ASPR TRACIE: Lessons Learned on Crisis Standards of Care in Kenema

Alternative Text for Time-Based Media

The following is a text alternative description for Lessons Learned on Crisis Standards of Care in Kenema

[The video begins with music and a title screen that includes the ASPR and TRACIE logos.] The video features John Hick, M.D. ASPER TRACIE Lead Editor (Detailed to HHS ASPR), Emergency Physician and Deputy Chief EMS Medical Director, Hennepin County Medical Center as the moderator of this session and Dr. Lewis Rubinson, MD, PhD, Director - Critical Care Resuscitation Unit at the University Maryland School of Medicine as the featured speaker. Dr. Rubinson will share various Powerpoint slides during his presentation titled, " Crisis Standards of Care. Real Life Lessons During Ebola Outbreak in West Africa".  Some of those slides will be highlighted as "Descriptive Text" in this transcript.]

[MUSIC]

Narrator: Welcome to “Lessons Learned on Crisis Standards of Care in Kenema” brought to you by the U.S. Department of Health and Human Services Office of the Assistant Secretary for Preparedness and Response, or ASPR.

ASPR recently developed the Technical Resources, Assistance Center, and Information Exchange (or TRACIE) to meet the information and technical assistance needs of healthcare professionals, emergency managers, and others working in disaster medicine, healthcare system preparedness, and public health emergency preparedness.   

Crisis standards of care is a critical issue facing the ASPR TRACIE audience. In this program, Dr. Lewis Rubinson discusses his experience serving as a physician consultant to the World Health Organization in Sierra Leone during the 2015 Ebola outbreak and how the United States might respond in future events.

 Dr. John Hick:  The TRACIE Team wants to welcome you to today's broadcast session with Lewis Rubinson, who is a associate professor of medicine at the University of Maryland and also the director of the critical resuscitation unit at the R Adams Cowley Shock Trauma Unit in Baltimore.

Lewis has an international reputation as both a critical care physician as well as a disaster responder and spent significant time in Sierra Leone during the 2014 to 2015 Ebola crisis.  He's going to share some of the lessons he learned while in country with us today. 

Dr. Rubinson:  So I'd like to take the opportunity to thank the TRACIE Program and the Office of Emergency Medicine and ASPR for inviting me to speak today about an experience in West Africa and its potential implications for crisis standards of care.  Before I get going with my talk today, it is important to put up my relevant disclaimers and disclosures.  Most importantly, I was a consultant for the World Health Organization in Kenema in September 2014.

And none of what I'm gonna talk today has been sanctioned or should be considered official WHO policy.  It just reflects my opinions and my experiences, but again, it should not be considered officially sanctioned conversation.  The rest of my disclosures are financial consultation that I've done, and none of those are relevant to this talk, but I've put that up there just for people's knowledge.

 So when I was asked to give this talk, I was asked to take a real-life situation, so what had happened in Ebola especially during the peak of Ebola outbreak in Sierra Leone, part of the West African outbreak, and use that as ground truth for being able to look at the implications for some of the crisis standards of care work that's been worked on in the U.S. in the past decade, decade and a half.  I'm gonna assess what I believe worked and also discuss where opportunities for improvement exist. 

I'm gonna highlight some crucial shortfalls -- and unfortunately there were some shortfalls, as with any overwhelming crisis and outbreak -- and, hopefully, that will ultimately lead to some actionable potential improvements.  And most importantly, this talk is to help part of the conversation so that we work as a community to make sure in the future we're better prepared and we respond better than we have in the past.

And then, importantly, my talk is not meant to be negative or pejorative towards folks.  There are many people who risked their lives, some who lost their lives during this outbreak, and many people who worked fervently for the outbreak.  I think there's a number of shortfalls that I will discuss during this talk, but again, it's not meant to point fingers at any one agency or organization to say that they did wrong. 

I just think that we could've overall done better.  So when we talk about crisis standards of care, as Dr. Hick and Dr. Hanfling and others have taught us, it's about a balance of need and resources.  And the expectations for an outbreak, I don't want this talk to be Pollyannaish and to have people believe that, you know, we're just gonna fly in a bunch of resources and everything's gonna get better because there are some uncomfortable truths that happen during outbreaks, especially in limited resource countries.

Those are: people will die, likely due to the outbreak.  But importantly, what this talk is gonna be about is separating out the deaths due to the underlying disease versus the deaths due to the secondary disaster of an unorganized and an inefficient response.  The difficulty is during any outbreak, resources which need to be brought to bear need to be linked to knowledge and to data. 

And situational awareness by definition is always lagging in a response, and it creates dilemmas when you're trying to appropriately allocate resources.  And they're also dynamic, right?  So resources and needs are changing, and our data systems don't always reflect those needs.  And then, lastly, the environments which are most susceptible to these outbreaks typically don't have the resources organically within their own country or within their own region to be able to respond. 

And they also have difficulty with situational awareness, so by the very nature of outbreaks in resource limited environments, there is going to be enormous impact on the population.  And as I mentioned, unfortunately some people are gonna die.  What this talk is about is trying to minimize the deaths due to just an inefficient or a chaotic response.

It's important to reflect that there's been a lot of excellent work, some of which has been sponsored by ASPR and AHRQ and other groups in terms of being able to develop the conceptual framework for crisis standards of care.  And I'd point you towards the Institute of Medicine work that Dr. Hick, Dr. Haflin, and a number of others have worked on in the past decade that really, I think, has advanced our thought process on how to start dealing with different mismatches between need and resource.

Dr. Rubinson:  And this slide again reflects a lot of the conceptual framework that has gone a lot further from, you know, in the late '90s, early 2000s; it was just a conversation.  I think the conceptual framework has now got a lot more meat on the bone, and there's a lot more detail on how to do this. 

Dr. Rubinson:  Even in the critical care world, I've been a part, as many others have been, in terms of trying to figure out how would we do this if the critical care needs in a developed country far surpassed the available resources. 

And in 2007, we had a meeting in which we put together a lot of government and professional society academic partners to be able to discuss how we should do this.  And I think it's worthy to go over a few of the key points.  The first thing is all effort should be undertaken to surge up, to be able to manage the need before we start triaging people to the potential for death. 

Secondly, modifications of care should be based on -- they should commensurate based on the mismatch of need and resource, and they require data to be able to make those choices.  And they also need to be dynamic because as more resource is brought to bear, it's important to try to provide the usual care as soon as possible again.

