As a kid, I watched my mother go through a weekend bill-paying ritual. She would sort through a large and sometimes teetering stack of mail, setting aside certain envelopes. Then she would pull out her checkbook, a bulky, Costanza-like item, and initiate scribbling. Writing checks, addressing envelopes, licking stamps, and balancing numbers on her ledger. It was quite a production, though it seems quite archaic to those of us who use online banking.
Bills come by email, and are paid with a click or two— no checkbook, no stamps, no hassle. Other bills are paid automatically each month, an innovation so simple that the danger is that you might lose track of who you are paying and why. To me, and to others, I suspect, this represents technology at its best, making our lives easier so we can focus on more worthwhile endeavors (e.g., fantasy football waiver wire pickups). Other examples of such sublime simplicity are GPS navigation, online flight check-in, and certain (but not all) smartphones.
But for every divine invention, there is an example of technology gone awry, technology that makes lives complicated and aggrieved, such as iTunes upgrades, the remote control for my ceiling fan, and until recently, the electronic health record (EHR).
EHRs have received considerable press, and most certainly hold the promise to make our lives simpler and healthier. But they have not yet clearly delivered on their promise despite their central role in the federal health care overhaul and the billions spent on them. They are expensive and they glitch, and they can be clunky chains that shackle health care providers to their computers at the expense of face-to-face time with patients. Take, for example, the recent study entitled “4000 Clicks: A Productivity Analysis of Electronic Medical Records in a Community Hospital ED,” which precisely documented the secretarial nature of an EP's job in the current EHR era. (Am J Emerg Med 2013 Sep 20.) Hill and colleagues observed that EPs at a community ED with an EHR spent a whopping 43 percent of their time on data entry (orders and charting). Well, maybe not so whopping; there are some shifts where it feels like I barely look up from my keyboard long enough to realize it is time to go home.
I'll readily admit that I've suffered from click fatigue as I navigate through the process of computer-based patient care, often bombarded by excess information that is redundant or irrelevant. Nonetheless, EHR technology has improved vastly over the past few years, and it will not be long until the paper medical chart has joined relics of the Jurassic era. Like it or not, medicine is going electronic. The question is, however, will EHRs, like online banking, improve the quality of our lives? Up until now, the evidence of the benefits of the EHR — to patients and providers — has been mixed. That, however, is starting to change.
A recent JAMA study by Mary Reed, DrPH, and colleagues (including me, by way of full disclosure) adds to a growing body of evidence supporting the beneficial effects of EHR implementation. We examined the staggered implementation of the EHR across Kaiser in Northern California between 2005 and 2008, and compared “negative” outcomes such as visits to the ED and hospitalizations in 169,000 diabetic patients whose care was administered through either an EHR or a traditional paper chart. The study found that annual ED visits declined 5.5 percent, and annual hospitalizations declined 5.2 percent. There was no significant change in the number of doctor's office visits for patients with diabetes before or after the EHR was implemented. (2013;310:1060.)
“Since our study finds EHR-related improvement in care quality and outcomes without changes in office visit rates, this may reflect greater efficiency during visits or care delivery between visits,” Dr. Reed wrote.
These results are promising, but must be considered cautiously. Any study that looks back in time and attempts to discern which medical interventions were associated with improvements in patient health is subject to confounding, such that the findings could also have been affected by other factors. The confounders in this study were minimized (but not necessarily eliminated) by looking at an intervention (the EHR) that was implemented in a staggered fashion over a number of years and by using statistical correction that included fixed effects estimation and running multiple sensitivity analyses. We can be reasonably confident that the results of the study demonstrate a real benefit to patients, but we must also recognize that the actual magnitude of that benefit is difficult to estimate and may not be all that profound. And, of course, this study just scratches the surface of the issue of cost and time-effectiveness.
Nonetheless, this is a promising study that brings me pleasant visions of the future of the EHR, visions of a time when medical care across all venues and hospital systems can be seamlessly integrated and provide real-time feedback to patients and clinicians, a time when the EHR only asks me to click through alerts that actually matter and to when the screen itself presents a pleasant and intuitive interface. (See photo caption.) To a time when the individual variations in patient characteristics, even down to their genetic sequences, can help generate automatic recommendations about treatments and prevention strategies. A time when patients can be active participants in their own record, helping to verify and augment their information and use the tools of the EHR to make shared decisions with their doctor effectively.
Much work must still be done, but it seems to me the EHR has now turned the corner and that we can be confident that it will continue to evolve into a technological tool that will make our lives easier, better, and healthier.