The Engine Powering Digital HEDIS® and Quality Modernization

Building a Universal Foundation with OmniQ

For decades, quality measurement has operated on fragmented views: within a Payer enterprise separate teams manage data packets with limited context, locked in proprietary formats; Provider data and workflows are managed and reconciled on different systems.

NCQA’s Digital HEDIS and various CMS mandates towards interoperability and data sharing are changing that status-quo. CMS's Universal Foundation reflects the policy of aligning quality measures, so one governed clinical data set can serve multiple purposes.

The Way Forward: Essentials 

Every requirement in healthcare — in-room patient care, compliance, prior authorization adjudication, quality measurement, automation, analytics, and population management — needs real-time access to high-quality, trusted data. None of that can be assumed any more. In order for data to be trusted at a massive scale on a continuous basis, the following elements are now non-negotiable for any healthcare organization: 

Complete Data

To achieve complete, comprehensive, computable FHIR^® data, all data in all formats must be ingested from legacy systems, supplemental data sources, EHRs, pharmacy, CDAs, claims, and clinical sources. The engine then normalizes and de-duplicates the data while it is continuously ingested, maintaining traceability, security and auditability as data flows in and out. For Digital Quality Measurement, data has to be clinically contextualized in FHIR with CQL logic before it is usable for measurement.

Continuous Data

Digital measurement is not truly continuous if the underlying platform architecture depends on legacy batch preparation and sequential processing. Many vendors in the Digital HEDIS market support FHIR APIs to accept data. Accepting data through a FHIR API is not the same as evaluating measures natively in FHIR. Where the calculation still runs through a translation layer into a proprietary engine, batch and sequential processing remain, and so does the multi-weekweed cycle.

The mechanisms that allow a continuous flow of validated data must be built into the architecture. A FHIR-native engine that runs CQL logic removes that translation layer, allowing measure evaluation to run concurrently across measures, in minutes, without a batch queue. This open-standards architecture makes validated and complete data available continuously.

Trusted and Validated Data

One of healthcare’s most expensive assumptions is that an organization’s data is ‘good enough’. On the surface, data is available in FHIR, can be accessed and shared, and reports seem to run. But as initiatives scale across organizations (or larger data sets are ingested) for digital quality measures, prior authorization, or running AI analytics, legacy architectures risk cracking under pressure. Resolving these risks means additional patchwork fixes, accumulated technical debt, additional IT resources, and unplanned budget. This risk impacts executive decisions (and trust in those decisions), and a Payer’s ability to govern effectively, respond to regulatory changes, control costs, and move toward value-based models of care.

Trust is the prerequisite for scale, and validated, high-quality, continuous data is now an operational requirement for every healthcare organization.
 

What Changes for Quality Teams

Payer quality teams are used to working on static, batched data sets, retrospectively with heavy manual and administrative burdens.

Over the next few years, the changes to quality workflow design, data and technology architecture, and overall governance will allow quality teams to impact care decisions proactively.

With data that is validated, governed and continuously available, quality teams will be able to run massive data sets against concurrent measure evaluation, and start to see results in minutes (rather than weeks). Net new quality initiatives that align with value-based care and impact Star Ratings can be added, without technology limitations holding the teams back.

Technology That Delivers the Future of Quality

With real-world data from early-adopter Payer clients, we ran a comparison between traditional HEDIS results and our dQM - Measures for HEDIS solution. Over 5 billion data points from over 5 million members were processed by the engine in under 12.5 hours. Data sets included up to 6,000 claim resources per individual, reflecting the complexity of real-world payer populations, versus test or sand-box environments. Across 80 HEDIS measures, the median variance comparing traditional HEDIS results and OmniQ results, came in at 0.3%, with average variance at approximately 1%. The expected industry variance was estimated at 5-10%.

Architecture that Scales for Billions

Healthcare data is growing faster than any other industry's, with an estimated compounded growth rate of 36% annually. Traditional enterprise architectures in healthcare are at their limit when it comes to the sheer volume of data that will continue to move through health systems.

Globally, we are moving towards technology architectures that can handle planetary-scale data. Data at this scale needs to be managed in three pillars: volume, velocity, and clinical complexity. The quality lifecycle and measurement processes require pioneering strategies, rather than incremental or cosmetic upgrades to legacy systems.

Smile's proven technology architecture is built for the future of healthcare on open standards that break data and workflows out of proprietary vendor silos, and allow Payers to innovate and adapt to changes as they come.

The OmniQ architecture is also built on Apache Iceberg and Apache Spark, both open-standard technologies that work alongside FHIR and CQL. Iceberg makes planetary-scale FHIR data queryable and analytics-ready. Spark distributes the actual measure calculations across the platform, so results are produced continuously, at scale, without the delays of legacy systems, and far beyond existing industry standards.

OmniQ powers concurrent measure reporting for over 5 billion points without the bottlenecks traditional database structures hit at that volume. Scaling traditional architectures to handle that load is resource-intensive, and is often associated with hidden infrastructure costs. Smile's approach is built to make real-time analytics at scale financially sustainable.

Beyond OmniQ dQM - Measures for HEDIS

dQM - Measures for HEDIS extends to other high-value use cases, including automation of prior authorization and the authoring of custom, state and jurisdictional measures. Payers can author and update their own guideline policies directly, rather than waiting on vendor-led projects. Smile's platform gives payers that autonomy by allowing additional content to be ingested and managed on the same OmniQ foundation, without significant additional fees for doing so.

Unified Quality and Compliance

OmniQ is built on a fully computable foundation. This single engine powers not only HEDIS and Custom quality measurement, but also MIPS, CMS compliance regulations, real-time prior authorization automations, and prospective care gap management.