SCCM Resource Library-Rapid Response Teams: 15 Years Later Post-COVID!

Video Transcription

Hello, my name is Dr. John Whitcomb, and I am the moderator for this session titled Rapid Response Teams 15 Years Later Post-COVID.  Topics covered in this session will be team leadership, how has it evolved, and response activities impact on patient destinations. Thank you and enjoy the presentations.  Good morning, everyone. Thank you for joining us. I am here, and I'm going to talk a little bit about rapid response and team leadership  and how it's evolved over the last few years. My name is Fiona Winterbottom. I am a clinical nurse specialist at Ashna Health in New Orleans, Louisiana, and I have  led implementation of our rapid response team over the last five years. I have no disclosures. And I'm going to start by just talking a little bit about how we set up our program.  We really looked at all of the things that contributed to early mortality and preventable mortality in our hospital, and so we looked at what were the drivers of those things,  and how could we mitigate those. So proactively preventing inpatient failure to rescue on the floors, enhancing early recognition, and then leveraging technology for surveillance.  So a lot of work on our electronic medical record. And then making sure that we had experts to resuscitate patients, so that's where we did  a lot of focusing on staff training, simulation, and then also our rapid response team. We wanted to make sure that we also were focusing on non-beneficial care, if that was  appropriate for patients. So we worked with our palliative care teams to establish training for that.  We looked at how we could optimize our care, and how we could be more timely, and how we could give feedback to staff quickly.  And then we also looked at how we could engage our staff further, and engage patients and families in their care.  So you may have seen this before. This is kind of the deterioration slope, and when we first started our program, we had  a heavy focus on advanced life support, and how we could really prevent cardiac arrests. And gradually, as our program has grown and matured, we have kind of worked backwards  up the slope, and so we've moved towards rapid response, and then really the clinical review and prevention of deterioration altogether.  And so this slide is really, it's taken from the Between the Flags program in Australia, and it really looks at the different modes of governance, calling criteria, emergency  response, education, and evaluation. And I'll explain how we kind of set our program up in a similar way.  We used a similar structure with our local, thinking about our local resources. So we really developed task forces underneath our program restructure.  We had a lot of work happening on technology, artificial intelligence, things like that. We looked at process of care is really our team, our rapid response team.  Equipment and supplies, we were constantly looking for supplies, and who owned them, and who paid for them. And so we did a lot of work on that.  We actually introduced a new defibrillator fleet, because all of our equipment was end of life. And so we introduced that.  Regulations, obviously, we want to make sure that we're following all our regulatory guidelines and we participate in the Get With The Guidelines program.  Education, obviously, nothing can be done without education. So we had a big training program that launched with this.  And then data and outcomes, we wanted to make sure that everything was very targeted within our program. And so we had expected outcomes.  But this discussion is really going to be about our process of care and our rapid response team.  So a rapid response team is defined as a whole system that provides a safety net for patients who suddenly become critically ill and have a mismatch of needs and resources.  So we had had a rapid response team for many years. We started in 2006. But it was really a consult team that was an ICU nurse.  That was called to help staff on the floor. And what we found was that that was effective for a while.  But as the hospital grew, as the complexity of patients grew, we really needed a different model. So in 2016, we did an evaluation.  And in 2017, we actually launched a dedicated 24-7 rapid response team.  So just kind of a summary of what we did over the last three years, we introduced a training  program, a fairly significant training program, not just for the rapid response team, but for all of the staff in the hospital.  The nurses, the rapid response nurses were trained to be experts in that program and actually taught that program.  In 2018, we launched the program and we really got our medical record working for us. So we were able to do risk stratification.  We had proactive rounding by our rapid response nurses. And of course, they responded as they had always had to our codes and other emergencies.  We did deploy a large defibrillator fleet, which was we have on one of our biggest academic campus, we have over 100 defibrillators.  So we did have to replace all of those in a staged manner over about a year. We use those to also train the staff in goal-directed resuscitation.  So we looked at prevention of clinical deterioration, and we looked at the actual code and use of the defibrillators.  By mid-2019, we realized that we needed a second nurse, and I'll show you that when I show you some of our data, because we had so many calls.  We added a respiratory therapist in 2019 also, because we found out that most of our deteriorations were due to respiratory events.  And then mid-2019 around June, we started to do a fairly rigorous case review process. So we would go and look at all of our cardiac arrests and work backwards to see what issues  we could mitigate and then put things in place to overcome those. And then in 2020, as we, you know, as COVID hit us, we were hit early in March of 2020, right after Mardi Gras.  