Healthcare IT News ran a story regarding Texas Health Resources, Ebola, and the failure to link CDC listed symptoms for Ebola with the proper clinical workflow, and to record those symptoms along with the patient travel history from Africa in the Epic Systems E.H.R.
The question is not whether people or technology is to blame. The question is, now that the event occurred, how can we respond rapidly and how to prevent this from happening again?
Flexible and agile systems and processes (E.H.R. workflow, training) are required. Disease management protocols should be built into the workflow to capture the right clinical concepts. Hopefully, Texas Health Resources will be re-tooling their E.H.R. and people process immediately.
As we move from these quick reactions for this single event to avoid a crisis epidemic, we can think about better data. The transition from ICD-9 to ICD-10, will move the U.S. health system from a number of codes describing hemorrhagic fever symptoms–which might mean Ebola–to one specific code for Ebola. ICD-10 provides a specific code for Ebola virus, A98.4. Biosurveillance is enabled by better, more accurate data, to quickly track and catalogue diseases and readily share information. Clinicians don’t have to be ICD-10 coders, but they do have to accurately describe and capture the complete condition of the patient using the right clinical concepts to enable better data to work. But that is what clinicians were trained to do in nursing school, and medical school.
Then, we need to ensure that this information is entered – completely – into the EHR via agile workflows and that it can rapidly shared – in real time – not just as a retrospective report in 12 months with CDC or CMS for an Accountable Care Organization. Some commenters on the Healthcare IT article said they thought paper charts were better, and that if only we didn’t have an E.H.R. we would not have gotten into this mess. Those who pine away for a return to paper must be forgetting that paper is harder to share. We need rapid consumption of shared knowledge across the U.S. healthcare system. Paper doesn’t support that. But does Epic support ubiquitous sharing? That is, can you get this data if you you aren’t running Epic? How long and how much money will it take to re-engineer the Epic system and process to ensure that the question …have you recently traveled to Africa?…
is included when the patient presents with the CDC listed symptoms for Ebola the next time? What if the patient were being admitted and diagnosed during an encounter where the healthcare provider is using Epic E.H.R. in Tennessee? Could the health care provider and Epic customer in Tennessee immediately benefit from the modifications that were made to the Epic system in Texas? Sharing across the continuum of care, by setting, system, geography, and health organization is the future. It is mission critical in emergency responses to stemming a potential epidemic.
The answer seems to be to share more of this information in the cloud and present it easily to clinicians, health workers, the CDC, and the public, in abstract depersonalized form using mobile health, not in on-premise software systems in one health organization in Texas.
