Beyond Compliance: Scale Clinical Intelligence with FHIR® & CQL

As the CMS-0057-F enforcement date draws closer, more and more Payers are interested in how to meet the mandate requirements while simultaneously  realizing better efficiency, scale and return-on-investment. 

Recently, our Payer clients have been asking us questions about HL7^®’s other open standard, CQL (in addition to the widely known FHIR standard). This blog unpacks what FHIR and CQL are, why FHIR and CQL matters for Payers, how it is leveraged in the Da Vinci Implementation Guides (IGs), and the benefits that are realized from this standard.

Standards like FHIR and CQL lay the groundwork for a scalable, interoperable future of healthcare. Let’s look at how.

A Quick Refresh

Current State and Future Potential: The ROI

Due to global adoption and the CMS mandates in the US, most Payers already know the benefits of FHIR data interoperability and modernization. What most don’t realize is that this is just the beginning. 

FHIR certainly makes data accessible in a standardized format—but the technologies that power workflows like ePriorAuth or quality measure evaluations are proprietary, siloed, not-interoperable, even within an organization. Without a shared way to define clinical logic and processes, scalability is impossible, operations are manual and inefficient, and analytics are often inconsistent and difficult to trust. This is where CQL comes in as the next evolution.

By combining the data structure of FHIR with the clinical logic expression capabilities of CQL, healthcare organizations can finally begin to scale clinical intelligence.

What Is Clinical Intelligence?

Clinical Intelligence is the ability to derive patient-specific, actionable insights leveraging high-quality healthcare data and automation. This is an area that greatly benefits from responsible and ethical use of AI. As such, it needs to be reliable, repeatable, and scalable to address many use cases:

• Authoring (aka codification of policies) and sharing of Clinical Practice Guidelines (CPGs)
  
• Automating Prior Authorization (PA)
  
• Conducting prospective Care Gaps Management
  
• Running real-time Quality Measures (eCQMs, dQMs)
  
• Delivering Cognitive Support

Trustworthy clinical intelligence depends on having:

• Interoperable, standardized data sources with high quality clean data (i.e. FHIR data repositories)
• Integrated, transparent, auditable workflows (process and knowledge interoperability)
• Codified reusable logic that can be maintained and shared from one central source
  
  

Today, these capabilities across healthcare organizations are fragmented across tools, vendors, data sources, and siloed proprietary internal systems or just not possible due to the sheer quantity of rules and guidelines that exist. 

Smile’s Digital Health’s FHIR and CQL engine and AI-assisted guideline/policy codification powers clinical intelligence in a way that makes it efficient and scalable.

How FHIR + CQL Creates a Compliant-ready Foundation for Modern Clinical Intelligence

1. Breaking Data and Process Silos

FHIR defines standardized "resources" (such as Patient, Medication, Observation) for representing clinical data. CQL provides the mechanism to write rules and logic using these resources. Together, they eliminate inconsistencies by providing semantic accuracy, reusable logic, and interoperability. This integration democratizes clinical intelligence across fragmented Payer systems, and even between Payers and Providers. 

2. Optimizing Business Processes

With FHIR and CQL, business workflows—such as Prior Authorization or Clinical Practice Guideline application—are both computable and human-readable. That means each step (inputs, decisions, triggers, outcomes) is standardized, reusable, and strategically utilizes human expertise. This drastically reduces manual overhead and accelerates execution.

3. Responsible Use of AI

In recent years, there has been increased concern in regards to the use of unregulated AI tools for PriorAuth adjudications, as they have caused systemic denials of requests. This is because most Payers rely on probabilistic AI models, which assign likelihoods to potential outcomes. Smile’s approach is different and leverages AI as a tool to assist in the creation of coded logic, not the execution of it. Smile’s tooling can ingest narrative text PDFs of Clinical Practice Guidelines and established industry best practices, and transform them into computable policies. These codified policies can then be executed by our FHIR and CQL engine for prior authorization workflows, and other use-cases like cognitive support and quality measures. This approach is called ‘deterministic AI’ as it is based on published medical research and clinical policies.  

By integrating this deterministic methodology with the FHIR and CQL engine, Payers can build trust in automated systems, ensuring that AI supports—not overrides—good medical judgment. This approach not only enhances compliance with regulations like CMS-0057-F but also lays the foundation for scalable, reliable clinical intelligence.

4. Future-Ready Design

Unlike legacy tools and formats, FHIR and CQL were designed for extensibility. They are already integral components to:

• Davinci IG for Prior Authorization DTR
• NCQA HEDIS Digital Quality Measures (dQMs)

Investing in a FHIR and CQL solution ensures that Payers are not just on target to meet the upcoming CMS-0057-F, but also future mandates, innovations and regulation that will direct the future of healthcare.

FHIR + CQL = ROI Beyond Compliance

Today, Payers are focused on meeting minimum requirements for CMS-0057-F. But there is a bigger opportunity here as well. Investing in a FHIR and CQL engine to scale clinical intelligence and unlock ROI across multiple business functions.

The Benefits of FHIR + CQL for Clinical Intelligence:

• Scalability: A shared infrastructure supports more use cases, across sites, systems and networks, without requiring more tools or vendors.
• Accuracy: CQL’s semantic clarity (i.e. clinical logic expressed in standardized, human readable ways) helps reduce manual abstraction errors in quality reporting.
• Interoperability: Clinical logic and data can be reused across departments and organizations.
• Cost Reduction: Streamlining prior auth, quality measure reporting, and predictive care gaps with automation reduces administrative burden.