SNOMED CT, Clinical Decision Support Systems, and CQL
Collection: Technology adoption
Summary: Yes, SNOMED CT supports interoperable CDS systems by leveraging intensional Value Sets and ECL to dynamically define clinical criteria while using SMART Guidelines and CQL to translate clinical knowledge into actionable, machine-readable formats.
Technology use case
Developing effective Clinical Decision Support (CDS) systems requires rules that are reusable, shareable, and maintainable. Healthcare organizations often struggle to achieve consistency across multiple systems while ensuring that the CDS rules remain up-to-date with evolving clinical knowledge. Creating such rules manually is resource-intensive and risks inconsistency, making it challenging to achieve scalable, interoperable solutions.
What Is a CDS System and Why Is It Important?
A Clinical Decision Support (CDS) system is a health information technology system designed to assist healthcare providers in making informed clinical decisions. CDS systems provide clinicians with patient-specific assessments or recommendations based on clinical guidelines, evidence-based practices, or patient data. By integrating CDS systems into healthcare workflows, providers can enhance the quality of care, reduce errors, improve efficiency, and ensure adherence to clinical standards. CDS systems are crucial for supporting complex decision-making, reducing variability in care, and improving patient outcomes.
What Are SMART Guidelines?
SMART Guidelines1 is a set of best practices and standards that help translate clinical guidelines into machine-readable formats that can be easily implemented in electronic health records (EHRs) or other digital systems. SMART Guidelines enable the integration of clinical evidence into workflows, making it easier to apply standardized healthcare practices across different environments.
The concept of SMART Guidelines was introduced by the World Health Organization (WHO) as part of their effort to transform traditional, text-based guidelines into digital, actionable tools that can be integrated into health information systems2. The goal was to create a more efficient way for healthcare providers to access and use clinical guidance at the point of care, leading to improved health outcomes. SMART Guidelines have gained traction as the need for interoperable, standards-based health IT solutions has become increasingly important in global health settings.
What Is CQL?
Clinical Quality Language (CQL)3 is a high-level language designed to express clinical knowledge in a way that is human-readable and executable by machines. CQL is often used to write CDS rules and clinical quality measures, enabling consistent and scalable decision-making across health systems. By using a standardized language like CQL, clinicians and developers can ensure that their CDS rules are interoperable, more easily maintained, and adaptable to changes.
CQL was developed by the Health Level Seven International (HL7) organization as a part of their Clinical Quality Framework initiative. The need for CQL arose from the desire to create a unified language that could express clinical quality measures and decision-support logic in a standardized format. Since its introduction, CQL has become a widely adopted language for expressing computable clinical knowledge, making it easier to share, reuse, and implement clinical decision support across different health IT systems.
CQL is also closely related to SMART Guidelines, as it provides the machine-readable language needed to encode the clinical decision logic derived from these guidelines. By using CQL in conjunction with SMART Guidelines, healthcare organizations can create adaptive and automated decision support that aligns with standardized, evidence-based practices. This synergy ensures that clinical guidelines are consistently translated into actionable decision-support rules, enhancing the quality and efficiency of care.
Advice
SNOMED CT plays a fundamental role in providing the terminology underpinnings of reusable and maintainable CDS rules, ensuring that clinical concepts are represented consistently and accurately. Value sets are collections of SNOMED CT concepts used to define clinical criteria, and their use enhances the maintainability of CDS rules by standardizing the clinical data used in decision-making. By utilizing SNOMED CT, CDS rules are grounded in a globally recognized terminology system, which supports interoperability and ensures that clinical logic can be shared across different systems.
Intensional value sets that are automatically updated can take full advantage of SNOMED CT's hierarchies, reference sets, and attributes, ensuring that Clinical Decision Support Systems (CDSS) stay up-to-date as clinical standards change4. SNOMED CT hierarchies allow for the efficient selection of all descendants of a concept, enabling flexible yet precise definitions of clinical criteria. Reference sets (refsets) provide pre-curated collections of SNOMED CT concepts that can be reused across various clinical scenarios. At the same time, attributes allow for detailed refinement of these concepts, increasing the specificity of value sets. The Expression Constraint Language (ECL)5 provides a structured syntax to define these dynamic value sets, leveraging SNOMED CT's rich structure for creating, maintaining, and updating clinical decision support in a scalable and efficient manner. SNOMED CT hierarchies allow for the selection of all descendants of a concept, enabling broad yet precise definitions of clinical criteria. Reference sets (refsets) provide curated collections of SNOMED CT concepts that can be reused across different use cases, while attributes allow for more detailed refinement of concepts, enhancing the specificity of value sets. This approach helps ensure that CDSS are always aligned with the most current clinical knowledge and standards.
Many FHIR servers support the execution of CQL-based CDS rules, providing a standardized platform for implementing clinical decision support logic. HAPI FHIR server is one such example that supports these capabilities6. By integrating a HAPI FHIR server into the workflow, organizations can automate the retrieval of SNOMED CT concepts using dynamic value sets, ensuring that the CQL-based logic references the most current clinical knowledge. This approach significantly reduces the burden of manual updates and enhances the scalability of clinical decision-making processes, leading to consistent and reliable care.
Tips for best results using SNOMED CT with Smart Guidelines and CQL
Leverage Intensional Value Sets: Use intensional value sets to define clinical criteria dynamically. This allows you to take advantage of SNOMED CT's hierarchies, attributes, and reference sets, enabling automated updates and minimizing manual maintenance.
Utilize Expression Constraint Language (ECL): Use ECL to define value sets in a flexible and structured manner. This ensures that the defined sets are consistent and precise, supporting effective and reliable clinical decision-making.
Integrate with FHIR Servers: Deploy a FHIR server, such as HAPI FHIR, to facilitate the execution of CQL-based logic. This will help automate the retrieval and application of SNOMED CT concepts, ensuring your clinical decision support system remains up-to-date.
Adopt SMART Guidelines for Standardization: SMART Guidelines ensure that clinical guidelines are translated into standardized, machine-readable formats. Combining SMART Guidelines with CQL and SNOMED CT can improve interoperability and consistency in CDS implementations.
Focus on Interoperability: Use SNOMED CT as the standardized terminology to ensure your CDS logic can be shared across different healthcare settings and systems. This supports consistent care delivery and facilitates data exchange.
Regularly Verify Value Sets Syntax: The content of intensional Value Sets is automatically updated based on the latest SNOMED CT releases. This keeps CDS rules aligned with current clinical standards and ensures reliable decision support. However, the Value Set definitions themselves may require maintenance if any of the referenced concepts in the definition have changed.
Test and Validate Rules Continuously: Implement continuous testing and validation processes to ensure the CQL-based logic and SNOMED CT value sets function as intended. Regular validation helps identify gaps and improves the accuracy of CDS systems.
References
Smart Guidelines: https://www.who.int/teams/digital-health-and-innovation/smart-guidelines
WHO SMART guidelines: optimizing country-level use of guideline recommendations in the digital age. Mehl, Garrett, et al. The Lancet Digital Health, Volume 3, Issue 4, e213 - e216
HL7 - CQL Language: https://cql.hl7.org/
HL7 - FHIR Value Sets: https://hl7.org/fhir/valueset.html#compositions
SNOMED Expression Constraint Language: http://snomed.org/ecl
HAPI FHIR - Clinical reasoning: https://hapifhir.io/hapi-fhir/docs/clinical_reasoning/overview.html
Learn More
To learn more, please reach out to [email protected].
Last updated