AMIA questions whether EHR data can be used for research
The use of electronic health records for clinical research offers great opportunities to facilitate medical research but there’s a long road ahead before digital records can reliably used for that purpose.
That’s the warning of the American Medical Informatics Association (AMIA), which yesterday filed comments responding to the Food and Drug Administration’s proposed guidance on using EHRs for research purposes.
Both hospitals and physicians are gathering large amounts of information through the electronic records systems they’ve installed in recent years, AMIA noted. But the professional association, which represents the nation’s leading biomedical and clinical informaticians, said it doubted whether those systems contain data of high enough quality that could support randomized controlled trials.
“With more than 96 percent of U.S. hospitals and 83 percent of U.S. office-based physicians using EHRs to deliver clinical care, we have an unprecedented opportunity to utilize digitized healthcare data for supplemental uses, such as clinical investigations,” AMIA said in comments. “However, we strongly caution the FDA from assuming EHRs are readily configurable for clinical investigations, even among more advanced institutions.”
In proposed guidance issued in May, the FDA covers using EHRs that interoperate with electronic systems supporting trials, and discusses ensuring the quality and integrity of data collected and used. The new draft does not include provisions under which the FDA would assess compliance of records systems.
What the FDA eventually decides in this area is important because facilitating the correlation between EHRs and clinical trials would provide significant benefits in speeding and cutting the costs of clinical trials, as well as streamlining the process of finding patients for research.
FDA previously issued guidance on electronic source data in clinical investigations, which acknowledged that data can come from various sources and be entered into the trial sponsor’s Case Report Form (CRF), which is a paper or electronic questionnaire. This could include data from EHRs—this latest proposed guidance focuses on such data.
Another best practice for using EHR data in clinical trials is ensuring the data is attributable, legible, contemporaneous, original and accurate. Further, FDA recommends use of ONC-certified EHRs and other health information technology because of clear differences in interoperability and keeping data confidential.
Non-certified EHRs may be used, but should be assessed to determine if adequate controls are in place to ensure data confidentiality, integrity and reliability. The assessment should include limited access to electronic systems; identification of authors of records; audit trails to track changes to data; and availability and retention of records needed for FDA inspection.
Citing a lack of technical standards and an overreliance on the assurances resulting from ONC’s Health IT Certification Program, AMIA said data generated through the routine course of clinical care would likely fall short of more rigid research standards for data quality and integrity. EHR data is meant to support the care of individual patients, rather than generate research quality data, the organization noted.
“Ensuring data integrity and tracking data provenance in clinical settings is incredibly complicated because multiple, authorized individuals contribute to the EHR and the specificity of audit logs varies widely,” the association said in comments. “If the FDA is interested in which data populated the electronic data capture (EDC) system or electronic case reporting form (eCRF) and where they originated along the continuum of care, the answer could prove extremely difficult and burdensome.”
AMIA recommended FDA update the guidance by pointing clinical investigation sponsors towards data warehouses that utilize a common data model. “These sources may have better semantic interoperability and data integrity compared to sources that remain in the EHR default data model,” the group said.
“If we want to reach the goal of a learning health system, connecting our care systems to our research systems is essential, but we must do it right,” said AMIA President and CEO Douglas B. Fridsma, MD, PhD, FACMI. “Ultimately, we want every patient encounter recorded in the EHR to add to our knowledge about how to do a better job with all patients.”
Health IT Now reacted to the FDA’s proposed guidance, saying more direction is needed for data standards, saying the lack of standards could jeopardize the quality of data used in clinical research.
HITN, a diverse coalition of healthcare providers, patient advocates, consumers, employers and payers who support the adoption and use of health IT to improve health outcomes and lower costs.
“Data collected electronically in the clinical setting or by patients themselves through wearable technology can be useful in clinical research, pre-regulatory approval, and post-approval observation,” HITN’s statement said. “However, one barrier that remains for product sponsors and investigators in leveraging these data is a lack of clarity, especially around acceptable data standards and formats, from the Food and Drug Administration (FDA) and other regulators. Valuable electronic data can be sourced from data registries, wearables, and other forms of health IT which this guidance does not include.”