By RICHARD GUNDERMAN, MD and MOHSIN MUKTAR (3)
Ironically, the same electronic health records (EHRs) initially designed as a tool to help physicians diagnose diseases have largely evaded diagnostic scrutiny. Talk to physicians who utilize them on a daily basis, however, and it becomes abundantly clear that today’s EHRs are ailing. They are adding hours to the physicians’ workday, siphoning attention from patient care, and sowing the seeds of demoralization across the profession of medicine. To address this problem effectively, physicians need to shift their focus from the symptoms associated with EHRs to the underlying diagnosis.
A key to arriving at the most accurate diagnosis is to cease treating EHRs as information technology problems and instead regard them as organic problems, not so different from the categories we would use in diagnosing a patient. Specifically, we need to seek out a known disease or diseases onto which many of the problems with EHRs can be mapped. In so doing, it is not our intention to stigmatize any disease or the patients who suffer from it, but instead to help physicians peer more deeply into the nature of the electronic malady with which they are wrestling.
We have concluded that contemporary EHRs meet the diagnostic criteria for autism. Introduced in its modern sense in the early 20th century, the term autism is derived from the Greek word autos, meaning self. Fittingly, autism was originally understood as a morbid degree of self-obsession. Today, we associate autism with impaired interaction, poor communication, and restricted patterns of behavior. Each of these features of autism can be found in contemporary EHRs, and for this reason, we propose that such systems — and to some degree, the individuals responsible for their design and implementation — be treated as autistic.
That EHRs both exhibit and lead to impairments in interaction is, to physicians required to use them on a daily basis, so obvious as hardly to warrant discussion. They demonstrate a nearly complete lack of intuition. They cannot tell what physicians are thinking or feeling, which makes it hard to build a working relationship. They are unable to look at tasks from the point of view of end users and often fail to grasp what physicians are attempting to convey. Not only are they unable to look physicians in the eye, but they also draw the physician’s gaze away from the patient. In the patient-physician relationship, they promote isolation.
This impaired interactivity negatively impacts the physician’s ability to remain focused on the most important aspects of a patient’s care. The physician is asking, “How do the patient’s symptoms, physical exam findings, laboratory tests results fit together into a coherent diagnostic picture?” while EHRs are stubbornly demanding adherence to a protocol of data entry grounded more in coding and billing than care. The physician wants to map out the patient’s disease and an appropriate management plan, while EHRs divert attention to another kind of map, the obtuse cartography of the system itself. In effect, EHRs supplant patients as the primary focus of the physician’s attention.
EHRs are equally poor at communication. They tend to operate with an idiosyncratic language that obscures the understanding of both physicians and patients alike. For example, they exhibit a marked predilection for checkboxes and pull-down menus. They do not respond to the particularities and nuances of each patient’s situation. They make it difficult for physicians to share the full range and depth of their experiences in a meaningful way. Focused on parameters too often tangential rather than directly targeted at clinical reality, they often seem out of touch, as if they existed in a world all their own.
EHRs also exhibit highly restricted behavior patterns, which manifest a high degree of compulsivity. The system may insist that certain information be submitted whether it is appropriate or not, fail to provide an opportunity to record other information, or demand that information be supplied in a specified order, even when it makes no clinical sense. Some behaviors, such as tortuous log-in protocols, are ritualistic in the worst sense of the term. Likewise, EHRs display remarkably restricted interests, often completely disregarding the most salient aspects of a case. Most such systems are extraordinarily resistant to change, even when enhancements would make life better for both patients and physicians.
As Osler said, “The good physician treats the disease, but the great physician treats the patient who has the disease.” To remedy the EHR’s autism, above all, we must know our patient, and we must accept that radical therapy will be required. We must insist that healthcare organizations put patient narratives before datasets, truth before expediency, and people before systems. Treating an autistic patient can be frustrating, and a favorable outcome is by no means assured. But surrender is not an option. Rather than following like sheep along the electronic paths that have been laid out for us, we must walk boldly yet humanely in new directions.
Richard Gunderman is Chancellor’s Professor at the School of Medicine, Indiana University. Mohsin Mukhtar is a medical student. This post first appeared on KevinMD. Republished at the request of the author.