Ebola, EHR and Teaching Moments

An Ebola patient is seen in the ED and discharged. The hospital points to its EHR software as the culprit. Alas, if it were only that simple.

In fact, as we all are now aware, the EHR has been exonerated (in an unusual and carefully worded clarification). The patient has died. And we as physicians and administrators are left to sort through the varied and conflicting media reports and question whether our own workflows, software and our training would have prevented a similar set of outcomes.

I was frankly skeptical of the initial explanation: A supposed flaw in the way the physician and nursing portions of EHR interacted. I didn’t see it that way. The problem was more textured, the solution more demanding, and the payoff transforming.

(By way of background, I have been certified and recertified by the American Board of Emergency Medicine four times. I have directed four emergency departments, including a Level II Trauma Center ED. I have spent 10 years as a Chief Medical Officer.)

From where I sit, the diagnosis and disposition of the patient remains solely with the treating physician. A careful travel history will always be part of the assessment of any patient presenting with symptoms (fever, cough, rash, vomiting, diarrhea, etc.) suggestive of a communicable disease.

Whether the triage nurse or the nurse performing the primary assessment captured, recorded and communicated an accurate travel history does not change responsibility of the physician to thoughtfully solicit and assess the patient’s travel history in making a diagnosis in formulating a treat plan.

The role of human error
I’ve also spent five years as a Chief Quality Officer. That, too, adds to my perspective. Physicians and nurses are human beings. They can be distracted; they can be fatigued. Even alert, well-rested humans are subject to making mistakes.

Consider the environment in which medicine, emergency medicine in particular, is practiced and you appreciate that error comes naturally. Knowing this, professionals working in quality use a systems approach to improve quality. Dr. James Reason advised that “we cannot change the human condition, but we can change the conditions under which humans work.” Using systems is the antidote to error prone humans.

One aspect of a systems approach to quality is identifying high-impact, mission critical events. A correct travel history with an infectious disease presentation is such a mission critical element.

Where you have a mission critical node, redundancy has long been advocated as an antidote. If system A fails, default to system B. In the case of travel history, two separate clinicians independently solicit and record a travel history.

Prior to formulating a diagnosis and treatment plan, the decision maker, usually the treating physician, reviews the other clinician’s notes and documents the concurrence of their findings, or where there is disagreement, addresses and resolves the variance.

In monitoring the systems operation, auditing the documentation of the review and resolution of the findings of all clinicians is a basic and effective action. Simply repopulating the physician’s electronic record with the nurse’s findings defeats the redundancy safeguard.

Big data and analytics
We all seek to learn from adverse events, and many hospitals and EHR vendors are reviewing and implementing Ebola-specific prompts as we speak.

But the real promise of our shiny new and powerful EHRs is not a system to respond to Ebola, but a system that utilizes big data and analytics to respond to dynamically emerging disease patterns.

We are in the earliest stages of our current IT transformation, one that treats the EHR as little more than an electronic phonebook. We provide the physician with all the data, in binary not paper, and let the decision-maker look for the information they need.

The next wave will bring profound systemic changes in the delivery of healthcare. It will bring an agile, analytic engine humming in the background of our EHR, scanning for a pattern in the structured and unstructured data, both internal and external, that flows from and around the individual patient.

We are nearing the time, as Stephanie Kremi, MD, writes this week, when “we will see headlines about how EHRs prevented outbreaks from occurring, not blaming them (even incorrectly) for contributing to a potential disaster.”