A new study appearing in the Journal of the American Medical Informatics Association has concluded that a sample group of physicians made more data entry errors with a new EHR than in comparable paper records, according to a HealthcareITNews item.
Researchers studied progress notes created at a Michigan hospital, Beaumont Hospital of Royal Oak, Michigan, between August 2011 and July 2013. They looked at 500 notes created during that period, some of which were prepared before the EHR implementation in 2012 and some after. The charts contained five specific diagnoses which always include physical findings, including permanent atrial fibrillation, aortic stenosis, intubation, lower limb amputation and cerebrovascular accident with hemiparesis.
Upon analysis, they found that rates of inaccurate documentation were 24.4% with the EHR, versus 4.4% with paper records. Residents had fewer inaccuracies (5.3% vs. 17.3%) and omissions (16.8% vs. 33.9%) than attending physicians.
While this is no reason to throw the EHR baby out with the bathwater – after all, the physicians in question were learning a system for the first time – it’s still a troubling set of statistics. They are even more troubling given that EHR documentation errors can sometimes create patient safety problems of their own, especially in fast-moving care settings like the emergency department.
“There are new categories of patient safety errors” taking place in EDs that didn’t exist before EHR use became commonplace, according to Raj Ratwani, scientific director for MedStar Health’s National Center for Human Factors in Healthcare in Washington, D.C., who spoke with Kaiser Health News. For example, EHRs that only allow doctors to edit records for one patient at a time can make it harder to track ED patients, according to MedStar physician Zach Hettinger.
Without a doubt, the healthcare industry can’t afford to have its IT infrastructure creating new categories of safety errors or even making mistake-ridden documentation more common. Not only does this defeat the key goals for putting EHRs in place (improving care quality and efficiency), it could lead to a net increase in safety problems.
But as peanut-gallery observers like myself have been shouting for ages, the answer to the problem is fairly straightforward. EHR user interaction design has to be improved dramatically, and soon. This isn’t exactly a secret, but it seems that the issue is still treated largely as an academic discussion rather than one of immediate practical importance for providers.
I’m not sure why we haven’t made more progress on the user experience front in EHR design – or rather, which of the reasons can actually be addressed in our lifetime – but something’s gotta give.