When lifesaving care is being withheld, and there will be some events where that is true, it's very important that all resources are requested.  So you shouldn't isolate and start triaging and potentially triaging people to where they will likely die from not getting the usual care without requesting the cavalry to come. 

It doesn't mean that the cavalry will be able to get there immediately.  But sometimes people can get lost in their own isolated community of not recognizing that there are other resources that can be brought to bear, whether they're just down the street or coming from afar.  And most importantly, when we undertake these efforts, there needs to be iterative improvement.  So if this is a sustained response, six hours, there's gonna be a lot of chaos. 

Six weeks should be better than six hours.  Six months needs to be better than six weeks, and the care needs to get smarter and better over time with a strategy of how we're improving.  Ultimately, that requires some sort of data stream to be able to understand, what did we get wrong?  What did we get right?  And then continually improving the process so that again at six months, we're not at the same point as the first day of a response.

Well, how have these concepts really played out?  In resource-limited countries, unfortunately they play out on a chronic basis.  In resource-intense countries, they actually have almost never had to be utilized in the past two decades, which is great.  So the concepts are well developed, but they actually have not really been tested in a realistic environment, and again, that's part of why I'm speaking today, although it's gonna be extrapolating from a resource-limited environment. 

Even during outbreaks that we have had, SARS and H1N1, the systems have come close to failing, but they have not failed.  So we really have not needed to implement these crisis standards of care in the extreme nature where we're actually triaging people to not get life supportive care.  Thank God we haven't had to go through that yet.  But again, we're not sure if our concepts are gonna be able to stand up if those events actually occur. 

So here's a few lessons -- at least Rubinson's lessons from West Africa.  Take them for what they're worth.  Other people may have different lessons, but these are a few things that I brought home with me.  The first is there are usual opportunities to mitigate public health emergencies before they become so catastrophic.  So when we look at Ebola, a number of people think this was a new virus and we were learning new things. 

Yes, there are new things that were learned because the numbers were so dramatic, but really, there have been responses for Ebola for at least four-plus decades, at least from the time that it's been identified in the mid-'70s.  And probably even before then, Ebola outbreaks have been occurring.  And they occur with relative common frequency, but they occur in places that typically are isolated and rural. 

So a lot of the outbreaks tend to peter out before international aid is necessary.  The public health measures typically work and have worked previously in the past.  That was mostly case finding, contact tracing, isolation.  Very little healthcare actually was part of the original outbreak responses and then proper management of dead bodies.

In terms of what care should we provide, again, even though outbreaks have been happening for a long time, there was some knowledge base, but honestly it was based on very limited data in very resource-limited environments, and the true knowledge base was suspect. 

Dr. Rubinson:  And this next slide just demonstrates the timeline where again there have multiple outbreaks in the past. 

The important thing to recognize, though, is the biggest outbreak previously, the Gulu outbreak, was under 450 patients.  And most outbreaks are considerably smaller. 

• 1976 – First outbreak of Ebola in Zaire. 318 Cases, 280 Deaths.
• 1976 – Lab infection by accidental stick of contaminated needle.   1 Case, 0 Deaths.
• 1976 – Outbreak in Gabon in gold mining camps in the rainforest. 1 Case, 0 Deaths.
• 1989 – Ebola Reston introduced into US quarantine facilities by monkey from the Philippines. 0 Cases, 0 Deaths.
• 1995 – Outbreak in DRC Index patient worked in the forest adjoining the city. 315 Cases, 250 Deaths.
• 1996 – Lab contamination in Russia. 1 Case, 1 Death.
• 2000-2001 – Outbreak in Uganda associated with funeral attendance and caring for patients with Ebola without proper PPE. 425 Cases, 224 Deaths.
• 2001-2002 – First outbreak reported in the Republic of the Congo. 57 Cases, 43 Deaths.
• 2007 – Outbreak in DRC. Last confirmed case on 10/4 and last death 10/10. 264 Cases, 187 Deaths.
• 2007 – Outbreak of a new strain of Ebola in western Uganda.    149 Cases, 37 Deaths.
• 2014 – Ebola outbreak in West Africa; largest in history.      27,000+ Cases, 11,000+ Deaths.

Source: https://www.cdc.gov/vhf/ebola/history/chronology.html

Dr. Rubinson:  When we get to -- got to this outbreak, we know that it probably started in December but wasn't really recognized until late winter, early spring of 2014, where there was an area of rural Guinea where the outbreak had taken hold, and then ultimately it took hold in Conakry, which is the capital of Guinea. 

This is the first time Ebola had really spread to an urban environment in west A -- not only in West Africa, but any urban environment.  And that should've set off an enormous amount of public health flares that this was gonna be a situation that we had never experienced before.  Certain groups like Doctors Without Borders recognized that this was a very, very different situation, but most of the world just waited to see how it was gonna play out. 

And unfortunately, the number of cases started to rise dramatically as the summer of 2014 happened and especially as we moved into fall of 2014.  So again, remember I had said that no outbreak really had exceeded more than 425 patients or so, and now we were talking about thousands of patients in an area of the world that had never seen Ebola before, West Africa, which had limited resources. 

And ultimately, this has led to just under 30,000 documented cases with an enormous amount of burden in the West African population.  When we think about whether or not these countries can really -- had been able to mount a response, Sierra Leone as well as Guinea and Liberia were countries that were coming out of civil war.  They were incredibly impoverished.  They have limited economies. 

Dr. Rubinson:  Wonderful people, but really, if you think about the cases in Dallas, the amount of resources that it took to be able to really reduce transmission in a resource-intense place like Texas, trying to do that in a country that more than half of the patient -- more than half of the people in the country are functionally illiterate; they're very young because of war and also not long life expectancy, you can imagine that this was gonna spread like wildfire.

Not because they aren't capable people, they just needed more assistance and infrastructure to be able to respond.  And this is just a picture which reflects kind of the economy of what you see. 

Dr. Rubinson:  There's a lot of -- there's a lot of street side economy that happens, a lot of limited infrastructure.  There's no real trash pickup or anything that we would think of for modern kind of conveniences to be able to reduce transmission. 

So you can imagine once a disease gets into these urban environments that it's very difficult to find who has the disease and to be able to put measures and to stop transmission right away.  The world's assistance, while assistance did come, it came very late.  Remember I showed the graph of where the cases really started taking off in the summer and the fall, it wasn't really until the end of fall and winter where most of the world responded. 

