Blog | Friday, October 24, 2014

What I learned from Ebola


I’m sitting at my dining room table trying to reflect on and process the events of what, without a doubt, will go down in the annals of infection prevention as a pivotal point in time. Hospitals across the country furiously raced to prepare for Ebola, propelled by the unfortunate news of transmission of the virus to 2 nurses at Texas Presbyterian Hospital in Dallas. I’ll share with you what I think are the lessons of this incredibly interesting week:

Texas Presbyterian Hospital isn’t the exception, it’s the rule. It’s easy to be the Monday morning quarterback and criticize the emergency medicine providers for initially missing the diagnosis of Ebola, but given that this was the first case to ever present to an emergency department in the U.S., it should not be surprising. In the process of diagnosis physicians are trained to use probability in their reasoning. And Ebola simply wasn’t on their radar screens.

It’s also important to keep in mind that even today given everything we know, fever in a returning traveler from Liberia is most likely not caused by Ebola virus disease. Malaria remains a much more common diagnosis. For this reason, our Ebola plan reminds physicians to consider infectious diseases consultation in the setting of a person under investigation for Ebola, so as to avoid having a patient die of falciparum malaria while Ebola is being ruled out.

In addition, there may have been, and likely were, systems issues at play. There are many distractions in the hectic environment of an emergency department that may have had impact as the physician worked through Thomas Duncan’s case. Nosocomial transmission to healthcare workers would have also likely happened at almost any hospital with the exception of the four hospitals that have a biocontainment unit. While American hospitals have made great strides in reducing health care associated infections over the last decade, the challenges posed by Ebola virus in terms of the prevention of transmission are unparalleled.

The efficacy and effectiveness of personal protective equipment (PPE) need to be considered. By efficacy we mean how well PPE works in the ideal setting to protect the healthcare worker. Effectiveness is how well it works in the real world. For most pathogens, this difference is likely quite small. Not so for Ebola. Removing PPE in the Ebola setting without contaminating yourself is a Herculean effort, and we are dealing with what Dick Wenzel calls “an unforgiving virus.“ Before Ebola, the implications of minor errors in doffing were trivial. Now they’re life-threatening. An article in today’s New York Times sums it up beautifully:

Debra Sharpe, a Birmingham, Ala., biosafety expert, has overseen safety at a nonprofit laboratory that researches emerging diseases and bioweapons, and has run a company that trained workers to handle biological agents ... “It’s totally shocking ... It would take me anywhere from 4 to 6 weeks to train an employee to work in a high containment lab in a safe manner. It’s ludicrous to expect doctors and nurses to figure that out with a day’s worth of training.

To her comments I would add that the challenging setting of an ICU with an Ebola patient having 10 liters of vomiting and diarrhea per day is nothing like the controlled environment of a specialized laboratory dealing with contained aliquots of the virus. How well PPE works in the lab approximates efficacy. How well it works in the ICU is a measure of effectiveness.

The most advanced ICU in the best U.S. hospital is not a biocontainment unit. It’s absurd to think that the standards of a biocontainment unit can be met outside of that special setting. These units have special physical layouts with lab facilities, specimen dip tanks, employee showers, and autoclaves. They were created and supported with federal funding, and their providers have had ongoing training over years. So we need to realistically attempt to match the facility with the expected function: all hospitals should be proficient at rapidly identifying a potential Ebola patient, quickly isolating them and providing initial care, but once the diagnosis is confirmed, these patients should be transferred to a specialized biocontainment unit if a bed is available.

We need to think about exposures differently. In infection prevention, we tend to classify exposures to infectious agents on the basis of whether the exposure was protected: Did the nurse have on an N95 mask when she treated the patient with tuberculosis? Did the young man wear a condom when he had sex last night with an HIV-infected man? Typically, unprotected exposures pose greater risk of infection than protected exposures. In Dallas, the same paradigm was applied: the unprotected healthcare workers in the ER who evaluated Mr. Duncan before he was suspected to have Ebola were thought to be at higher risk than those who cared for him in the ICU will full PPE. This turned out to be wrong. Early in the course of Ebola the infectivity is low, as demonstrated by the fact that none of Mr. Duncan’s unprotected household contacts became infected. Late in disease, infectivity is very high and two nurses in gowns, gloves and face protection became infected.

Equipment and supplies for state-of-the-art care are inadequate. Several of us tried to find a stethoscope without ear tubes so that auscultation could be performed without bringing a device close to your face. We had no success. Much has been made of the fact that the Dallas nurses used PPE that didn’t cover their necks. This was even noted in an editorial in the New York Times. However, almost all (if not all) products that provide neck coverage, including bunny suits, are difficult to doff, making self contamination likely. Fortunately, our hospital has an in-house seamstress who rose to the occasion and rapidly began designing an item to cover the neck that is easy to remove. In addition, the supply chain for PPE is tenuous. Already, many items are on allocation and the national supply for some is not robust. Just-in-time manufacturing processes are not advantageous in the current situation.

Investment in infection prevention infrastructure and research is necessary. The health care system in the U.S. has talked a good game regarding the importance of infection prevention, but if budgets are statements of what we value, infection prevention has been a stepchild. Ebola should be our wake up call. Funding is needed to answer basic questions of infection control and to train hospital epidemiologists. Mandates for all hospitals to have infectious disease trained hospital epidemiologists should be considered. New models for compensation of infectious diseases physicians must be developed to encourage young physicians to pursue training in our field.

It was a truly challenging week. But from an infection prevention standpoint, it was challenging in a really good way. It allowed us to collaborate with experts across the health system and think creatively with them, while providing us an opportunity to demonstrate the value we add. I am very lucky to work with an amazing group of epidemiologists and a strong leadership team at the University of Iowa. And the Society for Healthcare Epidemiology of America (SHEA) staff did an outstanding job of promoting what we do in the mainstream media.

Lastly, we must keep all of this in perspective. Every issue I have talked about in this post is a first world problem. The tragedy of what is happening in West Africa remains incomprehensible.

Michael B. Edmond, MD, FACP, is the Chief Quality Officer at the University of Iowa Hospitals and Clinics. This post originally appeared at the blog Controversies in Hospital Infection Prevention.