Then we added a third rapid response nurse for those COVID units and those high acuity respiratory units. This is just an example of our risk stratification system.  So you can see basically the red and the yellow indicates whether the patient is a high, the one on the MUSE column. And so that's what our nurses use for the high-risk screening.  We have other tools that we're working on, MSO for some sepsis prediction models, different things. We also have artificial intelligence alerts.  And so our nurses also get those to their phones or watch. And this is kind of an example of that.  We have been using continuous monitoring, which actually helps with triggering those  alerts because it means that there's a constant supply of data that the backend kind of intelligence can look at.  And so that has actually helped us be more targeted in our responses. So we have the monitoring system. We have also our routine vital signs that we've always had.  Our alerts can go to our phones, they can go to our watches, we can get notifications in multiple ways. This is kind of an example of some of the output that we have.  And this is really just designed to give you an idea. The black screens are really what comes from our continuous monitoring devices.  And then the white screens are the ones that really give us aggregate data so we can see  how many people are placed on continuous monitoring each day and then over time and what the utilization is.  So that gives us kind of a patient view and then it also gives us an aggregate view. When we look at what happens after we screen patients and then after we go to see patients  because they've screened out or because a nurse has called and consulted the team, we have some written order guidelines that the nurses have been trained to use.  So if they have a neurological event, these are the things that the nurses can order. If it's a respiratory event, these are the things the nurses can order.  So a lot of it's testing labs, ABGs, chest x-rays, things like that. So that's been really helpful and was particularly helpful in COVID, especially with patients  that have respiratory problems or were in distress and near the end of life. So that was a change we actually made. This is just an example of the notes that the nurses write.  And so it's very flagged in the medical record. If anybody's pulling it up, they can see it's labeled a rapid response or code note.  And then it's also got all the details of all the analytics really we would want to pull. So the nurses fill these out.  A lot of it is dot phrases and things that pull in from the records. So they don't really have to do a lot of typing. But it does help us track the interventions of the nurses.  And then this is all automated on the back end to be able to pull reports. And I'll show you the output from some of those.  This is just, again, not designed really for you to read the details, but just to give you an idea of what our case review process is. Every Tuesday at 4 o'clock we meet.  Everybody knows that. If they have a case they want us to review, then the resuscitation committee will review it. We can look at the top bar is really just our vital signs.  The black is our automated and continuous vital sign flows. This document right here with the blue lines that go across it is our CPR feedback.  So we can look and see how well we executed our CPR, our date and speed of compressions, our rate of compressions.  And then did we shock appropriately and use entitled CO2, et cetera. So did we use our goal-directed therapy?  And we found that we do very effectively and our time on the chest and time to the chest is faster than it was certainly a year ago. So hopefully that means our training is working.  And then what we do is we give our codes a preventability rating and a classification. So that way we can kind of aggregate problems and then we can work on them to mitigate them  later. So now I'm going to share some of our outcomes with you so that you can kind of see from that work what we have, what we can demonstrate.  This is our cardiopulmonary arrest rate from 2017 to 2020. It was actually higher in 2016 and 2015. So this is going down.  We did do some, we introduced our rapid response team right here and it was December of 2017.  So really the data shows that we, this bottom line is the non-critical care codes and the top line is the critical care codes.  So you can see at the beginning, our non-critical care codes were higher than our critical care codes. That was in January of 2017.  And so we've pretty much been able to keep our codes outside the ICU below our codes inside the ICU.  Right here you can see in March of 2020, we had a little blip and that was simply due to COVID. I have no other good explanation for that.  Our floor patients hit us earlier before our critical care patients. So you can see that that probably explains both of these peaks.  You can see where we added kind of people along the way. This is our proactive screening. So I try and put things in buckets of screening and then proactive and reactive work.  So our screening is our chart reviews, our consults, which is when the nurse calls. Some people call that the nurse worried and our artificial intelligence alert.  So a lot of times the nurses know something's wrong, they're just not sure, and they really want a second set of eyes.  So we've always kind of marketed our rapid response program as nurses helping nurses. Although I think it helps a lot of other people too, actually.  You can see that as we added more people, more nurses and respiratory, we obviously increased it.  It's kind of like if you build it, they will come while they came and they asked for help. So the nurses were good at going to the units, teaching, really trying to keep people out  of the ICU. This was COVID. So you can see we had a peak there and as everybody knows, we're going back up again here in the fall. So we expect it to peak up again.  We see about, we screen about 2000 patients a month on average, which is a lot. We have two rapid response nurses on each shift, ideally. We can usually staff that.  