Gaps in the quality and accessibility of EHR data and in interoperability can limit researchers’ ability to generate evidence to guide care decisions. A recent report from the U.S. Department of Health and Human Services highlights efforts to  improve the quality, accessibility, and interoperability of EHR data for patient-centered outcomes research (PCOR). These data infrastructure improvements support efforts to scale national research initiatives and accelerate research discoveries using real-world data.

The Office of the Secretary Patient-Centered Outcomes Research Trust Fund (OS-PCORTF) coordinates efforts across HHS agencies to improve the quality, accessibility, and interoperability of EHR data. 

The report notes that despite significant progress in improving data liquidity, adoption and implementation of data exchange standards vary. For example, lower-resourced hospitals engage less frequently in interoperable exchange compared to their higher-resourced counterparts. EHR developers also implement exchange standards within EHRs differently, limiting the availability of complete and longitudinal data for PCOR and comparative clinical effectiveness research.

The OS-PCORTF report highlights its projects that have developed and tested FHIR implementation guides and FHIR-based APIs that standardize EHR data exchange for key U.S. health priorities, including pregnancy and postpartum outcomes, birth outcomes, and childhood cancer.

For instance, data sets representing EHR data from across multiple health systems offer a valuable opportunity to fill a gap by providing detailed, longitudinal information on pregnancy, postpartum, and early childhood outcomes.

A National Institutes of Health-led project developed a suite of resources, including a FHIR implementation guide, an open-source measure-based software tool called MaternalHealthLink, and an API. The MaternalHealthLink tool and accompanying API extract, aggregate, and evaluate longitudinal data on maternal and infant health from EHRs to support research on maternal morbidity, mortality, and hypertensive disorders of pregnancy.

The project tested the FHIR implementation guide and data exchange tool for a new maternal health-focused use case, within an NIH-funded pediatric research effort focused on understanding links between structural birth defects and childhood cancer. Lessons learned from the pilot testing supported complete integration of the data exchange tool into the research platform. This integration now enables timelier extraction and exchange of data to support PCOR related to birth defects and childhood cancer, which can ultimately support the development of improved and targeted treatments, according to OS-PCORTF.

USCDI is the core set of health data elements developed by ONC and required for nationwide exchange—a shared language across health IT systems. It includes data elements such as clinical notes, labs, allergies, and intolerances, laboratory test results, and medications. USCDI evolves every year based on public input, ensuring that the standard keeps pace with real-world needs. USCDI Plus (USCDI+) builds on the core data elements defined in the USCDI by identifying and establishing domain- or program-specific extensions to meet use-case needs. These USCDI+ extensions are developed through a collaborative process with federal and industry partners to ensure they are harmonized with existing standards and taxonomies, enabling more comprehensive interoperable data exchange for priority health and care activities.

Two OS-PCORTF projects are developing a USCDI+ data element list for sickle cell disease and maternal health data element set. Both USCDI+ sickle cell disease and the maternal health datasets provide guidance to health systems about which data standards to use for exchanging electronic health information between health systems.

The USCDI+ maternal health dataset, which contains 145 maternal health data elements including 24 new data elements related to care delivered in the postpartum period, will support higher-quality EHR data for exploring PCOR and CER questions related to maternal health care during and after birth.

Building on this work, a health system will pilot test a postpartum transitions of care FHIR implementation guide that uses these data elements to improve gaps in postpartum care.

OS-PCORTF said its projects also strengthen the governance structures that guide how these data are shared and used for research.

A CDC-led project developed the Making Electronic Data More Available for Research and Public Health (MedMorph) “reference architecture,” a framework for using FHIR to access electronic data from EHRs and other data sources. The reference architecture describes, at a high level, a consistent and reproducible approach for health systems to automate data exchange to facilitate reporting and data collection, thus providing a consistent foundation upon which use case-specific implementation guides can be developed.