The U.S. did send a number of folks to be able to assist.  Much of the U.S. assistance was in Liberia, which was very much needed, but on the most part, even with the U.S. dragging much of the rest of the world along trying to respond, we brought very limited resource for what was a tragedy that was unfolding in West Africa. 

Dr. Rubinson:  And as I mentioned before, there were pillars of Ebola virus disease control that have worked in the past. 

Dr. Rubinson:   Clinical care was really one of the new ones, and the idea was if you could put people in an Ebola treatment unit and you can reduce their mortality, you would get people to come out.  And either the communities or themselves would self-refer and identify people who are symptomatic so that you can reduce transmission in the community. 

In addition, it's just the right thing to do as human beings.  When people get sick, we're supposed to be able to provide the resources as the world to be able to care for them.  So I arrived six months into the outbreak.  I went to Kenema Government Hospital, which was in southeast Sierra Leone.  It is a hospital that has a tradition of being a loss-of-fever referral center.

And when I got there, I was informed -- or prior to going, I was informed -- I would be the only WHO clinician on site at Kenema, and I was supposed to be a consultant to the Sierra Leonean clinicians.  And the idea was, the WHO was providing expert consultation to the locals to be able to respond internally. 

But when I got there, it was clear that there was no Sierra Leonean physicians that were working in the Ebola treatment unit, and the person who I was replacing unfortunately contracted Ebola and was evacuated back to the U.S.  So, in fact, there were no clinicians that were working in the unit for several days.  Again this is six months in, and Kenema was considered the referral facility for the entire country. 

So Kailahun, which was an MSF Ebola treatment unit, was still operating on the easternmost part of the country, but everyone from Freetown and areas where the outbreak were starting to take hold were being put in ambulances and sent to Kenema, and there was no real clinical staff at Kenema to care for them. 

So when I got there, ultimately, there were five WHO clinical staff for the ETU, some community epis and two additional docs who were sent as part of the outbreak investigation for the physician who contracted Ebola prior to me getting there.  We had the CDC lab as well that was embedded with us.  And that was -- we didn't know how many patients were in the unit, but ultimately turned out to be between 100 and 140 were in that unit with that -- with that footprint. 

And just if we think about what we normally bring with the National Disaster Medical System for far fewer patients who have far less illness, we were incredibly understaffed.  And there was actually staff in the country, but most of the international staff were working in Freetown.  And there was a small satellite staff that were working in the only referral ETU in the country.

Lesson number two:  Healthcare settings are likely to be the concentrated source of transmission in most outbreaks that cause serious illness.  Transmission tends to increase with the level of illness, and people who are more ill are likely brought to healthcare settings.  So you can imagine it's the perfect storm for being able to transmit disease, and unfortunately transmission will cause a problem with capacity and capability. 

So just as the healthcare setting is trying to scale up to be able to care for these patients, they're struggling with trying to keep their staff safe and their other patients safe.  And so the biggest part of lesson two is -- a "failure is not an option" lesson must be that we must understand transmission early on and it must become a priority that's fully resourced to truly understand modes of transmission rather than just using dogma to be able to guide how to keep people safe. 

So by the time I got to Kenema, a number of very heroic and experienced people had already gotten sick and had already died from Ebola.  Some of their senior nursing staff at Kenema Government Hospital and Dr. Sheik Humarr Khan, who was the loss of fever clinician at Kenema, had all gotten sick, and they had all passed away in the summer before I'd gotten there. 

And you can imagine that that is -- creates an enormous struggle for people to continue to carry on and to continue to care for their community.  Many of the nursing staff that were at the Ebola treatment unit came from outside areas and were paid an additional risk pay to be able to provide care, but there was no experienced senior leadership that was still guiding them because unfortunately, the senior matrons had either been sick and died or seen their fellow matrons die and were still recovering from the overall tragedy that was going on. 

Some of the international workers had also gotten sick, so Will Pooley from the UK had gotten ill in Kenema and Dr. Ian Crozier, who got ill right before I got there, also had gotten sick at Kenema.  So you can imagine there was a level of extreme desire to try and figure out, how do we stay safe and still manage the overwhelming number of patients that we had? 

Dr. Rubinson:  And healthcare workers were at high risk during the entire Ebola outbreak.  You don't want to be one of the big circles, which is where a number of healthcare workers got sick.  Kenema again is in the southeast side where a number of workers got ill.  Throughout the Ebola crisis, healthcare worker infection really created a struggle to be able to respond. 

And while I was there in September, it was starting to get reduced.  We were starting to maybe figure out how to keep people safe, but it was still happening. 

Dr. Rubinson:  So even though I was there six months in, where you'd think we would've really figured out by then how to keep safe, we were essentially trying to regain a safety culture at Kenema, which was most of our focus. 

I was lucky that I happened to be deployed with people who had a lot of experience either with MSF or with other viral hemorrhagic fevers, and we were able to figure out how to put together a personal protective equipment program and an infection control program to keep us safe.  But again, it was not a turnkey approach that you would think should've happened by then.  It was really, we were making it up on the fly. 

And -- and the struggle was, the recommendations were all over the place.  The recommendations in the U.S. were really saying that you could use limited PPE to be able to manage these people, and that was at direct odds with what Doctors Without Borders was saying.  And, unfortunately, as healthcare workers continued to get ill, the investigations were not very strong in terms of being able to decipher, did people get ill in the Ebola treatment units using proper PPE? 

Or did they get it from an unknown, unprotected exposure?  Ultimately, it just created a lot of fear, and almost all the attention was spent on “how do we stay safe” rather than “how to we strategically manage patients?”  And obviously, the personal protective equipment that we started using there was similar to what MSF and others had been using for -- for years, and it came with a price. 

And the price is that it was very, very hot, and patients with Ebola need 24/7 care.  Many of them need to be able to -- need people helping hydrate them and watch them as they get more ill, but the personal protective equipment really restricted the amount of hours that you could be in and caring for people. 

So unfortunately, the things we needed to do to stay safe were putting patients at risk, and we really never came up with a strategic way on how to balance those two things.  It mostly had to stay on safety because we were still -- we were still having so many healthcare workers getting ill.  And the ambient conditions in West Africa, as I mentioned, are quite difficult. 