And then we have one respiratory therapist, 7P, 7A to 7P.  See the COVID spike there. This is, as I told you before, we kind of track our notes. That's how we get our measures.  It's an automated report where we can pull how many notes were written. So this is actually the notes that were documented when the nurses went to see patients.  So we use our proactive notes and our follow-up notes. Each note has a category and an operational definition. And so we use those to capture the work of the rapid response team.  And that's about 250 when you add those together each month. When you look at our proactive rounds and our RN consults, obviously we have limited resources as does everybody else.  And as the consults go up, which is more our, you know, kind of reactive, then our proactive rounds tend to go down.  And so this is just really showing you that, obviously, you can only do so much work. And so we become more reactive when, you know, when we're pushed for the consults.  And again, you can see those consults went up when COVID hit us early on in March.  Here's our, I call it nurse facilitated upgrades. Because we do have patients that move to the ICU that are not rapid, not facilitated by the rapid response nurses.  Some services have primary services that can move their patients and have admitting privileges to the ICU.  But for those others, we usually, it's about 80% of patients, probably even more than that, above 80% of patients that are rapid response nurse facilitated upgrades.  This is really showing you both the upgrades from, to the ICU, and then the upgrades that are to a higher level of care, which would be from our med-surg units to our step-down  units.  Usually we upgrade about 100 patients a month. That's about where we live. That's actually less than when we started.  This gives you another reactive response, our cardiopulmonary arrests. You'll see the blue is code blue, and then the floor intubations are the orange.  I'm really, it's, I think through COVID, we saved a lot of patients by intubating them on the floor. We definitely saw a rise there.  And then here is a cumulative look at all of the interventions that our rapid response teams do. And so, you'll, you know, you'll see from some of the other speakers,  they don't just do rapid response and code blue, they do IR coverage for strokes, they do rapids, they do STEMIs, they do multiple other things  that will come in that bucket of work. And so we wanted to make sure that we captured that.  Here's our COVID data, and you can see that our intubations went up 220%. During that time, our consults went up 375%.  This is just an example of an automated dashboard that shows our work. It's a work in progress, but I thought it'd be interesting to share what our risk adjusted mortality is.  We're trying to be a little bit more sophisticated with our program to show how we can look at lives saved based on what the staff does.  And so you can see that this is a little dated, but you can see that we did have some lives saved and then we set goals to see how much we can actually move the needle on those.  This is some data that we just finished that looked at the statistical significance of our program. And you can see the line on all of them is where we made an intervention.  And we basically statistically, it was statistically significant that we improved in all domains.  We did some financial analysis that year one, and we basically paid for our program.  And this is where we just also made a community model in this last year during COVID. And you'll see that it worked for them just like it worked for us at our academic center.  So we can use it in both settings. I think it just shows that we can use rapid response in a pandemic and it helps support the nurses  and it helps support the team and allows flexibility.  And here are some of our team members and our references.  Thank you.  Hello, my name is Michael DeVita. I am a critical care physician. And I want to provide for you today some context for rapid response teams in the COVID era.  I will do that by reviewing a little bit of the history and discussing how and why the term RRT is a bit of a misnomer. I'm gonna be highlighting instead the rapid response system.  And then at the end, I will circle back to how it is impacted by COVID.  The first thing that's important to recognize is critical illness is common. It occurs outside the academic setting. It occurs outside the ICU. I think there can be a misconception  that critical care only occurs in the ICU. And that is false. Patients can become critically ill any place. And when that happens, they need to be treated quickly.  You can see a graphic on the left by Dr. Kwok many years ago, that this is one of many studies that shows that failure to respond results in long-term increase in mortality.  The criteria listed on the right is something that we developed at the University of Pittsburgh. And the point of this is not the content of the slide,  but that there is content relating to when to trigger a condition C, which is what we called our rapid response team.  Having objective criteria makes the call about the patient instead of about the caregivers caring for the patient. What we wanna do is find situations  where the resources needed by the patients are out of proportion, they're not adequate. So it is a system issue. We need to make sure that needs and resources are matched.  If they are matched, that's a happy situation. If there's more resources than the patients actually needed, well, that's okay as far as the patient's concerned,  but in terms of us being efficient caregivers, it's not as good as you would like. And of course, if the needs are in excess of what is available to the patient,  that is a bad situation. And if the patient is critically ill, it is extremely dangerous. When that occurs, you have two choices.  