stnazkul - stock.adobe.com
ONC-Funded Project Taps FHIR for Clinical Research EHR Data Access
An ONC-funded research team has developed two new platforms that leverage FHIR to support standardized EHR data access for clinical research scalability.
MedStar Health Research Institute, a 2020 ONC Leading Edge Acceleration Project (LEAP) in Health IT awardee, has developed new Fast Healthcare Interoperability Resources (FHIR) platforms to support EHR data access for clinical research.
In collaboration with the Georgetown University Medical Center and HealthLab, the MedStar project team developed two new tools to provide researchers access to standardized data for at-scale extraction and analysis.
The tools, available via ONC’s GitHub repository (registration required), include the Bugs & Drugs FHIR Factory and the Trend Engine FHIR Factory. The framework for both data platforms performs automated, iterative data extraction, as well as transformation and integration of the data using FHIR application programming interfaces (APIs).
The Bugs & Drugs FHIR Factory provides near real-time, continuous antibiograms that define drugs of choice for treating urinary tract infections. Antibiograms are profiles of how susceptible certain microorganisms are to antimicrobial drugs.
The Trend Engine FHIR Factory leverages EHR data to monitor illness rates, utilization of health services, laboratory testing, and clinical diagnoses by area of interest, such as trends of diagnoses of COVID-19, influenza, and stroke.
The project team focused on an opioid use case using the Trend Engine FHIR Factory to look for patterns in the rates of emergency room admittances from opioid use. The researchers replicated the Trend Engine for more than 50 other disease conditions. Hospitals can also use the platform to track the effectiveness of intervention efforts across communities.
According to ONC official Alison Kemp, future use cases for FHIR Factories could include automatically updating data within research articles.
“Because FHIR Factories can automate the repeatable actions of a data scientist, one unique output is the ability to generate living research articles and easily scale healthcare research,” Kemp wrote in a HealthITBuzz blog post.
“Articles can be automatically rewritten each time the data is refreshed, thereby creating results that are near real-time and virtually always up-to-date,” she added. “The articles can include statistics, simulations, and visualizations.”
