SCCM Resource Library-What Data Do I Need to Collect to Measure Bundle Compliance and Performance?

Video Transcription

Good afternoon, good morning, good night, depending on where in the world you're watching this video. Welcome to the 51st Critical Care Congress.  I wish, I so wish that we could all be together in Puerto Rico celebrating this wonderful Congress but unfortunately we're here on video instead. My name is Michelle Balvis.  I am currently the Associate Dean of Research at the University of Nebraska Medical Center College of Nursing and I am here today to give you a presentation regarding how you  are going to collect, measure, define success, know that you're doing what you should be doing, proving your worth, but really what we're going to discuss today is how we're  going to use data to facilitate implementation and sustainability of the ICU liberation bundle. Before I begin the presentation, I would like to just pass along my personal thanks  for everything that you guys have done for your patients and colleagues in the last, I guess we're going on two and a half years now.  My hearts are with you and my prayers daily thinking of you and all the work that you've done in terms of your critical care practice.  So I thought I would divvy up this lecture in a way that is more kind of like tip focus, so kind of things maybe you could take back to your actual practice when you're going  to implement the bundle. Again now that we're regaining some sense of normality, recognizing that perfect sense of normality yet, but some sense of normality back in our ICUs.  So the first tip that I'm going to give you in terms of data collection with the ICU liberation bundle is to stop.  So before you go and start collecting and measuring data, you really need to define the things that you're measuring. Sounds simple, right?  But we know that the ABCDF bundle, the ICU liberation bundle does have a lot of different parts. So the first kind of tip that I'm going to give you is don't reinvent the wheel.  We recently, the Society of Critical Care Medicine conducted the ICU liberation campaign and it involved I think 68 adult intensive care units and some really highly motivated  pediatric colleagues as well. But we've done a lot of this data measurement definition work for you.  So with the ICM ICU liberation collaborative, we did define the individual ABCDF bundle components.  Within the collaborative, the A, the assess, prevent, and manage pain part, we defined the A part as a patient in the ICU had at least six pain assessments per day and those  pain assessments occurred using one of the valid and reliable instruments that you see on the PowerPoint here.  So either the numeric rating scale, the CPOT, or the behavioral pain scale, all right? So the A, when you're going to define before you measure A, it is focused on the assessment  prevention and management of pain. And we operationalized how frequently those assessments should occur to be at least six times a day.  The B in the bundle is both spontaneous awakening trials and spontaneous breathing trials.  And obviously the SATs were defined as being performed daily on patients who were receiving  continuously infused sedatives, so both analgesics and sedatives, or if the patient, and I think this is rather rare, but should a patient be ordered pain medications every hour or  every two hours around the clock, regularly scheduled sedatives were also considered in the SAT definition.  SVT, we wanted SVTs to be performed on patients, on all patients requiring invasive mechanical ventilation.  And obviously those SATs and SVTs, it's beyond the scope of this presentation, but we want those coordinated, right?  We know the best outcomes come about when the sedation shut off before the spontaneous breathing trials are performed.  But at least once a day, a patient on a continuously infused drip had their sedation shut off and  had an SAT, and all patients on the mechanical ventilator that were eligible, that were on the mechanical ventilator had a breathing trial once a day.  The C in the collaborative was focused on the choice of analgesics and sedatives. That was a little bit harder to operationalize.  So during the collaborative, we decided the C will be measured as the frequency and level of arousal assessment.  So meaning how many times a day did the patient get their sedation assessment? We use the goal of at least six level of arousal assessments per day.  And similar to pain, we want those assessments to occur using valid and reliable tools. So the RAS or the SAS.  The C also component of it, although we didn't really measure it as much, but the goal being, we want light levels of sedation, and also it's great to set target levels of sedation.  To operationalize, it's that they got those six level of arousal assessments. The D, the frequency of the delirium assessment.  In the collaborative, we operationalized that they should get at least two delirium assessments  per day on all patients on the vent, not on the vent, all patients should have the delirium assessments.  And it's the same thing actually with C. That's one of the frequent questions that I get is, well, how often should the patients on the ventilator get, how often  should we document their level of arousal versus how many times should we do it? For consistency sake, we just kept the number of assessments for vent and non-vent the same.  So the D, again, two delirium assessments per day using the confusion assessment method or the intensive care delirium screening checklist.  The E, this was a rather controversial one, but how are you going to know your patient got the E?  Well, in the collaborative, we had a lot of great back and forth discussions, but in the collaborative, we decide that the patient got the E if they had mobility activities  that day that were higher than active range of motion. So if a patient received passive range of motion, active range of motion, or on a bed,  or you're just turning them, that doesn't count as early mobility. We defined early mobility is that it was documented that the patient at least dangled at the side  of the bed or was actually up walking in the room or in the hall. All right, finally, our F, we operationalize family engagement empowerment, meaning either  a family member or significant other came and visited the patient and was educated on the ABCDF bundle, or that family or significant other participated in rounds, conference plan  of care, or ABCDF bundle. So either of those would be counted as the patient received the F that day. Now, again, I said a lot of this work has been done for you.  