You either have to move the patients to the resources or move the resources to the patients. And I'm sure everybody listening in today would agree that it is much easier  to move critical care resources to patients than it is to affect a transfer of a patient from a ward bed into the ICU.  I'd like to talk a little bit about the rapid response system structure. When we held our first conference on rapid response back in the early 2000s,  most of us came to the meeting prepared to describe, discuss what eventually became termed the afferent limb. We had all created a team that could jump in  and fix any problem that happened or at least get the person transferred to the ICU. As we discussed it more, we realized that we had all created a list of triggers  and the triggers were somewhat similar, but they were different. And the key point was we had a trigger that told people, this is an event that needs a special team  and then we'll send the special team. And so this came to be known as the afferent and the efferent limb.  In the last day of our conferencing, we recognized that there were two other limbs  that we were all doing that we also had not previously recognized as being important. The first is a quality improvement arm and this is in the bottom left-hand corner.  And this is collecting data around the event that then leads to future improvements and then an administrative limb which oversees all these moving pieces  to make sure that they work well together and that they are in a position to improve. I will talk more about each of these components later.  Gary Smith described the different components of the rapid response system in a chain of events that has to occur at the patient's bedside. He called this the chain of prevention.  And it starts with education of the staff and it moves towards eventually an effective response. And I wanna talk about how each of these fits  into the structure that I previously described, the four limbs. The education is part of the administrative limb. This is people who are designing the response  and then are making sure that everybody knows what to do when a problem occurs. The efferent limb is the responders and this is teaching people who should go  and what they should do when they get there. In between is the afferent limb and this spans everything from monitoring through recognition that there's an abnormal situation  and then how to call for help.  The last limb is a quality limb and this moves as we collect information on the backend of a response, then feeding it back into our educational system  and our policy system in order to create a newly more safe environment.  So the administrative limb is responsible for creating the strategy, focusing on what behaviors we wanna see and then education for key constituencies.  And I'll talk about that in the next slide. There needs to be reinforcement after events to help people recognize what I did well and what I did poorly.  And mnemonics are very helpful. Most hospitals I walk into now will have some posters someplace regarding what are the rapid response criteria and how to call.  And I wanna help everybody recognize that it takes a long time to change culture. It takes at least one generation. So if your median duration of people working at the hospital  is three years, you should expect the educational process to take at least three years before it takes hold.  Now, the administrative limb, you have to be cautious about what you're teaching. So on the afferent limb, the entire hospital needs to be educated.  Every person in a hospital has the potential of stumbling into a situation that is a crisis and they need to know what to do. The content of this course,  we've entitled the first five minutes, which describes not only why an RRS is needed, but then also how and when to trigger a response. So this is basically focused on detection  and then recognition and then calling for help.  In situ practice can help with this. So this is not something that should be done merely with pocket cards or educational pieces, but actually bringing it to the clinic  and the bedsides will be very helpful. The efferent limb, by contrast, we're targeting just the responders. This is a much smaller group of individuals  and they actually need much more detailed education regarding who is going to show up at the events, what equipment do you expect to show up at the events,  how I am supposed to move and coordinate with those people, whose responsibility it is to get key equipment, and then do training to make sure  that everybody is well-practiced in what to do when they get to the event.  Gabrielle Gossman at McGee Women's Hospital in Pittsburgh  created a novel rapid response system for obstetric crises and they call this condition O. And I wanted to show you, highlighted in pink here,  the huge amount of effort that they put into the educational component of their system. They didn't just create a bunch of responders and then set them out there.  They did a series of events over a period of about a year and a half that resulted in a change in culture. And you can see on this next slide  the impact in that change in culture. So the white bars are condition C events. These are typical medical crises. And the black bars are condition O events  and those are obstetric crises. And you can see when they started the educational program, it was about April or May, there were no condition O events and they gradually increased  until early in the following year where you can see how they peaked.  Interestingly, the condition C events also increased. So education regarding how to manage one crisis in a particular way increased the likelihood  that people would respond in the appropriate way to other types of critical situations. I think it's fascinating to see how those black bars decreased.  And I can tell you that this was not a result of failure to call, but actually improvements in their health system.  There are different ways of triggering the events. This is some work by Gary Smith where he looked at different types of early warning scores.  