Reimagining the clinical trial registry

Taking better advantage of EHR data

In addition to the features and functions proposed in the registry proof of concept, the non-profit TransCelerate Biopharma is exploring other areas of informatics, including better integration between clinical trial registries and electronic health records (EHR), which it says would provide an opportunity to link clinical research opportunities to points of care. Its eSource Initiative aims to modernize the way clinical trial sponsors and other stakeholders in the clinical trial ecosystem collect data and to facilitate optimal use of electronic data sources.

One use case involves examining EHRs for protocol feasibility and clinical trial recruitment for a specific clinical trial. TransCelerate is working with the health standards organization HL7 to facilitate interoperability by enabling the use of EHRs in clinical research, enabled by the use of the HL7 FHIR (Fast Healthcare Interoperability Resource) standard.

“The problem this is addressing is the ability to utilize clinical EHR data to support research activities,” explains Wayne Kubick, HL7’s chief technology officer. Data in the EHR may be used to feed clinical trial databases to ?eliminate the redundancy of re-entering data. He says, “You have to make sure that the health record is consistent with the clinical trial record, because if not, you run into trouble later with traceability and audits that may question the integrity of the data. The idea is to be able to tap into the EHRs and pull data to prepopulate case report forms in study databases, as well as a number of innovative capabilities to build on these linkages.”

Standard data framework

Kubick explains why that type of interoperability has been difficult in the past: “On the pharma side, there was a general lack of trust that the data was of sufficient quality for research purposes,” he says. Or it might be presented in a format of scribbled provider notes, rather than structured data. But now EHRs are collecting structured data that fits nicely into study databases: demographics, labs, vital signs, problem lists and treatments.

Pharmaceutical researchers wanted to get that data, but because every clinical information system was essentially unique and you had to have researchers deal with the IT department, who are trained to not let anyone touch their data, it was nearly impossible to do in a cost-effective way, according to Kubick.

So how does the FHIR standard potentially come to the rescue? “It gives you a standard, generic application programming interface that eliminates the need to develop custom integrations,” Kubick says. “It gives you a standard data framework based on the FHIR resource model. Basically, by eliminating the complexity of individualized, point-to-point interfaces, it provides the ?possibility of tapping into the generic APIs provided by the major EHR vendors, all of which give you a pretty consistent view of the data.”

In a recent “Connectathon” meeting in San Diego, HL7 and TransCelerate worked on several use cases: One is being able to tap into the EHR API and prepopulate the case report form for a study database; another is to try to gain more insight into when a patient enrolled in a clinical trial visits a care provider, which could provide much more visibility to the clinical research community about what is going on with the patient.