ONC data shows potential link between EHRs and patient harm

Register now

Publicly available electronic health record surveillance data shows a possible link between EHRs and patient harm, according to researchers from the MedStar Health National Center for Human Factors in Healthcare.

That’s the finding of a research letter published last week in the Journal of the American Medical Association.

MedStar researchers contend that their results represent the “first-of-its-kind” examination of EHR surveillance data collected by the Office of the National Coordinator for Health IT, with all publicly available surveillance data—from Jan. 28, 2016, to June 24, 2019,—analyzed.

“This is a unique analysis,” contends Raj Ratwani, director of the MedStar Health Human Factors Center and an author of the JAMA research letter. “Nobody’s really done this kind of analysis with those data before.”

Investigators looked at more than 350 reports of EHR issues believed to violate ONC’s health IT certification program to determine whether identified issues had the potential for patient harm.

Each EHR product certified by the ONC receives a unique certified health product listing identification. What researchers found was that 39.5 percent of surveilled product identifications had an issue with a potential for patient harm.

In addition, based on required product use reporting, researchers estimate that 786 hospitals and 37,365 individual providers may have used the products with these potential patient safety issues. At the same time, they acknowledge that “although products associated with possible patient harm issues are widely used by healthcare facilities, it is unknown whether these issues resulted in actual patient harm.”

Nonetheless, Ratwani makes the case that “four out of 10 products that were surveilled have a potential safety issue and that’s certainly of concern” as well as the fact that the product IDs involved nine different vendors, “giving an idea of the reach of these potential issues.”

Ratwani gave an example of a possible patient harm found in ONC’s EHR surveillance data that involved one particular product. “When you enter the medication dose, the decimal point is removed from the entry,” he notes. “So, if a physician is placing an order for a medication at 2.5 milligrams that would get processed by the EHR as 25 milligrams. Depending on the drug and the patient, that could be a significant overdose that could lead to patient harm.”

Ratwani and his colleagues point out in their research letter that ONC’s EHR surveillance of vendor products may occur in response to a reported issue (reactive) or can be randomly selected for investigation (randomized).

“Most issues were identified through reactive surveillance, with 41.8 percent of product IDs reactively surveilled having a possible patient harm issue, although randomized surveillance was infrequently performed,” state the letter’s authors.

In response to MedStar’s research letter, an ONC spokesman contends that the agency’s EHR surveillance program is “working and doing what it is supposed to be doing—ensuring that products meet certification requirements.”

Further, the agency spokesman points to the Certified Health IT Product List (CHPL), a listing of all products successfully tested and certified by the ONC health IT certification program, and the fact that all products listed on the CHPL have been tested by an ONC-Accredited Testing Laboratory (ONC-ATL) and certified by an ONC-Authorized Certification Body (ONC-ACB).

“The open CHPL is important and a very important public good our program provides,” adds the spokesman. “It shows that reactive surveillance is working. The findings demonstrate the need for continued assessment of conformance in the field, which is why we have a complaint process in place, beginning with the ONC-ACBs and including our health IT feedback system for people to submit their complaints and concerns.”

In addition, he notes that in ONC’s proposed interoperability rule, the agency has proposed real-world testing that will be reported publicly on the CHPL.

However, looking at the body of evidence from MedStar’s research as well as other studies, Ratwani warns that based on the data, “it’s overwhelmingly clear that there’s a connection between the use of EHRs and safety issues—and those are not just usability issues.”

Separate from usability, Ratwani concludes that “there are issues with the technology that have direct consequences for patients,” and they must be addressed. He contends that what’s needed in healthcare is a “more robust reporting program that would allow clinicians to voice those concerns.”

Currently, EHR product surveillance is primarily focused on ONC’s health IT certification rules, “but there may be a whole host of other safety issues,” Ratwani adds. “We also need to be developing better proactive safety surveillance techniques. Instead of relying on reactive surveillance, we need to be thinking more proactively and look at using machine learning approaches and other computer science techniques to identify potential issues before patients are actually harmed.”

For reprint and licensing requests for this article, click here.