Safety performance of EHRs vary widely

ByJanel Miller

Despite the broad adoption and optimization of electronic health records over a 10-year period, researchers reported “wide variation” in the safety performance of operational EHR systems.

Perspective from Andrew Gettinger, MD

David C. Classen, MD, MS, research professor in the department of internal medicine at the University of Utah, and colleagues noted that EHRs have been used to safeguard against medical errors, but recent evidence suggests that medication safety and overall safety problems are “still unacceptably high.”

Physicians have come to rely on [EHR systems] to ensure their patients get safe care.” The source of the quote is David C. Classen, MD, MS.

Researchers analyzed data from the National Quality Forum Health IT Safety Measure — a computerized physician order entry and EHR safety test — from 2009 to 2018. The test uses simulated medication orders that have either injured or killed patients in the past to determine how well hospital EHRs can identify those errors. Most of the 2,314 hospitals that participated had 100 to 399 beds; were part of a health care system; were located in urban areas; had private, nonprofit ownership; and/or were in the Southern United States. All hospitals had at least 1 year of test results.

Researchers found that among 8,657 hospital-years, mean scores on the overall test increased “modestly” from 53.9% (standard deviation, 18.3%) in 2009 to 65.6% (standard deviation, 15.4%) in 2018. During the same timeframe, mean hospital scores for basic clinical decision support increased from 69.8% (standard deviation, 20.8%) to 85.6% (standard deviation, 14.9%), and scores for categories representing advanced clinical decision support rose from 29.6% in 2009 (standard deviation, 22.4%) to 46.1% (standard deviation, 21.6%).

Classen and colleagues emphasized that there was significant variation in safety performance across vendors, and vendor choice explained about 10% of the variation they observed.

“One of the selling points by EHR vendors to physicians was that these systems would markedly improve the safety of care for their patients,” Classen told Healio Primary Care. “This was felt to justify the expenses and big impact on physician workflow. Our study shows that is not the case. This is important because physicians have come to rely on the system to assure their patients get safe care.”

In a related editorial, Raj M. Ratwani, PhD, director of the MedStar Health National Center for Human Factors in Healthcare, encouraged hospitals and other health care facilities to immediately begin using existing EHR safety assessment test case scenarios regularly, particularly after EHR system downtimes, upgrades, or other changes “to advance EHR safety.” He added that long-term solutions include making EHR safety assessment methods part of a hospital’s accreditation process.

References:

Perspective

Andrew Gettinger, MD

The HHS Office of the National Coordinator for Health Information Technology has worked with Dr. Classen for many years. He and his colleagues are established and well known researchers in the field of health IT safety. Consequently, the JAMA Network Open article is an important report from a very respected team.

Classen and colleagues indicate that EHRs have improved, but not to the level that they, we, or the clinical community would want, which suggests that there are continuing opportunities to make EHRs better. For me, it was interesting to see that the drug-allergy important interaction was identified 98.4% of the time, whereas the drug-diagnosis contraindications were only identified only 33.2% of the time. More importantly, there was great variability between institutions using the same underlying EHR software, suggesting that initial implementation decisions and keeping the software updated may have contributed to the findings.

As required by the 21st Century Cures Act, my agency is currently developing a system where clinicians and hospitals can report issues or problems with software. The public can review the drafts to read more and to provide comments on our proposed system. We anticipate that when completed, the system will allow anyone to see how some of these EHR software systems are actually performing.

Andrew Gettinger, MD

Director, HHS Office of the National Coordinator for Health Information Technology

Disclosures: Gettinger is employed by HHS.

Sources/Disclosures

Collapse

Source: Classen DC, et al. JAMA Netw Open. 2020;doi:10.1001/jamanetworkopen.2020.5547.

Disclosures: Classen reports receiving grants from the Gordon and Betty Moore Foundation and Robert Wood Johnson Foundation and serving as an employee of Pascal Metrics, a federally certified patient safety organization, outside the submitted work. Ratwani reports receiving funding from the Agency for Healthcare Research and Quality. Please see the study for all other authors’ relevant financial disclosures.

Read more

Safety performance of EHRs vary widely

Perspective

Related Content