Skip to site content
New to Availity? Get Started

Fuel your FHIR with Better-Quality Data

Fast Healthcare Interoperability Resource (FHIR®) is a relatively newer standard that utilizes up-to-date web-based norms for both data and API mechanisms, facilitating efficient data exchange. The implementation of the FHIR release four (FHIR R4) based API has been mandated by the Office of the National Coordinator for Health Information Technology’s (ONC) 21st Century Cures Act Final Rule and the Centers for Medicare & Medicaid Services’ (CMS) Interoperability and Patient Access Final Rule for the purpose of health data exchange.

Although FHIR is not yet widely employed for seamless patient transitions across care environments like Consolidated Clinical Document Architecture (C-CDA), its availability is essential for enabling patients to access their data, such as through downloads to their mobile devices. Moreover, FHIR is progressively gaining traction for novel applications within and between healthcare entities.

C-CDA documents are a standardized format for the exchange of clinical information between healthcare providers and systems. Recent estimates suggest that billions of C-CDA documents are exchanged annually in the United States, making it one of the most prevalent data standards in the healthcare industry.1

To effectively utilize the widely available C-CDA data to meet the federal mandate and requirement of other FHIR-based applications, Availity has joined others to lead a Health Level Seven International® (HL7®) cross-group project which develops mapping guidance between the C-CDA and FHIR standards. The project group recently submitted the first mapping guidance publication for the HL7 ballot. For more details, checkout my C-CDA to FHIR Mapping blog.

While being a critical step towards interoperability, the mapping between standards only addresses technical interoperability. Both C-CDA and FHIR face data quality challenges relating to how clinical data are recorded, structured, and codified within electronic health records (EHRs). Consequently, despite creating flawless mappings for transferring data between standards, the resulting information might remain unusable if it was inadequately entered at the source. To better illustrate the data quality challenges, we presented the following examples at the American Medical Informatics Association (AMIA), which received significant attention and won the “Distinguished Paper Award.”2

Figure 1. These examples were identified from the artifacts submitted for Meaningful Use Stage 3 C-CDA Certification.

  • Panel A presented a mismatch between the recorded code and its description, which could adversely impact drug-allergy checks.
  • Panel B had a mismatch of vital sign unit between the human-readable and the machine-readable record, which could impact drug dosing.
  • Panel C presented inconsistent dosing information between the pre-coordinated medication RxNorm code and the structured “doesQuantity” field, which could impact medication administration and reconciliation.
  • Panel D omitted a code and unit for a common laboratory result which adversely impact data usability in care transitions or research.

When we convert such data to FHIR, it produces an observation resource that has a data-absent-reason code, such as “unknown”. Consequently, when querying for patients who have met certain clinical criteria (e.g., those who took an HbA1c test, a commonly employed blood test for diagnosing pre-diabetes and diabetes), this specific data point will remain undetected, as it lacks a vaild Logical Observation Identifier Names and Code (LOINC).

Figure 2. C-CDA to FHIR Mapping with Source C-CDA Quality Issues

A research team led by Dan Gottlieb reported similar data challenges at FHIR Dev Days 2023 as they embarked on the creation of quality standards for FHIR data. This underscored the existing interoperability disparity within raw clinical data. To address this issue, Availity Fusion™ employs its Upcycling Data™ technology to standardize the source data into a uniform code system. Once the data is cleaned and coded in this manner, the resulting mapped data transforms into a semantically interoperable resource, poised to power downstream applications effectively.

Figure 3. C-CDA to FHIR Mapping when Source C-CDA Quality Issue is Fixed

Beyond the isolated concerns, our clinical data informatics group has consistently noted substantial data variation. To illustrate, in a collaborative study involving HealtheConnections, a health information exchange (HIE) client, and the National Clinical Quality Assurance (NCQA), we demonstrated that the utilization of the standardized LOINC terminology for laboratory information displayed noteworthy discrepancies across facilities, even within a shared EHR system.

Figure 4. Use of LOINC (y-axis) by Facilities (x-axis, Colored by EHR clusters, most are composed of a single EHR)

As data inherently differ in their origins, it could be prudent to avoid treating them with uniformity. This could involve establishing a data hierarchy, whereby sources of higher quality are given precedence over those of lesser quality in cases of conflict. Alternatively, it might be necessary to exclude consistently low-quality sources that lack credibility. In this scenario, utilizing all data sources could necessitate subjecting lower quality inputs to heightened scrutiny or more rigorous monitoring to guarantee data reliability before use. There are many applications of this information. For example, for the NCQA Data Aggregator Validation certification program, the data quality information can be used to help further define the current clusters, empowering validators to concentrate their efforts on more problematic data sources while conserving time for those with superior data quality.  

To meet the needs of evaluating data quality, Availity has developed a data dashboard module. The dashboard runs on 300+ interoperability rules including both data completeness and content checks and are highly configurable to fit the needs of different use cases. It is deployed at multiple customer sites including Veterans Administration, who routinely use this module to audit and improve their clinical data quality from various data contributors.

In conclusion, the pivotal role of high-quality clinical data cannot be overstated, as it lays the foundation for interoperability and all the downstream use cases. The consequences of embracing and maintaining high data quality resonate across the healthcare ecosystem, leading to enhanced diagnostic accuracy, streamlined operations, reduced costs, and increased patient safety. By prioritizing data accuracy, completeness, and consistency, healthcare organizations can harness the power of data-driven insights to drive innovation, optimize resource allocation, and shape a better future of medical practices. To learn how you can ignite your FHIR strategy with high-quality data, download our FHIR e-Book.


1“Epic Announces Plan to Join TEFCA, Champion Next Step in Evolution Toward Universal Interoperability” Epic Newsroom, 20 June. 2022,

2D’Amore, John, et al. “Interoperability Progress and Remaining Data Quality Barriers of Certified Health Information Technologies.” AMIA … Annual Symposium Proceedings. AMIA Symposium, U.S. National Library of Medicine, 5 Dec. 2018,