This is a great resource for you. This is the SCCM's Adult ICU Liberation Minimum Dataset already here.  It's on the website, I think it's on the website, that gives you a data collection form with some of the definitions that I just described.  Some of them might be a little bit different, and you'll learn that some measures might need to be tweaked for your individual ICU based on where you are and where you want to go.  But again, if you look at this form, and we won't go through all of them, but you see the A here, it just asks on this data collection form, the number of times a pain assessment  was documented using any of those measures that I mentioned before. So the self-report or the behavioral pain schedules.  And then after that question, the follow-up question, again, just for this element would be, out of all those assessments, how many of them were significant pain?  Because you might have a quality improvement initiative in your unit, right?  Ongoing quality improvement initiative in your unit that might look in that, you know, a lot of patients are having pain, what can we do about it?  Well, you'll need to know how many patients were having significant pain. So great resource for you here, already developed.  You could peruse, and it does go through each of the elements of the ABCDF bundle.  So tip one, clearly define your bundle. Clearly define what means the team members did the bundle.  So again, just because it's the A and it's easiest to go to, our operational definition was A was a success and A was given if they had those pain assessments.  The next thing, my next tip, there's a lot of controversy surrounding, you know, what's more important, performance or compliance data?  And it's beyond a theoretical discussion, to be honest with you.  So we know that parts of the ABCDF bundle, the ICU liberation bundle, many of them need to, it's recommended that you have safety screens before, right?  So before we shut off the patient sedation and do an SAT, we do a safety screen to see if it's safe.  Before we shut off the SBT, I mean, before we shut off the patient's ventilator, we determine if it's safe to shut off a patient's ventilator. Seems kind of common sense, right?  But those safety screen criterias, they're important in that the measures that are in them may actually serve to inhibit your performance data. Now, what do I mean by that?  So when I'm talking about performance, what I'm really referring to is more of a yes, no concept.  So if you're thinking performance, you're thinking, was the patient eligible for the intervention and did they get it? You're not going to be thinking about safety screens at all.  Performance is, if the patient's on a continuously infused sedative drip, did they get an SAT? It's a yes, no. There's no, not applicable, anything.  They either, in the last 24 hours, got that SAT or they did not get that SAT. No judgment there. There might be a very good reason why the patient didn't get an SAT.  Maybe they're on 100%, 20 a peep, pressers the max, but that's not what the performance data is going to ask. It's just simple, yes or no.  Did the patient get the individual bundle elements?  With that performance, we've defined, you have to think about what being performed. Do you want to know if the entire bundle's being performed?  If so, you want to know complete performance. And that being defined, they got every single element of the bundle that they were eligible for, right?  Eligible meaning, again, take out that safety screen aspect of it. If they were on a drip, they got the SAT. If they were on the vent, they got the SBT.  If they were in the ICU, they sat at the edge of the bed or higher, right? So all of the individual elements, they got all of them, 100% complete performance.  We also know that you could then, another way of looking at it is, is you could look at it proportional, right?  So how many of those bundle elements that they could have got, did they actually get? Because you might see as you develop over time in your ABCDF bundle performance, you  get better over time and they're getting a little bit more and more and more of the bundle.  The compliance is very different. And again, I personally am not a believer in compliance because I've seen so many problems  with having these safety screens with 25 criteria that the patient has to pass in order for them to have their sedation.  Most of them not based in evidence that then serve to inhibit.  But if you're a really big believer in that, it's really important to capture that data about why they're not getting it, you might want to look at compliance data, right?  So with compliance data, you're looking at, yes, no, the intervention was received in those that were deemed eligible. So they're redeemed that person eligible.  So whoever deemed that person eligible to get it, meaning they pass those safety screens,  they got it.  Both again, very important, but I think they both have very different reasons for existence.  Excuse me for one second. Let's go back here.  All right. And this is why I say they're both performance. If you look at the big outcome paper from the SCCM ICU liberation collaborative.  So when we looked at those 68 ICUs and we looked at performance, and that's, I think what we reported in that paper was mainly complete and proportional performance.  So simple, yes, no. Did the patient get it? Again, not looking at whether they were, whether it was safe or not.  We found that every, except pain, every outcome that we looked at got better  with complete performance, right? Patient's more likely to go home, less likely to die.  But we also observed, so it was great. The best outcomes were if the patient got the complete bundle, they got every element that they were eligible for, they had the best outcomes.  