You know, 120 minutes in the suits were probably about what you could do.  We pushed it a little further some days, but -- that would also come with a cost where the next time you go in, you had less reserve than prior because it was so warm.  In addition, the infrastructure, there was a lot of money that was being thrown at the Ebola problem by then, but there was really a poor infrastructure to make sure that just patients could be reliably cared for, such as that the generators worked 24/7. 

So the generators would be turned off just haphazardly.  So we had to make decisions that we just couldn't go in at night because there were trip hazards and there were no lights, and in your PPE, you would potentially rip open your PPE.  So we unfortunately were not able to care for patients overnight on many of the nights, and again, that was -- early on in an outbreak, I understand that's just part of an outbreak, but we were six months in and we were no closer to solutions when we were there. 

Dr. Rubinson:  So most of what we knew about how Ebola was transmitted was from past outbreaks.  Again, you would think six months into a current outbreak with thousands of patients, our knowledge base would've changed.  And Dan Bausch did a great job in earlier outbreaks to try and start looking at what are potential sources of Ebola RNA, and can those sources transmit to other people?  But that's really on a very limited number of samples. 

Now we saw this enormous outbreak, and we were still using the lessons from older outbreaks rather than learning and getting smarter to figure out, how can we keep people safe?  So, again, it just led to a lot of distraction because we weren't getting any smarter six months in to be able to ensure that we could either reduce PPE or this amount of PPE was sufficient.  We just didn't have those answers. 

And then as lessons do happen, that creates more uncertainty.  So in -- early on, Germany had a patient that was sent back to them, and again in Dr. Bausch's series, you did not see RNA collected from the sweat of a patient.  But in this patient we actually saw that Ebola could be found -- at least the RNA could be found -- in the sweat of a patient. 

So that leads you to going, well, what's right?  And while the answers are more complex than that, this really were the discussions that needed to happen because when there is a lot of uncertainty and people's uncertainty is not really validated, then that creates fear, and fear creates a distraction from being able to manage patients.  And that was really what we should've been doing. 

So what I propose is that very, very early on, we deploy a multi-professional team whose goal is to be able to understand transmission.  While I understand there are some politics of needing to be invited in by outside countries, I believe that if we had a soup-to-nuts team that could do all this -- and can do it immediately -- that that would be invited in, and it would reduce a lot of the chaos, where, you know, even now, nearly two years after the outbreak, there's still so many unanswered questions about how Ebola is transmitted. 

If we do the Manhattan Project of sending a group of folks who have a variety of different expertise to be able to figure out some early answers on transmission, we can then get people focusing mostly on the clinical care. 

You'll hear some people say this is not a good use of resources because it's all about clinical care, but if clinical care is being distracted and is inefficient, it's better to get other boots on the ground so that, you know, again, in six weeks or six months, you can have a much more robust clinical response than still living in the world of uncertainty like we did when we got there. 

Also, people who are providing the recommendations probably need to be some of the people who are in harm's way.  And this is, again, no -- it's not meant to sling mud at folks who are creating recommendations, but just like you want someone who's gonna pack a parachute to jump with one of those parachutes, if people are giving personal protective equipment recommendations and saying it's sufficient, but they're not necessarily putting their own people at -- in harm's way, then there's gonna be an inherent mistrust of what those recommendations are. 

So there needs to be a way for some of the boots-on-the-ground people who are truly gonna be in harm's way to be part of the conversation of what is the right PPE for being able to manage people.  Lesson number three:  The public health agencies actually never intentionally implemented or officially said that we were doing crisis standards of care, and I think in large part, it's because I'm not sure that many of the agencies truly understood that catastrophic triage was actually occurring. 

And the dilemma is many of the public health agencies, they were working hard.  These were people who were, you know, at meetings all day, working political angles, doing many, many things to try and bring the outbreak into a more manageable scale and ultimately to go away.  But the same people who were doing all of that were also responsible for the clinical management of patients ultimately. 

And I think the bandwidth starts to get pretty narrow if folks are at meetings all day long about case findings, about a variety of different issues, and also trying to figure out what's the right way to do clinical response.  And I think ultimately one of the lessons should be that there needs to be a group of people in operations who are simply tasked with supporting the clinical delivery of care. 

When I got there, essentially the WHO had tried to recoalesce knowledge into an updated clinical guidance book, and it was OK, but really there was no -- there was no scope of care that was set up.  So be -- when you got there, no one said, “This is what we're doing and this is what we aren't doing.”  It essentially was left to you to show up and make a decision of this is what we're doing, and this is what we aren't doing. 

And it also wasn't built iteratively on the experience of the previous clinicians who were there.  We had some email traffic going back and forth of people who had been there before in conversations, but there was no official jurisdictional sanctioned method to say, “Right now, we are not going past this on patients.”  And if we do go past on that -- if we can't do that currently, this is what resources we're requesting so that we can ultimately provide some meaningful standard of care. 

We tried to have that discussion, and there was just no venue to have that discussion.  In addition, I -- you know, people would talk about the case fatality rate, so at Kenema our case fatality rate was quite low compared to historical average, and so we were getting a fair amount of “attaboys.”  And I could tell you we were not providing any meaningful clinical care at all. 

And I think what that also reflected is the case fatality rate may not be the right data point in Ebola care, and it may just reflect a changing virus, a changing host, or timing to intervention, or in fact just some people actually don't get that sick with Ebola and we were starting to recognize that. 

And I -- I think case fatality rate actually led us astray of just feeling like everything's going OK 'cause we could not provide any sort of meaningful care on a daily basis to people who are sick.  So the reality is we didn't even know who was there.  The patient census were inaccurate.

So as clinicians, some days, that's all we did, and we couldn't provide any care because while we requested people to come and provide assistance with us to gather data, they never came.  Our clinical teams were overwhelmed, and ultimately it came down to there were two docs.  And there were -- of the 30-some-odd nurses from KGH, really only four or five were providing clinical care. 

So we would -- if you figure there's 80 to 100 patients, you essentially get 60 to 90 seconds per patient to provide some sort of care.  So we would do rapid diagnostics, we'd try and intervene, and then we would move on.  And ultimately, we learned that public health was actually capturing data, but they weren't giving it to the clinical team, and all that was, was essentially they were counting cases and the number of people died. 

And there were a number of people who probably could've been brought to bear for data gathering, and again, a lot of people don't see that as useful, but I would argue it's the same thing of when utilities are out in this country.  If we ask the utility companies to bring back on the lights but don't give them any maps of where the utilities are actually out, then they would probably be quite inefficient at bringing those resources back online. 

