, consistent and available for quality measurement, but standards are needed to certify that aggregators maintain and ingest clinical data that is valid and ensures comparability across organizations.
The explosion in electronic clinical data stemming from adoption of EHRs makes it even more essential that CMS evolve technology-enabled approaches to validate and leverage clinical data sources for use in quality and incentive programs. The federal government bases many performance incentives on insufficiently validated data processed through systems that are prone to error, undermining CMS’s goal of rewarding high-quality care. CMS should create policies and procedures to advance a national data validation model that crosses all data sources and levels of aggregation. An end-to-end calculation solution, as described above, cannot be fully realized without a federal model where seamless flow of data is trusted.
In July, after two years of pilot testing, we will launch the NCQA Data Aggregator Validation (DAV) program. Data hubs that earn NCQA Validation become reliable data sources, making data aggregators attractive partners for health plans and trusted sources for state and federal regulators. NCQA checks that aggregators can produce data from Continuity of Care Documents (CCD) in a standardized format that they can then pass to health plans for quality measure reporting.
The DAV program validates electronic data collected and shared with vendors and health care organizations that undergo audit for reporting HEDIS measures. The program ensures that NCQA’s standards and protocols are met and that data provided from the original source accurately reflects data reported for use as standard supplemental data—in other words, validated DAV data can forgo additional audits for reporting to programs, such as NCQA’s Health Plan Accreditation and/or CMS’s Medicare Stars. We encourage CMS to consider this option.
The foundation we created with the DAV program will allow us to easily support CMS in testing and will strengthen digital clinical data gathered through FHIR-based APIs. Our initial work supported organizations producing and transmitting standardized aggregated CCDs—a logical first step, given that CCDs are the industry standard for sharing clinical data. But CCDs have limitations, and other formats for clinical data might hold promise. In our next phase, we aim to standardize flat files, FHIR and Bulk FHIR, based on industry need and readiness. One benefit of a standardized flat file is expanded, usable and trusted data validated through a data stream, allowing organizations to make more informed decisions about closing gaps in care. The FHIR and Bulk FHIR format will not only encourage organizations to drive business value from policy-driven requirements (ONC, TEFCA, 21st Century Cures) for industry standard APIs, but will also ensure that data can be easily used across multiple organizations and applications. NCQA is the only organization committed to validating and standardizing clinical data for use in quality measurement, and we will be a premier partner for CMS in strengthening and testing data obtained through FHIR.
Potential Future Alignment of Measures Across Reporting Programs, Federal and State Agencies and the Private Sector
NCQA supports the federal government’s alignment efforts across agencies and the initial steps to drive consensus with industry partners. The Core Quality Measures Collaborative is an important preliminary effort to align measures, reduce burden and identify high-impact core measures for value-based payments. In parallel, it will be essential for CMS to coordinate and support HL7’s ongoing work to advance FHIR resources and the Gravity Project’s efforts to code data elements and value sets for SDOH. NCQA proudly supports both efforts.
NCQA has been a leader in educating commercial plans, employers and clinicians on the benefits of digital measurement. We look forward to continuing our work with industry partners to align the digital data infrastructure needed for all dQMs. But while it is important that measure requirements align across federal, state and private sector programs to reduce burden, we must also have industry alignment around data standards and formats. Without alignment, we cannot realize the goal for all dQMs by 2025.
As CMS prioritizes new digital measure development, we encourage the agency to focus on health equity and “high-impact” measures where evidence shows additional access to care can improve quality outcomes (e.g., heart failure, behavioral health). Additionally, collection, standardization and interoperability of race, ethnicity, language, gender identity, disability and SDOH data should be a priority for CMS to advance equitable outcomes.
NCQA remains an ideal partner to support this work and continues to be the national leader in digital quality measurement. We welcome the opportunity to discuss our experience and findings, and we remain committed to working with CMS to build a more equitable, sustainable and responsible American health care system.
