Are e-health records at fault for Ebola mistakes?

A design flaw in the e-health records system in Dallas may have made it difficult for an ER physician to see the patient’s travel history.

Ebola virus and electronic medical records
Thinkstock / Thomas W. Geisbert, Boston University School of Medicine

A Dallas man who became the first person diagnosed with Ebola in the U.S. has died. And while the hospital that treated him has retracted a claim that its electronic health records (EHR) system contributed to a lapse in his diagnosis, experts are skeptical that the system worked properly.

The problem isn't that the patient's recent travel to West Africa hadn't been documented. Instead, it's that a physician would have had to search the EHR for that history in order to find a link to Ebola, according to Judy Hanover, research director at IDC Health Insights.

"They'd have to click through to look at it and read it deliberately," she said in a blog post about the incident. "And in this case, those few extra clicks and their impact on productivity in a busy emergency room may have been a problem, if not an actual flaw in the EHR system."

Thomas Eric Duncan died this morning 7:51 a.m., Texas Health Presbyterian Hospital Dallas announced today.

Duncan had travelled to Liberia to visit family and, after returning to the U.S. on Sept. 20, he visited the hospital's emergency room with fever, headache and abdominal pain. The hospital, however, did not test Duncan for Ebola and sent him home. On Sept. 28, Duncan was taken by ambulance back to the hospital where he was admitted with acute symptoms.

When Duncan first visited the hospital "his condition did not warrant admission. He also was not exhibiting symptoms specific to Ebola," the hospital said in a press release. However, Duncan had told nurses that he'd just returned from Liberia and they did document it.

Hospital officials initially indicated that a nurse may have failed to tell the emergency room doctor about Duncan's recent trip to Liberia. Then hospital officials blamed a "flaw" in the hospital's EHR system for not communicating the information that had been inputted by the nurse.

"We have identified a flaw in the way the physician and nursing portions of our electronic health records (EHR) interacted in this specific case. In our electronic health records, there are separate physician and nursing workflows," the statement reads.

The very next day, however, the hospital retracted that statement and instead said "there was no flaw" in EHR system from Epic Systems and confirmed that the doctor did have access to Duncan's travel history. An inquiry to Epic by Computerworld was not immediately answered.

The question is, how easy was it for the doctor to access that information?

In an Oct. 3 statement, hospital officials said, "There was no flaw in the EHR in the way the physician and nursing portions interacted related to this event. The patient's travel history was documented and available to the full care team in the [EHR], including within the physician’s workflow."

IDC Health's Hanover, however, said a common design flaw with EHRs is that physician and nursing documentation is separate and that an entry by a nurse might not be immediately obvious to an emergency room physician or subsequent healthcare provider. In fact, physicians and other healthcare workers often complain they are less productive using EHRs than they were when they used paper records.

"The inability to restore productivity with EHR has clearly affected the business outlook and perhaps the quality of care for many providers," Hanover said. "We clearly can't afford lost productivity or mistakes in our emergency rooms."

Errors in patient care due to EHRs are "incredibly common," according to a report released last year by the American College of Physicians. Hospitals that are under EHR vendor contracts, however, can prevent physicians from speaking publicly about those problems by using gag clauses.

Most EHRs are still using first-generation software that took a paper process and simply transferred it to an electronic format without defining proper workflow practices or covering every scenario for an alert to be issued, Hanover said.

Hanover believes that problem may be solved with future generations of EHRs that are either cloud-based or connect to a cloud service that will allow for alerts from the U.S. Centers for Disease Control and Prevention, or from other agencies.

The use of EHRs has skyrocketed over the past few years due to regulatory pressure to implement the technology. Hospitals, physicians and other healthcare providers that don't implement EHRs face monetary penalties.

By the end of last year, about six out of 10 (58%) of hospitals had adopted EHRs -- four times the amount that had done so in 2010, according to recent studies published by the U.S. Office of the National Coordinator for Health Information Technology.

Additionally, the studies revealed that 78% of office-based physicians had some type of EHR system, but only 48% of them had the capabilities required by government regulators to meet specified criteria under its "Meaningful Use" program.

The studies also found big gaps in electronic health information exchange between one health provider and another, with only 14% sharing data with providers outside their organization.

Almost nine in 10 physicians responding to another survey indicated that at least one data management function was slower after adoption of EHRs. The study, performed by the American Medical Association about members of the American College of Physicians, also found that nearly 34% of respondents “saw more time spent on finding and reviewing medical record data with the EHR than without it.”

Dr. Randall Case, an emergency room physician and data infomatics expert who has worked for EHR makers such as Cerner, Siemens and IBM, said the problem with EHRs is multifaceted. The systems are "thrust" on physicians and nurses, who sometimes receive minimal training and then must adapt their workflows around them.

Just logging into an EHR can take 30 to 60 seconds for every patient and, in a busy ER, that can add up to "hours" when physicians have to walk to and from a computer terminal to do it, he said. And each EHR must be customized for every hospital's needs. Some undergo that customization process better than others.

"They essentially come up to loading dock with a truck full of parts and dump them. Then you hire an army of guys like me to assemble them," said Case, who now runs the consultancy Healthrecord Solutions in Irving, Texas.

EPIC's EHR, as other do, uses navigation icons that require several "click-throughs" to reach different sections of a patient's chart.

However, regardless of the complexity of the EHR, Case said he would not blame the EHR when a physician has a patient in front of him.

Another problem with EHRs is they create cognitive dissonance for users. Rather than writing symptoms on paper, as physicians and nurses have traditionally done, icons and links must be clicked through, and in emergency situations that can cause stress.

Years ago an ER doctor told Case it took 27 clicks to order a test, Case said. "Having worked for several of those [EHR] vendors, I think they've continued to improve workflows, [the] user interface and the learnability of their system, but they’ve made [only] marginal progress along those lines."

This story, "Are e-health records at fault for Ebola mistakes?" was originally published by Computerworld.

Copyright © 2014 IDG Communications, Inc.

7 hot cybersecurity trends (and 2 going cold)