So data is not about research.  It's about operational guidance to figure out what you're doing so that you can efficiently manage your resources.  And unfortunately, a number of patients probably died just because they were ignored and we just couldn't do anything to care for them. 

Dr. Rubinson:   And I think it's important to look at the difference between deaths versus preventable deaths. 

Now, I understand that a bit of this is in the eye of the beholder, but what we have learned from cholera is if your case fatality rate, which I think in cholera is an appropriate number; in Ebola, I'm not sure.  But in cholera, if your case fatality rate is more than a few percent, it would argue that your cholera treatment unit still does not have its processes honed to be as good as they can. 

And that's a good benchmark of figuring out, does there need to be more educational resources brought to bear?  In trauma, we've learned that if you simply look at number of patients and case fatality rate, and you have a bunch of people with orthopedic injuries after an explosion, no matter what you do to care for them, most of them are gonna live, but it doesn't mean you provided good care. 

So Eric Frykberg some time ago talked about the critical mortality, which was the denominator, only had the patients who actually had a chance of dying, and then you looked at what the mortality in that group was.  We were trying to argue that the same should be done for Ebola.  Whether it's, you know, a person who could no longer take PO by -- and couldn't hydrate themselves, and then looking at the case fatality rate because there again are a number of people who were just gonna do fine no matter what we do for them, and there are a number of people who are gonna die no matter what we do for them.

What we're trying to figure out is the group in the middle.  Can we come up with proper processes, where if we're still at too high of a rate of fatality, then we ask for more resources, ultimately to where we get -- to where we're really a good Ebola treatment unit?  And that experience was really -- at least in Sierra Leone when I was there -- was never established.  We had no idea what was -- what was the right ability of an Ebola treatment unit to provide care. 

So unfortunately, we had to make a lot of triage decisions without any operational top cover, and you made them on the fly.  And the crisis standards of care model people talk about, that typically you do this as a team.  We tried to do it as a group, but it really had to be the clinicians and the team because we were so overwhelmed, you didn't get a second chance.  You can't have someone coming by three hours from now because probably the decision was already made on them. 

So the pace is much faster than I think has been reflected in the Institute of Medicine document, but that just also may be that nature of the level of mismatch that was happening in West Africa.  And we struggled with a lot of difficult decisions, and we talked about this at night.  “Should we be trying to get people out of the Ebola treatment unit who are inappropriately put in there because they had symptoms consistent with but didn't have Ebola, or should we concentrate on treating patients?” 

Because both take resources.  And it was -- it was a struggle to try to figure out what the right balance was.  We didn't want anyone to get Ebola simply because they were mis-triaged into an Ebola treatment unit.  At the same time, we also didn't want to have to keep walking past a bunch of patients who were really struggling and potentially had survivable disease but were moving on to dying simply because we were, you know, trying to get the other patients out. 

And I tell you, we've really never figured out the right balance, but we -- those conversations we had quite often.     And more importantly, we -- we needed public health agency guidance on that as well as additional resources being requested to know that we were making those difficult decisions.  And when we did send out information -- and again this is not meant as an individual hit -- essentially, we were told can we change the format of our request into tabular form because we didn't have it in the proper format. 

And then they went unanswered from there.  So obviously we were in a situation of crisis standards of care where there wasn't enough time to perform all the essential tasks.  And what we really needed is, we needed additional people helping us figure that out rather than us coming up with it on the fly. 

And most importantly, we needed data so that, you know, our instincts could be reinforced, meaning if you thought someone was too sick and you were gonna pass on them and move on to the next patient, you wanted to be pretty sure that you were right.  And we had very limited data at that point in time to make those decisions.  So we were making big decisions really on the fly without any information. 

Dr. Rubinson:  So, lesson number four: the public health response was more focused on disease control, political interface, and building ETUs, but the processes within the ETUs, at least when I was in Sierra Leone, largely went unnoticed and misunderstood. 

And to expect that the frontline clinicians are gonna build that and also care for patients is really a setup for failure, and especially when there's human-to-human transmissible disease as well where there's, you know, the whole issues around the emotional struggle of trying to stay safe.  So we needed a much bigger bench, but we didn't necessarily need a whole bunch of additional clinicians. 

We needed a bunch of different skills to be able to help us, whether it was data gatherers, whether there were industrial hygienists, etc., and unfortunately all those requests went unheeded.  So how does an ETU work?  Well, the first is the suspect ward, and that's where people who have symptoms consistent with the case definition are put in, but there's no confirmation that they actually have Ebola, and they're put in a unit where there are people with Ebola who haven't been confirmed yet either.

Dr. Rubinson: So you can imagine you want to separate those people, and most of the patients who are still well enough have the survival sensibility to separate themselves from other people.  But the key thing is you want to get the patients with Ebola into the confirmed ward 'cause that's the best way to concentrate your resources to provide care.  And you want those without Ebola out as soon as possible so that they don't get ill from being mis-properly triaged. 

And then there's a whole period of Ebola as people are convalescing where you want them out of the treatment unit as well so that they're not -- they're not taking up a lot of the resources for treatment.  Or you can repurpose those people to help provide treatment for when the caregivers are not able to be in the treatment unit.  The annex is where we did triage.  It was a dungy, dreary place. 

It was a place that had an enormous amount of trip hazards, and what we had to learn is to walk past the people in need.  So when you first get in, the last thing you want to do is take care of someone who likely has Ebola, is very sick, and touch them and care for them and then go and take care of people who don't have Ebola.  So we had to be able to come up with a pathway to make sure we don't transmit to people who don't have Ebola and then go back and deal with the folks who do have Ebola. 

We had to figure out the census and update the duration of symptoms because unfortunately, the PCR tests could miss Ebola in the first three days of symptoms.  So we needed to truly understand when someone's symptoms started so that once we had that duration, as soon as they're more than three days, we could get 'em out of the unit.  But then we had to prioritize their phlebotomy. 

So the goal of the suspect ward is get the people who don't need to be there out, get the people who are clearly sick into the treatment ward, and try and avoid cross-contamination.  And one of the key features of whether or not we could reduce the burden in the annex would be could we have changed where the previous studies had told us that the PCR is only fully sensitive after three days. 