Thank you again for the opportunity to comment. Although we answered many questions in our comments, for your convenience we have also included direct responses to questions posed in the RFI. Please refer to the Appendix. If you have any questions, please contact Eric Musser, Director of Federal Affairs, at (202) 955-3590 or at email@example.com.
Margaret E. O’Kane
Appendix: Solicitation of Comments (pp 448-489)
|column 1||column 2|
|Do you have feedback on the dQM definition?||We support defining dQM as “a software that processes digital data to produce a measure score or measure scores.”|
|Does this approach to defining and deploying dQMs to interface with FHIR-based APIs seem promising? We also welcome more specific comments on the attributes or functions to support such an approach of deploying dQMs||Yes.|
|Do you agree that a transition to FHIR-based quality reporting can reduce burden on health IT vendors and providers?||Yes.|
|Would access to near real-time quality measure scores benefit your practice?||We believe the dQM infrastructure as described by CMS can supplement clinical decision support.|
|What parts of the current CMS QRDA IGs cause the most burden?||The schematron technology incorporated into the CMS QRDA IG is outdated and the number of IG requirements is burdensome.|
|What could we include in a CMS FHIR Reporting IG to reduce burden on providers and vendors?||Requirements of a CMS FHIR Reporting IG need to be sufficient and efficient for quality measurement. Rigorous requirements beyond minimum necessary standards will add burden on providers and vendors.|
|Do you agree with the goal of aligning data needed for quality measurement with interoperability requirements? What are the strengths and limitations of this approach? Are there specific FHIR Implementation Guides suggested for consideration?||Yes. Quality reporting and value-based payment are high-value business cases that support adoption of interoperability and data-sharing standards. With strong leadership from CMS and other federal agencies, this should drive interoperability and accelerate the move to dQMs while eliminating burdensome and costly manual processes.|
|How important is a data standardization approach that also supports inclusion of PGHD and other currently non-standardized data?||We believe PGHD and other nonstandardized data should be incorporated into CMS dQM vision, but we must address the lack of standardized electronic clinical data in parallel. The future of digital measurement should incorporate a whole-person, patient-centered approach, in which standardize PGHD and other non-standardized data (e.g., SDOH) are aggregated for a holistic view of individuals and populations.|
|What are possible approaches for testing data quality and validity?||NCQA is committed to validating and standardizing clinical data committed for use in quality measurement and looks forward to partnering with CMS to strengthen and test data obtained through FHIR and FHIR-based APIs, through our Data Aggregator Validation program.|
|What functionalities, described in Section (4)(b) or others, should quality measure tools ideally have in the context of the pending availability of standardized and interoperable data (for example, standardized EHR data available via FHIR-based APIs)?||Connect resources that enhance the utility of shareable, computable knowledge to foster alignment across quality measurement processes and deliver real-time, evidence-based decision support informed by recent data.|
|How would this more open, agile strategy for end-to-end measure calculation facilitate broader engagement in quality measure development, the use of tools developed for measurement for local quality improvement, and/or the application of quality tools for related purposes such as public health or research?||The more open and agile strategy for an end-to-end measure calculation can fill an important role in situations such as the COVID-19 pandemic, when rapid dissemination of new knowledge, guidance and feedback, based on trustworthy data, can inform care and save lives. It can also be a critical link in identifying and closing the gaps in health equity.|
|Do you have feedback on policy considerations for aggregation of data from multiple sources being used to inform measurement?||We encourage CMS to establish policies for data aggregators (e.g., HIEs, vendors), as well as for EHRs, clinical registries and similar data sources, to support CMS quality reporting. CMS can leverage NCQA’s Data Aggregator Validation program to ensure data is properly aggregated, standardized and validated for the use in CMS quality programs.|
|What are initial priority areas for the dQM portfolio given evolving interoperability requirements (for example, measurement areas, measure requirements, tools)?||We encourage CMS to focus on health equity and “high-impact” measures where evidence shows additional access to care can improve quality outcomes (e.g., heart failure, behavioral health). Additionally, collection, standardization and interoperability of race, ethnicity, language, gender identity, disability and SDOH data should be a priority for CMS to advance equitable outcomes.|