Building Next-Generation Digital Quality Measures

June 7, 2023 · Andy Reynolds

You don’t have to be an informaticist to learn from Anne Smith and Michael Ryan’s presentation at the AMIA 2023 Clinical Informatics Conference, Building Next Generation Digital Quality Measures.

Anne and Michael packed valuable context and reminders into less than 15 minutes.

Highlights include:

The Difference Between dQMs and eCQMs

According to CMS, digital quality measures:

• Use standardized, digital data from one or more sources of health information, captured and exchanged via interoperable
  • NCQA recently hired a Vice President of Interoperability. That’s a new role, and we’re thrilled that noted FHIR expert Amol Vyas is in it.
• Apply quality measure specifications that use code packages and are standards based.
• Can be computed in an integrated environment without additional effort.

Electronic Clinical Quality Measures (eCQMs):

• Are the most widely used and familiar dQM program.
  • As CMS puts it, eCQMs “fit within the dQM umbrella.”
• Started in 2013 as a voluntary CMS program for hospitals and became mandatory for some CMS programs in 2016.
• Use data from EHRs and other sources.
• Use the Quality Data Model (QDM).
  • A “data model” is how data elements are organized and relate to each other.
  • FHIR is the future: FHIR eCQMs will replace QDM.

HEDIS and dQMs: Past and Present

NCQA’s Healthcare Effectiveness Data and Information Set (HEDIS^®[1]):

• Offers digital versions of 23 measures for health plan reporting for measurement year (MY) 2023.
  • The symmetry of “23 for 23” is a coincidence. Expect more than 25 digital measures by MY 2025.
• Transitioned to FHIR in 2021 for MY 2022.
• Used dQMs (QDM-based) for the first time in MY 2018.

The QDM has shortcomings, but we’re all-in on FHIR to replace it, because FHIR:

• Has better functionality, flexibility and adoption.
• Allows more complete expression of HEDIS specifications.
• Offers a collaborative, effective stakeholder network.

Three Recurring Themes

The deeper we get into digital measurement, the more desirable and feasible these three themes become:

1. HEDIS requires data from many domains and settings.
   • Because of the variety of HEDIS data and sources, using one implementation guide or set of FHIR profiles just isn’t feasible.
     • We combine an array of established standards, including US Core and QI Core.
   • We need ad hoc selection of data constraints based on diverse use cases.
     • Brad Ryan, NCQA’s Chief Product Officer, acquainted the HEDIS community with use cases in our September 2021 Future of HEDIS webinar.
   • Standardizing data using Clinical Quality Language (CQL) works better than using structured FHIR artifacts.
     • CQL is an “authoring language” that makes measures understandable to people and is useful for computer queries.

2. “Quality care” looks different to different people.
• Generalized traits don’t always capture the nuances of what qualifies as “the right care” for an individual.
• Technology helps make measurement more specific.
• We’re expanding measurement so measures can be individualized.

3. Expand beyond retrospective measurement of health plan performance.
• Expand allowable adjustments.
• Develop flexible digital specifications.
• Configure HEDIS for a variety of cases.
• Align content used for clinical practice guidelines with clinical decision support.

We thank AMIA for hosting Anne and Michael.

Check back for future summaries of our digital experts’ public presentations.

[1]HEDIS^® is a registered trademark of the National Committee for Quality Assurance (NCQA).