The good news is we also saw improvement with proportional performance, meaning the more bundle elements the patient got, the better their outcomes.  And again, this is, if you haven't looked at this paper, if you just look at the consistency of results, it's just absolutely amazing to see the consistency in terms of proportional  performance. So for example, if we're looking at ICU discharge, the more bundle elements, so they're down  at 33%, they got a third of the elements that they were eligible, the third of the elements that they were eligible for, they had better earlier discharge, right?  60%, even better. 100%, even better. And it's the same thing with, again, all those outcomes. Look at death, right?  And this one's a little bit opposite because you want less than one to be favorable. But the more bundle elements that the patients got, the less likely they were to die.  So both complete and proportional performance have been strongly demonstrated associated with improvements in a number of patient outcomes.  But that's, again, could be another seven-hour lecture.  Another really fascinating thing that we observed, and when you're considering your data collection efforts too, this might be a little bit heartwarming.  And again, we all acknowledge what a big setback we're taking in our ICUs during this pandemic, right?  We know that we've lost years and years of improvement in terms of evidence-based ICU care delivery.  But this might be a little bit good news for you, right? So we asked the question of whether participating in the collaborative led to increased bundle performance.  And in fact, it did, right? We had these great changes, and you can see that in the top slide. Again, it's illustrated in the proportional performance.  But when you're looking at that complete bundle performance, we made really great progress. If you're looking at that pre, the flat line, and then the sloped line, we made slow and  steady progress, right? Isn't that beautiful? You see that? It's going up and up and up. Each month, it goes up and up and up.  It's great. Almost tripled, right? You're looking at, we almost tripled ICU performance.  So if you look where we're starting at, like 3% there and went up to 12%.  Oh, no. So that's even higher. Anyway, the point being, you can see this nice, steady increase over the months in complete bundle performance.  Unfortunately, when we're looking at that complete, you can see even in these highly motivated, wonderful, progressive, all insights, over that period of time in the collaborative,  we had that improvement, but that improvement was indeed slow, right? And there were still really great opportunities to make some more improvements.  So, steady, they say, always wins the race.  Tip three with data collection.  In addition to those measures that we just talked about that you will, as a committee in your hospital, think about what's most important for you to capture, I'm also going  to throw a couple other variables out there that we've been finding based, again, on some analysis that we're currently still doing on the ICU liberation collaborative.  It's just the gift that keeps on giving.  About some other things that you might want to consider in terms of data collection.  So tip number three, we're going to consider some patient factors that might affect your ABCDF bundle performance.  So this is a paper that might be published by the time you're watching this slide. It'll be published in CHESS.  We asked the question of, we wanted to know what are the factors that were associated with SAT and SBT performance, right?  So everybody has their ideas of why people get SATs and why they don't. There was really kind of gray area in terms of objective data.  So we took the data from the ICU liberation collaborative and we asked, what factors are considered with next day SAT or next day SBT performance?  And we won't go through all of these, but you can read the paper when it comes out. But some of the things that we found that you might want to consider when doing this  data collection is in terms of, you know, if your performance might be low, there might be a reason for it or things that you might have to modify later.  Things that we found associated with next day SAT performance was age and SBT performance was age, BMI, and diagnosis, right? So not really too surprising.  The diagnosis came up there. You know, the sicker you are, the less likely you are to have that SBT perform the next day.  The age was a little bit counterintuitive as well, meaning that the older people we found were actually more likely to have a SAT and an SBT than the 18 to 39 year olds.  So that was kind of surprising. So think about the patient population, obviously, that you're dealing with.  We also want to consider some practice factors, right? So we talked about that significant pain a little bit, but we did find that an episode  of having an episode of significant pain actually was associated with a lower chance of having that SAT perform the next day, but higher chance of having that SBT.  Pretty consistent, we did find if you're in a deep sedation, if you're deeply sedated or in a coma, so again, your RAS is minus four or minus five, guess what?  Next day, you think you're more or less likely to have an SAT or SBT, much less likely to have both. But again, the importance of, if at all possible, avoiding deep sedation.  Well, that's, you know, deep, deep, deep sedation. Interestingly, we also found some practice related factors were associated with having those SATs and SBT performed, right?  So patient being physically restrained, I hate to say this and it breaks my heart, but those, I'm not in any way advocating, and I never will to the day I die, applying physical  restraints, but we did find, you know, an association between physical restraint use and the increased SAT, next day SAT and SBT.  Look at this though, this is probably the most amazing thing to me because we toiled and toiled, how often should we do level of arousal assessments?  How often should we do pain assessments? How often should we do a delirium assessments? And honestly, don't tell anybody, we kind of made it up because there wasn't a lot of  good data, right? We knew deep sedation bad, light sedation good, but those definitions were kind of all over the place.  