Semantic Alignment of Korean EDI Codes with SNOMED CT: A Focus on Procedures in Anesthesiology and Pain Medicine

Background: Fee schedule codes, also referred to as Electronic Data Interchange (EDI) codes, are administrative codes primarily designed for reimbursement purposes and have limited ability to represent precise clinical meaning.This limitation constrains the accurate interpretation of healthcare claim data.Using Systematized Nomenclature of Medicine Clinical Terms (SNOMED CT)'s attribute-value concept structure, we aim to clarify the embedded operational meaning of EDI codes within the national health insurance system by mapping them to detailed SNOMED CT concepts.Methods: Nine EDI codes in anesthesiology and pain medicine from Chapter 9 (Procedures and Surgical Fee Schedule) were analyzed.Source data, including fee schedules, reimbursement criteria, and coverage guidelines, were integrated.The manual mapping process followed a structured workflow consisting of candidate SNOMED CT concept identification, assessment of billability, evaluation of semantic sufficiency, and application of post-coordination or concept modeling where necessary.One-to-many mappings were allowed to reflect composite meanings.Validation included 18 expert consultations with 6 clinicians and 3 terminology specialists. Results:The nine EDI codes were mapped to a total of 42 SNOMED CT expressions.One code was fully deferred.Five EDI codes were mapped to a single SNOMED CT expression each, while the three (including one partially deferred) were linked to multiple expressions with EDI-to-expression ratios of 1:3, 1:7, and 1:27, respectively.A representative code (S4825) was linked to up to 27 SNOMED CT expressions, demonstrating substantial clinical heterogeneity within a single billing code.Among 30 SNOMED CT procedure-defining attributes, 10 key attributes-such as Method, Procedure site -Direct, Using device, and Has intent-were identified as essential for structured representation of clinical meaning.Conclusion: Semantic alignment using SNOMED CT enables decomposition and structured representation of the complex clinical meanings embedded in EDI codes.This approach improves interpretability and reliability of claims data and provides a foundation for refining procedure classification systems and enhancing interoperability with international standards.