And he showed that there was a receiver operating curve, the area that was about 0.78, 0.8. Now with the analysis of big data and real-time systems, that is up near 0.9.  The important point that I wanna make  is that it's the area above the curve which represent missed opportunities to identify patients who will have a crisis. And this system is only as good  as the data that goes into it. So I think the key point is to recognize that there will be events that are unpredictable. And because there are events that are unpredictable,  intermittent monitoring can never be satisfactory because random events will occur to otherwise stable patients. We investigated this when we looked at  the diurnal variation of rapid response in our hospital. And we thought that we would have more events at night because staffing was worse. And actually it got lower at night.  We had higher rapid response events during the day. And then we looked at staffing and we thought it might be more visits during the day,  which made us ask the question, does monitoring matter?  So we stratified the events by day and night cycling in three different levels of monitoring. One is intermittent monitoring, that's on the right.  Then there's light continuous monitoring in the middle and then ICU monitoring on the left. And you can see as you increase the monitoring, the diurnal variation goes away,  leading me to believe that the more monitoring you do, the more likely you are to catch these random events.  Dr. Khanna explains this well. In the graphic on the left, you can see how two patients are stable at 4 a.m. At 4.40, they both turn blue.  One has a continuous monitor on, the other one does not. And obviously the one with the continuous monitor is detected sooner and gets therapy in the form of transfer  or therapy provided at site within minutes of a deterioration and is returned to normal. The one on the left who does not have continuous monitoring,  we have to wait for the next scheduled visit by a staff member, and that can be hours later. In fact, it could be so far later that the patient could have expired  by the time somebody sees the patient. There are a number of studies now looking at continuous, the impact of continuous monitoring. A recent meta-analysis looked at 40,000 patients  in nine countries in 24 studies. And the majority of these showed a decrease in critical care use, decreased hospital length of stay. Three had cost-effectiveness.  And the key barriers that were encountered by hospitals was nursing and patient satisfaction because of the burden of false alerts for the nurses and patients having to wear devices  that were impeding their ability to their mobility.  So I think we need to recognize that continuous monitoring improves detection.  So when the event happens, we need to know about it as soon as possible. I think it is very important to do good prediction  because we can prevent the deterioration in the first place for the majority of patients if we have accurate prediction and intervention.  It is interesting to me that the gap in monitoring is not at home. It is actually on the wards of the hospital. And this is because in the ICUs, OR, ED, PACU,  we have high-level monitoring. And when patients go home, they are actually monitoring themselves. And the percentage of the population  that is monitoring through pocket phones and watches and other wearable devices is becoming very high and is actually higher than what happens on hospital wards now.  So the danger zone for continuous monitoring is in the hospital. Let's talk about the rapid response systems quality then.  We would think mostly that the quality limit is intended to figure out how good the rapid response team is but actually it has a more powerful impact  if you collect all of those events and look for the antecedents. And in that, you will see collected a story that emerges like radiology is a dangerous place to go  or Ativan may be a dangerous drug to give to patients who are seizing. And so on. There are a number of events that you can find that are easily detectable  by looking at the things that cause crisis. And I can't tell you how many administrators or doctors told me, if only I knew this stuff was happening,  I would have done something different. Well, this is an opportunity for them to know. So that brings me to the point of rapid response systems.  This system of detecting and responding to an event can be done for any type of emergency. At the University of Pittsburgh, we had a huge list of crisis events  and created a special system of response for each one.  The list here is quite old. There is a new one code lavender, which many places are using. And this is to provide support to a secondary victim.  And this would be say the nurse or doctor who had been caring for a patient who died or had an untoward event happen to support them through the psychological  and emotional trauma of been through an event. I think the key point here is that you can design a specialized response for any particular problem that requires action fast  with specialized equipment and personnel. So let's turn back to COVID for a moment. And is there a rapid response system needed for COVID? And I think the answer to this is yes.  First off, we need to make sure that responders who are on route know that they're going to a COVID room. I think they will also infer that the response  may be delayed because many of these patients are on isolation. It gives them the opportunity to protect themselves on route.  It also, one might consider having specialized responders  or specialized equipment at the site. And when I say specialized responders, I'm thinking of people at a senior level who might make rapid sequence decisions  regarding code status changes, speaking to family and triage decisions in hospitals, which are overwhelmed with their ICU needs.  So in some, I'd like to say that rapid response systems are still evolving, and I think they hold merit for providing benefit for all types of crises, including those with COVID.  Thank you very much. ♪♪