But now we had thousands of patients to learn from, and we don't know if that situation were still true.  Even if we had knocked that down to 48 hours, that could've reduced the burden in the Ebola treatment unit again, but there was nowhere to have this conversation to try and prioritize, what are the key questions we need to solve, again, that aren't necessarily about getting docs and nurses to West Africa, but can we use some of our data to change our processes to reduce the burden so that the docs and nurses who are there can be more efficient at providing their care? 

So after the annex, we would -- or in the annex once we dealt with the folks who were not that ill, we would take care of those who were quite ill.  And we realized after a time, it was just not the right place to try and manage care.  So we would again try and prioritize patients to get phlebotomized, to find out if they do have Ebola, so we can send them into the treatment ward. 

And one of the first things I noticed at the treatment ward was there are a whole bunch of people holding court on the porch outside of the treatment ward.  And some of those people were on their way to getting sick, but a number of them never got that sick.  They just add muscle aches.  They had some fever, but they never got seriously ill.  So those folks became our extenders and our community folks who would watch to make sure other people were doing OK. 

And then we would also get fairly good at being able to find who was really sick.  One of the telltale signs was when it was 90 some odd degrees and humid and someone's under three blankets in the Ebola treatment unit, that person's probably fairly sick.  So we got better and better at concentrating those patients in a particular place so we could be more efficient on rounds to try and give them IV fluid, antibiotics, and antimalarials. 

Another thing we learned, unlike on TV, Ebola is not -- the predominant feature is not hemorrhagic disease.  It's a GI disease, so they have profound nausea, vomiting, followed by diarrhea.  It's a neurological disease, and actually the agitation is a huge problem because trying to get people safely managed with IV fluid when they're agitated and they want to keep walking around or they want to hide under a bed was a huge struggle for us to be able to provide appropriate care. 

And occasionally they have bleeding.  So the clinical features are key to understand because that's the requirement for being able to provide care.  And again, they're -- most of the features of the disease were, I think, slightly different than the reported literature.  The CNS agitation was profound.  There was some early information from Conakry and Gueckedou in Guinea where some of those clinicians had at least by email taught us some of their features, and we were trying to provide care. 

But bleeding really just became evidence of someone who probably was gonna die, but it wasn't massive bleeding.  So if you saw a nosebleed; if you saw gum bleeding, those people were probably gonna be dead in the next day or two days.  We would still provide them some care, but we knew that as they got sicker, we would have to move on to other patients.  So what was proven about the care prior to 2014?  Almost -- almost nothing, unfortunately.  

There was a fair amount of dogma about whether or not we should be doing oral rehydration.  A lot of that came from people who treat cholera.  The nausea and vomiting is quite profound in Ebola, and the amount of oral rehydration you can do is quite limited in the disease.  And we're not really sure how much IV hydration to provide.  Nutrition is something that a lot of people have argued for.

But the critical care literature from the last three decades would tell us that for a limited disease, nutri -- meaning a time-limited disease -- nutrition doesn't usually play that big of a role.  So honestly, there was a lot of people who felt very strongly about what care should be provided, but we weren't really sure what it should be.  And when you're in a resource-limited setting, you want to just be providing what you think is really important.  And unfortunately, we had no data to guide us.  So we just again unfortunately made it up, and making it up feels pretty awful six -- six months in. 

And this slide just goes over the variety of things that we did for the patients,

Dr. Rubinson:   and this is ultimately as the international community was gearing up to provide help, this was the difference.  So this was the level of complexity of our census. 

Dr. Rubinson:  That census took me 40 minutes to do just writing that down and asking people.  And most of the international community was focusing on novel therapeutics. 

All important stuff, but we couldn't even provide the most basic of care, and that's really what, you know, the world could've provided early on rather than trying to figure out how do we bring in new therapeutics to treat this disease.  Can we provide even the most basic supportive care?  Then if we do that, then perhaps maybe we can provide additional therapeutics. 

And when we were overwhelmed, again, we kind of developed our own intuitive senses of who would do badly, but it wasn't until after we were already there that we start to see reports coming out.  So six months in, there was still very little information on who did badly. 

Dr. Rubinson:  We started to get the sense that if people were agitated or moving towards coma that they would do badly.

It doesn't take a rocket scientist to figure that out, but ultimately, data proved to that because when you're walking past someone who you know is gonna die with no care, you want to be pretty sure that you're right about making that decision.  And again unfortunately we were just making it up as we went on.  And our experience again from early bleeding, poor-tending outcomes ultimately did bear out in other cohorts, but this information should've come out much earlier. 

It's great that folks who put this information out were able to do this, but there really needed to be a concerted effort to get this data out really early.  Not just on the clinicians, but the world should've brought resources to bear to be able to do that, so that we're not leaving clinicians hanging and patients at risk by folks just making chaotic decisions in the midst of chaos. 

So the clinical response is not just about bedside providers.  Disease transmission assessment needs to be a priority.  We need to have a systematic and strategic clinical response.  I understand that the initial responses can be chaotic.  Of course it's gonna be, but there needs to be a strategy, short term, midterm, and long term, so that we iteratively get smarter as we continue in the response. 

We need appropriate metrics that are rigorous and that truly reflect the processes of care that we're delivering, and some of those metrics need to be created by seasoned clinicians so that we're -- again, we're actually measuring what's truly reflective of process rather than just what we think matters.  Patient data collection has to happen.  And this is not about research.  This is about operational guidance. 

We cannot be responding to these outbreaks without any information.  When we're making this up on the fly, we are inefficient, and we are misusing our resources.  And there are -- what I've found during multiple responses is there are many people who want to help, and the resource bench is incredibly deep.  Just, people want to know how they can help, and we need to get smarter about how to leverage all of the world's resources. 

Even if not everyone wants to travel to West Africa to assist, you can imagine we could've used a number of people who stayed in their own country, but who were extenders who were either providing clinical guidance or helping figure out what the equipment supply should look like, and then ultimately having the end users figuring out if that was right or wrong rather than asking the people who were managing patients day to day to figure it out on the fly. 

And it actually matters.  This was not an abstract effort.  Some of the crisis standards of care stuff sometimes feels like it's a -- it's when the asteroid is gonna hit the -- the Earth, but this stuff actually matters.  Getting it right matters.  It changes people's lives, and it's important that we do better next time.  Thank you so much.

