Converting X12 EDI Health Insurance Claims to FHIR

[Guest Luis-Apellaniz]

Processing health insurance claims can be quite complex. This complexity is driven by a few factors, such as the messaging standards, the exchange protocol, workflow orchestration, all the way to the ingestion of the claim information in a standardize and scalable data stores. To enable operational, financial, and patient-centric data analytics, the claims data stores are often mapped to patient health records at the cohort, organization, or even population level.

 

 

 

What is X12 EDI?

 

 

Electronic Data Interchange (EDI) defines a messaging mechanism for unified communication across different organizations. X12 claims based processing refers to a set of standards for electronic data interchange (EDI) used in the healthcare industry to exchange information related to health claims. The X12 standard defines a specific format for electronic transactions that allows healthcare providers, insurers (payers), and other stakeholders to exchange data in a consistent and efficient manner. This cross-industry standard is accredited by the American National Standards Institution (ANSI). For simplicity, we will refer to 'X12 EDI' as 'X12' throughout this article.

 

 

 

What is FHIR?

 

 

FHIR® (Fast Healthcare Interoperability Resources) is a standard for exchanging information in the healthcare industry through web-based APIs with a broad range of resources to accommodate various healthcare use cases. These resources include patient demographics, clinical observations, medications, claims and procedures to name a few. It aims to improve the quality and efficiency of healthcare by promoting interoperability between different systems.

 

 

 

Azure Health Data Services is a suite of purpose-built technologies for protected health information (PHI) in the cloud. The FHIR service in Azure Health Data Services enables rapid exchange of health data using the Fast Healthcare Interoperability Resources (FHIR®) data standard. As part of a managed Platform-as-a-Service (PaaS), the FHIR service makes it easy for anyone working with health data to securely store and exchange Protected Health Information (PHI) in the cloud.

 

 

 

Why Convert X12 to FHIR?

 

 

FHIR is a modern, developer-friendly, born-in-the-cloud data standard compared to the aging X12. Converting from X12 to FHIR has many merits; (1) take advantage of FHIR interoperability and adoption to exchange claim information across various systems using modern and secure protocols, (2) unification of patient health and claim dataset into a single FHIR service in the cloud (3) enjoy a larger community of developers and evolving ecosystem at the global healthcare stage.

 

 

 

The Azure Solution

 

 

In essence, this article describes how to orchestrate the conversion of X12 claims to FHIR messages using Azure FHIR Service (with Azure Health Data Services), Azure Integration Account and Azure Logic Apps. Azure Logic Apps is a service within the Azure platform that enables developers to create workflows and automate business processes through the use of low-code/no-code visual and integration-based connectors. The service allows you to create, schedule, and manage workflows, that can be triggered by various events, such as receiving an HTTPS request or the arrival of a new file in an SFTP service. The Azure Integration Account is part of the Logic Apps Enterprise Integration Pack (EIP) and is a secure, manageable and scalable container for the integration artifacts that you create. The X12 XML Schema will be provided through the Azure Integration Account. The complete implementation of the X12 to FHIR conversion in Azure is available on GitHub.

 

 

 

Orchestration of X12 to FHIR Conversion

 

 

Azure Logic Apps orchestrates the conversion process from X12 to FHIR resources, allows for additional data quality checks, and then ingests the FHIR resources in the Azure FHIR Service as depicted in the following 4 steps:

 

X12 to FHIR

 

 

 

1. First, we ingest the X12 file content into the Azure Logic Apps workflow. In this sample, we submit the X12 file content in the body of an HTTPS Post request to the HTTPS endpoint exposed by Azure Logic Apps.
   
2. Initial data quality check and decoding is done using the Azure Logic Apps X12 connector leveraging the X12 XML schemas associated with the transaction sets. This step will verify that the sender is configured and enabled in the system and pick the correct agreement that is configured with the X12 schema. This schema is used to convert the X12 data to XML.
   
3. Once the X12 file is validated and decoded into the XML format, the XML content can then be converted to FHIR using the Azure Logic Apps XML to JSON Liquid connector. This uses DotLiquid templates to map the XML content to the corresponding FHIR resources.
   
4. The output of the workflow is to store the data in Azure FHIR Service (with Azure Health Data Services) to support a unified view of the patient's record. The FHIR service supports an HTTP REST endpoint where individual resources can be managed or sent as an atomic transaction using a FHIR bundle.
   

 

FHIR Resources

 

 

Various FHIR resources can be generated from an X12 transaction set. Depending on business requirements and entities participating in the integration, these resources will vary.

 

 

 

Entity	Resource	Description
Patient	Patient	The person who the claim is for.
Sender	Organization or Practitioner	The organization or provider submitting the claim.
Recipient	Organization or Practitioner	The organization or provider receiving the claim for processing.
Claim	Claim	The details about the services, amounts, and codes associated with the claim.
Transaction Set / Message	Communication	Metadata bout the X12 message including the raw message.

 

 

 

 

Liquid Template Sample

 

 

A sample liquid template is provided showing how to extract data from the decoded X12 file. In the following snippet, the elements under 'content' correspond to XML elements in the decoded X12 file. The XML elements are being mapped to the 'total' attribute of the 'Claim' FHIR resource.

 

 

 

 

 

{

"resourceType": "Claim",

"status": "active",

"use": "claim",

"total": "{{content.FunctionalGroup.TransactionSet.X12_00501_835.TS835_2000_Loop.TS835_2100_Loop.CLP_ClaimPaymentInformation.CLP03_TotalClaimChargeAmount}}"

}

 

 

 

 

 

Considerations

 

 

 

 

• Patient and provider identifiers in the transaction set may not correspond directly to the FHIR identifiers for those matching resources. A lookup approach may be needed to map the data such as an EMPI (enterprise master patient index) and Provider Registry. These mappings can exist in a separate data store or using the FHIR Identifier data type for the corresponding FHIR resource.
  
• Various X12 EDI schemas may need to be managed across your provider base. Each version of the transaction set will have a corresponding Liquid template which will also need to be versioned to convert the correct XML to FHIR. An approach around modularizing templates will be crucial to find the right balance for reusability.
  
• Depending on the scale of the provider base and security requirements the architecture can be revised accordingly:
  
  • One instance of Azure Logic Apps can be created per provider providing full compute isolation.
    
  • Azure Logic Apps also support parallelization allowing for a batch of X12 files to be submitted and then processed in parallel.
    
  • One instance of Azure Logic Apps and Azure FHIR Service can be associated with a certain geographic region which may be needed if data sovereignty is required.
    

  [*]Depending on business scenario, the ingestion process can be trigger from an SFTP event. Health organizations and providers can be associated with an Azure Storage Account enabled with SFTP where they can securely connect and manage their X12 artifacts.

 

References

 

• X12 to FHIR Demo
  
• Azure Health Data Services - FHIR, DICOM & MedTech | Microsoft Azure
  
• Logic App Service – IPaaS | Microsoft Azure
  
• Exchange X12 messages for B2B integration - Azure Logic Apps | Microsoft Learn
  
• Transform JSON and XML using Liquid templates as maps in workflows using Azure Logic Apps
  
• GitHub - dotliquid/dotliquid: .NET Port of Tobias Lütke's Liquid template language
  
• Medallion Architecture
  

 

Continue reading...