Video Summary

In this video, Dr. John Whitcomb and Fiona Winterbottom discuss the evolution of rapid response teams over the past 15 years, particularly in the context of the COVID-19 pandemic. They explain how their hospital implemented a rapid response team to address preventable mortality and inpatient failure to rescue. They focused on early recognition, leveraging technology for surveillance, and providing expert resuscitation when needed. They also emphasized the importance of non-beneficial care and engaging patients and families in their care. They shared the various components and structure of their rapid response system, including education, monitoring, calling criteria, response teams, and evaluation. They highlighted the impact of their program, including a decrease in cardiopulmonary arrest rates, increased proactive screening, and successful upgrades to higher levels of care. They also discussed the challenges and adaptations made during the COVID-19 pandemic, such as the addition of a third rapid response nurse and respiratory therapist, and the use of continuous monitoring and artificial intelligence alerts. Dr. Michael DeVita further discussed the importance of rapid response systems in detecting and responding to critical illness outside of the ICU, and how the four components of the system (administrative limb, afferent limb, efferent limb, and quality limb) work together to prevent and improve the response to crises. He highlighted the importance of education, continuous monitoring, and quality improvement in the system. He also emphasized that rapid response systems can be adapted to different types of emergencies, including COVID-19, to ensure the appropriate and timely response to critical situations.

Asset Subtitle

Resuscitation, Professional Development and Education, 2021

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2021 Critical Care Congress held virtually from January 31-February 12, 2021.

Learning Objectives

Discuss the evolving changes in rapid response team makeup and participation modeling
Identify education thresholds and experience benchmarks required for participation in rapid response teams
Evaluate the impact of bedside interventions on patient care and the interactions of these interventions with standard unit policies

Team Leadership: How Has It Evolved?
Fiona A. Winterbottom

Response Activities: Impact on Patient Destinations
Michael A. DeVita

Meta Tag

Content Type Presentation

Knowledge Area Resuscitation

Knowledge Area Professional Development and Education

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Team Leadership

Tag Administration

Tag Healthcare Delivery

Year 2021

Keywords

rapid response teams

COVID-19 pandemic

preventable mortality

early recognition

expert resuscitation

non-beneficial care

continuous monitoring

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.