Dr. John Hick:  That concludes Lewis's formal presentation.  And now we have an opportunity to ask a few less formal questions about some of the material.  Lewis, thanks so much for taking the time to be with us today.  We understand how busy your schedule is, but fascinating conversation.  And I wonder just for starters, you know, you were dealing with a country that has very limited governance, very limited public health structure, and a lot of mistrust in fact of public health and of modern medicine.  Do you think that we would see similar levels of fear and distrust and issues here in the United States? 

Or is that just a phenomenon that we have to understand that we would be dealing with in less developed countries?

Dr. Rubinson:  I -- I actually was amazed at the resilience of the Sierra Leoneans.  So their country was on fire with Ebola, and life continued to go on.  And in many ways, they were much more resilient than the fear which kind of took over the U.S. during, you know, really a very small setting of secondary transmission. 

So albeit they have limited resources, and a number of the folks in that country don't follow -- don't prescribe to the germ theory, they struggled to be able to bring enough resources to bear, but at the same time, they did not let fear take down their entire community.  I think we can learn an enormous amount of lessons from that in terms of community coming -- coming together to be able to solve their problems. 

I think the risk communication, though, on Ebola failed whether it's resource limited or resource rich environments.  There was a fair amount of dogma about Ebola that I don't think we can be as sure what we know about Ebola from rural Africa and extrapolate that to resource rich environments and be as sure as we were 'cause then when that proved not to be the case, that led to a lot of mistrust. 

And I think Texas is probably the biggest area where that played out.  So again, I actually think that in West Africa, they were able to deal with the outbreak much better than we did with our very small outbreak in the U.S.

John Hick:  Interesting.  I think the point that you made about the -- the need to recognize uncertainty is critical you know because going into any new epidemic, we still don't know a lot of things that we want to know.  And if you're not upfront about that, whenever you're proved wrong, that's a huge trust issue for everyone. 

So the scope of clinical care that you were able to offer in Kenema, were -- did you have any basic laboratories that you could do blood or urine or anything?  Or were you mainly just dealing with visual assessment of patients and/or minimal vital signs?  What -- what did -- what tools did you have? 

Dr. Rubinson:  That's a great question, John.  So in Guinea, several months prior, they were able to utilize a point of care device, an i-STAT.  When we were in Kenema, there was no laboratories available at all except for molecular testing for Ebola and antigen testing for malaria was sometimes available.  That was the only two diagnostics that you could get. 

Everything else, so in terms of physical exam, looking at conjunctiva to try and figure out what the hemoglobin was, for trying to determine heart rate, we used a pulse oximeter not because we could provide any supplemental oxygen, but we couldn't bring our own wristwatch in to be able to count time so it was able to give us a pulse rate.  And we used a manual blood pire -- pressure cuff and then also on the most part didn't even use a stethoscope with it. 

We just used palpation to figure out what the blood pressure was.  And that was the extent of your level of diagnostics.

John Hick:   And I think the point you made about if you've got a person who is suffering from agitation and also dehydration and you're not sure what the risk benefit of trying to put an IV into somebody who's agitated without sedatives being available, you know, you can assess tactilely for fever and sweating and things like that, but you can't do any tactile analysis what their sodium is or, you know, how much that person needs IV fluids versus another person.  So that -- it makes it incredibly difficult to, you know, decide who is going to get what interventions, even if they're available. 

So was there any -- was there any thought about trying to predict, you know, sort of resource requirements or -- or assessing requirements for, you know, like for every 10 patients, you know, this is what we'd like to have as far as ORS or diagnostics or whatever?  And, you know, is that kind of requirement building?  Is that something that we need to be looking forward to as the future to say if we're going to level X level of care or Y level of care or Z level of care, then we have to be able to provide, you know, for X number of patients these resources, staffing and otherwise?

Dr. Rubinson:  Yeah, I think that's the exact right way of looking at it.  We -- we -- by the time we had left, I think we had a fair sense of what we thought we needed, but honestly, you know, we made it up just like everyone else, and it was not evidence derived.  I think just like I had mentioned before that there should be an early deployment team that's looking at early transmission.  Probably one of the sites should get concentration for resource utilization. 

You can't do it in every clinical site where you're trying to provide clinical care, but you can probably do it at your best resource site.  And again, it's not to do research on people.  It's to get an early understanding so then you can start leaning forward and being able to bring the rest of the treatment units up to the same level of capability.  But you're absolutely right, like,  we have no idea how much IV fluid it actually takes to resuscitate someone. 

And if you give too much, is that bad?  If you give too little, is that bad?  We have a little bit of information from the people who came back to Europe and the U.S., but that's less than 20 patients.  You know, there were 30,000 opportunities in West Africa that we really missed the boat on trying to understand how to manage them.  In terms of your question about agitation, even simple stuff. 

We -- we were given IVs that were metal cannulas that -- yeah, so 40- or 50-year-old technology, very dangerous to use, and there was no way to secure them.  So if the patient got agitated, they would pull 'em out and there'd be blood everywhere.  So much of this stuff was not the more difficult doing really, you know, resource intense data derivation.  It was just someone listening of hearing. 

You know, if we had some protect catheters and we had some devices to secure the IVs, and we're talking about pennies and dollars, right, rather than spending millions and tens of millions of dollars on clinical research, even though ultimately I believe that's important, we would've developed an infrastructure far more capable than spending almost all of our time on esoteric therapeutics when we couldn't even do the most basic stuff.

 Dr. John Hick:  Sure, and you have to be able to ride the bike first, right, you know before you start racing.  I mean, it was never gonna be that everyone could get ZMapp or get, you know, advanced therapies at the same time.  We didn't even know whether the basics made a difference, and if you took a select number of those treatment centers and really invested some time, we could've had some invaluable data. 

Dr. Rubinson:  I think that's right.  I mean, the patients who came back to Europe and the U.S., while there's a selection bias to that, the mortality is under 20 percent for that group.  And it's not clear whether novel therapeutics made any difference, but they were done in -- they were administered in such a way we'll never know.  But it looks like it's probably just good, supportive care. 

And if we could have knocked down the mortality 30, 40 percent by good supportive care, it's hard to imagine that a novel therapeutic would have that big of a treatment effect.  That's not to say novel therapeutics don't have a role.  They just -- they're -- they come further on into the outbreak.  We always know what the ethical issues and all of the other issues that go together with bringing novel drugs into other countries that it's gonna take some time. 

