ONC Project Taps FHIR to Extract Over 40% of Clinical Data Registry

Only about 46 percent of non-federal acute care hospitals in the United States participate in clinical data registries due to the daunting task of data collection. However, a solution may be in sight. 

A project funded by ONC's Leading Edge Acceleration Projects (LEAP) leveraged Fast Healthcare Interoperability Resources (FHIR) to simplify the historically complicated task of populating clinical registries and use the data to improve care decisions. 

Clinical data registries are vital repositories of patient health information and treatment details. These registries focus on specific conditions, enabling healthcare providers to analyze treatment effectiveness across diverse patient profiles. By leveraging these registries, hospitals and private practices can measure and enhance the quality of care provided to patients. 

However, the journey to compiling comprehensive clinical registries has long been plagued by challenges. 

“It’s not an easy process to gather all the data needed to complete a registry form,” Anmer Ayala, an ONC official, wrote in a HealthITBuzz blog post. “This data may be collected from multiple locations such as EHR, emergency medical services systems, or a legacy system that is not connected to an EHR. Data may need to be collected from multiple facilities or care settings to create a longitudinal portrait of the patient.” 

The existence of disconnected systems often results in arduous and time-consuming manual data extraction from multiple sources. 

Additionally, the need for discussions with clinical teams to interpret and derive supplementary data further complicates the process. After completing the registry forms, submissions undergo rigorous validation against unique data rules established by the registry. 

Using a case study from Chesapeake Regional Information System for our Patients (CRISP), ONC outlined key clinical data management strategies for data registries. 

The collaboration between CRISP and the American College of Cardiology (ACC) focused on the joint project titled "Development and Testing of Data Sharing Functionality for Health System Participating in National Cardiovascular Disease Registries," focused on enhancing the interoperability and efficiency of clinical data management. 

They focused on the CathPCI Registry, which stores essential information on cardiac disease, including treatments, outcomes, demographics, patient history, and risk factors. By mapping the registry's proprietary data elements to widely recognized standards like SNOMED CT and LOINC, they aimed to streamline data integration and address any mapping challenges identified. 

The project also explored the potential of using FHIR to extract and submit registry data effortlessly. During the January 2023 HL7 FHIR Connectathon, CRISP successfully demonstrated data extraction from EHRs and submission to the CathPCI Registry using a proof-of-concept application. 

James E. Tcheng, a cardiologist at Duke University School of Medicine and an informatics advisor to ACC, led efforts to normalize the National Cardiovascular Data Registry (NCDR) to universal standards. Tcheng reported that more than 40 percent of clinical registry data could be extracted using FHIR, marking significant progress. However, complete extraction has not yet been achieved. 

This innovative endeavor holds tremendous potential for advancing patient care and research in cardiovascular health, Tcheng said 

“Using FHIR to mitigate the pain points of registry completion can result in reduced processing times, less intervention for registry documentation, reduced submission delays, and improved data quality,” Ayala said. “As a result, the time required for clinicians to use the information to make clinical decisions and support patient care can be reduced.” 

CRISP and ACC have also created the Protocols for Clinical Registry Extraction and Data Submission (CREDS) FHIR Implementation Guide (IG). This groundbreaking guide offers a scalable solution for hospitals, health information exchanges (HIEs), and other contributing organizations involved in the National Cardiovascular Data Registry (NCDR) and other registries. Set to be published in the fall of 2023, this IG aims to streamline data extraction and submission processes, revolutionizing the way clinical data is shared and utilized.