• Tuesday, June 24, 2014RSS Feed

    EHR and Patient Safety: A Real Danger, Even for Experienced Users

    When it comes to electronic health records (EHRs) and patient safety, experience might be a great teacher, but it doesn't guarantee straight-A performance. According to a new study, even in longstanding EHR systems such as the one used by the Department of Veterans Affairs (VA) health care system, "many significant EHR-related safety concerns … remain."

    In a study of investigations of EHR-related safety violations launched through the VA's Informatics Patient Safety office (IPS) from 2009 to 2013, researchers looked at 100 closed cases at 55 VA facilities. Of those cases, 74 involved unsafe technology, and 25 involved unsafe use of technology, which authors write "most commonly involved the dimensions of people, clinical content, workflow and communication, and human interface." A majority of cases (70%) involved both unsafe technology and unsafe use. The study was published online in the June 20 issue of JAMIA.

    According to the study's authors, the problems documented in the research underscore the importance of constant vigilance for issues that could impact patient care—both at the purely technological level and within the "sociotechnical" realm, where users must interact with the technology.

    Researchers were able to use the data to tease out 4 types of safety concerns represented by the cases:

    • Concerns related to unmet data display needs. These concerns are related to what an EHR can and can't display, and occur when there's a "poor fit between information needs and the task at hand." Authors cited an example of a patient with a drug allergy who was given the drug due to a combination of human error and information gaps that made it seem as if an EHR warning was a "false positive."
    • Concerns related to software modifications. These issues center on the intended and unintended consequences of software upgrades or improper software configuration. By way of example, authors pointed to an upgrade to a disease management package that "erroneously escalated user privileges to place and sign orders."
    • Concerns related to system-to-system interfaces. Concerns in this area have to do with the ways systems talk to each other, and the possibility of one patient's records being mixed up with another's.
    • Concerns of hidden dependencies. Researchers identified several instances in which seemingly unrelated components of an EHR affect each other, such as the reassignment of a patient from outpatient to inpatient status resulting in the removal of certain medications from the patient's active medication list.

    Authors chose the VA system in part because of its standardized and well-documented approach to investigations of EHR safety problems, and in part because it was an early adopter of EHRs. The VA's level of experience, however, didn't eliminate the problems. "Having a mature EHR system clearly does not eliminate EHR-related safety concerns," authors write.

    Tackling these problems will take not only vigilance but an understanding that EHRs involve both technological and human elements. "Our study suggests that technology-based solutions alone will only partially mitigate concerns and that interventions to improve EHR-related safety should encompass the people, organizations, systems, and policies that influence how EHRs are used," the report states.

    APTA offers several resources on information technology and EHRs, including a webpage devoted to the use of EHRs.


    Comments

    Yes, we must constantly work on ways to improve EHRs. Even so, this study shows how GOOD we are already doing. Please share that message in Twitter as well as the "Real Danger" message shared earlier. Here is why: There are 1700 VA facilities and thousands of patient cases over the three plus years reviewed. Even so, closed cases with these errors are found at only 55 facilities (3.24%). Not perfect, but good. Also, with EHRs, we can have this discussion. With a paper process, it would have taken months and months to gather and review the data. Finally, even with the criticisms of how this is or could be shared and misinterpreted via social media, the four 'types' of errors are always worth reviewing by everyone managing or using EHRs, so thank you for sharing that message. Maybe share once again, with a positive twist on the message?
    Posted by Robert Latz on 6/25/2014 10:16 AM
    Leave a comment
    Name *
    Email *
    Homepage
    Comment

  • ADVERTISEMENT