And during that time, people are continuing to get sick.  So had we had a multipronged approach of let's get good supportive care; let's figure out, like you say, what are the resources, look at that in a few sites, then emulate that over multiple sites, then have other people working the novel therapeutics.  By the time the novel therapeutics were really ready to go, you would have an infrastructure ready to give them.

Dr. John Hick:  And I think, you know, you made a good point about we did get good data from some of the patients who were brought back to western environments, which was helpful because some of the assumptions that were being made about care in a first-world environment and -- well, we're not going to offer dialysis.  We're not gonna intubate these people, which I thought was particularly, you know, egregious example of taking historical data that was based on care in a very rudimentary healthcare system and try to extrapolate to, you know, a well-resourced Western environment. 

We have to be very careful about those assumptions.  So -- the thing that you mentioned with the -- the personal protective equipment that obviously had some personal issues for you is that there's always a tendency to say more PPE, better.  And at the same time, more PPE equals more physical limitations, you know, more physical impediments.  Any more comments that you have about, you know, contributing factors to, you know, falls, needle sticks, those sorts of things that the PPE winds up engendering?

Dr. Rubinson:  Great question again.  I -- I think taking a step back, the PPE has become the major focus, and I think -- excuse me.  What I'd really like to see, and maybe you can't do it as much in the resource-limited environments, is we need to get much smarter about our environmental controls, and PPE should only be a last fail-safe.  If you look at the nuclear industry -- if you look at most places, they do not put the risk of whether you get ill or not simply based on the PPE. 

The PPE is the last fail-safe just in case everything else fails.  We've turned that on its head in how we do outbreaks here where it requires a human to be 100 percent adherent all the time.  And the burden of the PPE is fairly high.  I'm not trying to say that PPE will go away, but I think, you know, what we've done in some ways is made it nearly impossible to provide clinical care. 

Most of -- like you talked about, the burden of PPE gets in the way of what you're trying to do with a patient.  And I think the discussions of PPE should be much broader than just with clinicians and infection control people.  I think it should be human factors, experts.  I think it should be industrial hygienists and other folks who can get us thinking different ways of how to keep ourselves safe.  And like you say, sometimes more is not better, absolutely. 

So we, we had to trans -- transition from facemasks to goggles 'cause there was a perception that mucus membranes were at risk with facemasks, which is true to an extent, but the goggles that we were given fogged up almost immediately.  And because of the trip hazards, we were at much higher risk of ripping open our PPE 'cause we couldn't see with the goggles. So it was well-intentioned, you know, recommendation that forced us to do that, that put us at much higher risk. 

So I think sometimes more needs to be put into place of the context, the situational context, and weighing all the different factors on what ultimately keeps someone safe.

Dr. John Hick:    That hierarchy of controls in infection control is a lot easier to implement when you're in a nice air conditioned hospital, but at the same time, you're exactly right.  PPE is always the last barrier, you know, not the first.  While you were there, you actually wound up suffering a needle stick and -- and were hospitalized for a few days, you know, during an observation period, which was -- has been described in journal articles.  From the perception of being a patient who, you know, had to deal with caregivers who were in PPE as well as a caregiver that's been in PPE, can you speak to just kind of the emotional issues and some of the things that we need to be cognizant of from the patients' perspective?

Dr. Rubinson:   Sure.  So at -- at -- at first when -- you know, when I got there, so I -- I, you know, essentially had a reaction to the VSV vaccine, which did what it was supposed to.  It -- you know, it essentially generated an immune response from me, and it made me feel pretty un -- unwell for a little while.  When I was feeling that way, I honestly -- I didn't care about the PPE.  You know, the staff were great.  They were wonderful. 

I went -- there was a small risk I had Ebola, very small, so of course you would never want someone else to get sick because you're sick.  So I think from that standpoint, all was fine.  When it became clear that I did not have Ebola and I had a vaccine reaction, I had to spend probably another five or six more days in the biocontainment unit. 

And that's where really it felt like it was starting to get at where I was being held against what really science would say.  And again nothing against the folks at the NIH -- they were wonderful -- but ultimately, many of us saw many bad things in Africa, and we were dealing with the demons of the things that we couldn't -- that we knew we could manage in an everyday basis, but we couldn't provide that kind of care there. 

And then to essentially kind of hold us as someone who is putting the population at risk, it obviously engendered mixed feelings.  From -- I -- I felt a lot of respect and -- and really a lot of gratitude to the folks caring for me, but it was clear that I wasn't a risk after a while, and my desire to get out was, let's just say, not heard very well. 

And -- and honestly, not by the NIH but later by public health folks, I was threatened with things such as if you walk your dog, you will potentially be incarcerated.  And just after the experience, I get that the country was fearful, but if we're saying that you can spread this disease asymptomatically, if I'm adhering to all those things, I had probably been tested for Ebola more than anyone at that time--

Dr. John Hick:   You were probably the safest person to be out walking your dog at that point, so...

Dr. Rubinson:   Exactly.  And just how to balance that with also -- I get that they have the public perception of needing to have people feel safe.  I chose to take a different tact than the person in New Jersey, but I got what she was experiencing, you know. 

It's -- and -- and I -- and I think we're gonna -- we're gonna create a fair amount of potential for PTSD in folks coming back, when they come back and really need to start just processing what they're going through, and instead they're going through legal issues that somehow the work they did is -- is putting the public at risk and they should be ashamed for it.

Dr. John Hick:  And -- and isolated, you know, which is the last thing you need when you're trying to kind of reconnect the dots and reenter, you know, into our environment.  So difficult, but at the same time, we're grateful.  You know, thank you for going, and thank you for the information you brought back.

Dr. Rubinson:  Sure.

Dr. John Hick:  So we appreciate Lewis taking time to share his thoughts and ideas about his time in Sierra Leone with us today, and thank you for joining us.

[MUSIC]

Narrator:   Thank you for joining this ASPR TRACIE program on Lessons Learned on Crisis Standards of Care in Kenema.
We encourage you to visit us at asprtracie.hhs.gov for more information on a variety of healthcare emergency preparedness and response topics, including a Topic Collection on viral hemorrhagic fever and Ebola, and a separate Collection and Information Exchange thread on Crisis Standards of Care. Issue 1 of our newsletter, The Exchange, focused entirely on Crisis Standards of Care. You can also submit resources to be considered for inclusion in our resource library and Topic Collections, participate in a password-protected discussion board, or submit requests for technical assistance.

[The video ends.]