This to me is fantastic and really intuitive to me, but what we found is the more often  the level of arousals were assessed, the more likely the patient was to have an SAT and an SBT next day, right?  And if you were only doing those level of arousal assessments three times a day, the odds of having that SAT and SBT went down the drain.  So when you're thinking about your protocols, please do consider how frequently you're doing your level of arousal assessments and the same thing for delirium, right?  Again, we guessed, so we said, let's do two.  And so that's how we operationalize it. And we found that actually in the more frequent delirium even assessments were associated with better SAT and SBT performance.  So those two, that finding alone is really pretty fascinating to me.  Finally, as I said before, the C as we define the C really does refer to choice analogies in sedation.  However, to operationalize that would have been a nightmare to do with 68 different intensive care units. So we chose to do that.  You know, are they getting their level of arousal assessments?  But clearly the meds patients get during their ICU stay matter, right? Medications matter. And we know that. And Patty's guidelines pick them up. We know what meds are good or bad.  Some people have their babies and love them more than others.  But what we found was actually, again, pretty consistent with the literature in terms of benzos. Benzos bad. There's a reason there's a B in benzos. Benzos bad.  You know, again, not for everybody.  I know some people need them. They're on them. Alcohol withdrawal. I get it. But in your head, if you think benzo bad ballast, you might think twice before doing that two  milligram IV out of hand push. Anyway, so they got benzos less likely to get the SAT, SBT.  Propofol you see more likely. Dex more likely.  Ketamine less likely. I have a bias against ketamine. I don't know why. But we did find in this analysis that ketamine, they were less likely to get both the SAT and SBT.  And also your typical antipsychotics for your SAT.  So kind of neat stuff to think about in terms of other data you might want to collect while you're going away.  So my last tip is to ask the right data related questions, right?  So ask the right questions. Do I have the right tools, right? Is your ICU using the right tools? Are they using those tools?  And again, I know it's hard and we fought with many ICUs. You know who you are out there. About why these, you know, why these recommendations were made.  But the current evidence did say that the ones that are recommended in the PAD-Ease guidelines were shown to be valid and reliable.  Again, you have special circumstances depending on your patient population. But the recommendations in the PAD-Ease guidelines in terms of the tools you should be using  are really the most up to date. Do I have the right frequency? And you saw that amazing groundbreaking news. There's a reason we do these assessments.  Do I have the right frequency? Am I doing the level arousal pain assessments, delirium assessments as often as I should?  Am I doing them the same frequency for both vent and not vent patients, sedated and non-sedated patients?  Sedated and non-sedated patients.  Am I doing my SATs once a day?  Some ICUs are doing SATs and SBTs two times a day. I am not aware of any data showing that, you know, doing them twice a day is any better.  If we could just get the once a day up across the nation and across the world, I'd be gleeful and I could retire and do what I really want to do, which is just struggle with my  new shih tzu. But do you have the right frequency for your SATs? Are you burning everybody out by making them do these SATs and SBTs two times a day for no reason?  Just think about it. Do I have the right standardization? Am I training people the same way so everybody's talking the same common language?  Those safety screen and success failure criteria.  Oh, I don't have it on here, but I shouldn't have made it tip number seven. Get rid of unable to assess on every form.  Trust me, you'll thank me later. No options for unable to assess.  But do I have the right safety screen and success failure criteria? Again, go to the evidence.  I'm not going to tell you here what the evidence is, but when you're making those safety screens,  base it on evidence.  Check your policies and procedures. Ask yourself, do I have the right policies and procedures?  You will be shocked when you see how many of your policies and procedures contradict each other.  Do I have the right educational modalities?  Are my trainers trained?  How am I doing the education? We found a lot of ICUs reported that teach back or those spot checks, super duper duper important.  If you have the absolute luxury of having a clinical nurse specialist or another clinical nurse leader who can take on this kind of initiative, that would be fantastic to do  those spot checks. So I'm going to tell you, your RAS score minus two is a little bit different than my RAS score to do the spot checks.  And do they have the right people?  We know and we all acknowledge exactly how much time, treasure, and talent it takes to provide our patients the best care.  So asking yourself, do I have the right people at this moment to even make a big push to get this done? You need the right people.  Including your IT support, who's going to help with all of your data collection, hopefully efforts. Hopefully all this will be extracted from the electronic health record.  And there's another wonderful presentation that you'll be getting from one of my colleagues in this session about all the stuff that's already built in.  You don't reinvent the wheel.  And that's it. That's the tips that I have. But again, the rest of this symposium is going to be fantastic and very, again, hopefully  pragmatic building on kind of the stuff that I mentioned, particularly with how to capture this data in electronic form to save yourself a lot of time and effort.  Thank you for inviting me to be here. I so enjoy and I so miss you and I can't wait to see you at next year's Critical Care Congress. I forget where it is.  Hopefully it's somewhere warm.  Have a great day, guys.  Thank you.

