A three-year study published in the May-June issue of Annals of Family Medicine, at first blush, doesn’t make a good case for EHR implementations. Its findings showed that for 16 practices in the eastern U.S. — treating almost 800 diabetes patients — outcomes measured by A1C, actually, were worse on average than 26 practices using dead trees for their health records.
That should give critics of EHRs and those who enjoy criticizing federal health IT policy in general as a giant government boondoggle plenty of ammunition. Great.
But, looking over the findings by the authors from the University of Medicine and Dentistry of New Jersey’s School of Public Health, it appears from their results that another principle’s in play we’ve heard consultants and analysts endlessly repeat: Installing an EHR system doesn’t fix anything. If there are problems in a provider’s workflow before flipping the switch on an EHR, those problems will persist despite the technology investment.
The researchers focused on EHR features that enable population care, a coming hotspot as accountable care organizations will hold physicians accountable for chronic disease quality of care measures among their patients.
“In EHR-using practices,” the authors write, “leaders should focus on engaging all members of the health care team in redesigning work to support efforts to improve population care–whether or not their current system supports these efforts.”
They authors go on at length to speak of meaningful use and education of physicians, making recommendations for the Office of the National Coordinator for Health IT and putting some responsibility on vendors for making EHR systems more usable. They also detail, quite thoughtfully, the limitations of their study (i.e., they did not choose the study participants to offer a true national reflection of the U.S. health system, and furthermore the providers in the study were early EHR adopters — things might be different now).
While the study does have its limitations, it’s good to see EHR implementations analyzed in the mode of a randomized clinical trial, which one would think would yield more accurate results than the more frequent vendor studies or surveys we see almost daily. More and more of these studies are cropping up in scholarly journals.
Now, all we need to guard against is a politically charged press manipulating the numbers to create falsehoods — such as “Hey! I thought EHRs were supposed to solve all our problems, and look here, these researchers call it a complete waste!” when in fact, time and time again these studies merely reaffirm that providers need to solve their workflow problems…before installing the EHR. And, as the authors put it, consider how that EHR implementation might change workflows — and prepare for it.