Video Summary

In this video, Michelle Balvis, the Associate Dean of Research at the University of Nebraska Medical Center College of Nursing, discusses data collection for the ICU liberation bundle. The ICU liberation bundle consists of different components that aim to improve patient outcomes in intensive care units (ICUs). Balvis explains that it is important to clearly define and measure the components of the bundle before collecting data. She suggests using the definitions provided by the Society of Critical Care Medicine's ICU Liberation Collaborative. Balvis also emphasizes the importance of assessing patient and practice factors that may affect bundle performance. Age, BMI, diagnosis, and medication use were found to be associated with SAT and SBT (spontaneous awakening trials and spontaneous breathing trials) performance. Lastly, Balvis advises asking the right data-related questions, ensuring the use of appropriate tools, standardization, policies and procedures, educational modalities, and having the right people involved in the data collection process.

Asset Subtitle

Quality and Patient Safety, 2022

Asset Caption

Utilization of the ICU Liberation Bundle (A-F) has been shown to decrease delirium, coma, restraint use, skilled nursing facility discharge rates, readmission rates, ICU and hospital lengths of stay, and mortality in ICU patients. Yet incorporating ICU Liberation work through the electronic health record (EHR) presents unique challenges and opportunities. This session will provide recommendations to clinicians to implement and sustain the ICU Liberation Bundle through use of the EHR, particularly leveraging the data.

Meta Tag

Content Type Presentation

Knowledge Area Quality and Patient Safety

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Evidence Based Medicine

Year 2022

Keywords

data collection

ICU liberation bundle

patient outcomes

intensive care units

components

definitions

	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.