Eric Hoyer, M.D., presents at the Johns Hopkins Department of PM&R’s Grand Rounds on September 29, 2021.
um So I think many of you know me from my work and quality and safety here in the Department of Physical Medicine and rehabilitation. Um but in in addition to that I'm also the co director for the activity mobility promotion program here which is um a program where we're really focused on addressing patients function and and functional impairments particularly when they're when they're in the hospital. So today what I wanted to talk about is this whole concept of setting patient mobility goals um and and also some of the barriers in particular related to falls in the hospital. Um and uh this is actually sort of a novel talk that I'm giving. Um So I really welcome any feedback if you want to give it to me offline because I do plan on giving talks in the uh in the upcoming months in a different area. So so so so please I kind of and taking this opportunity also to kind of show some new stuff and and and would appreciate any feedback. Um So just to start off with in terms of disclosures, I do provide some consultation services which are not relevant. Um uh so I'll skip that and then for our objectives. So we're gonna discuss our um approach to measuring patient mobility in the hospital setting. Um I'm going to be talking about the concept of patient mobility goals that are based on their individual capacity. Um I'm gonna be talking about how we're using this data that's being collected in routine care to understand this concept of the tension between fall prevention and patient mobility as well, which is a pretty hot topic and then um will be discussing the use of this uh individualized mobility goals to to uh to overcome barriers and some of our strategies for that. So that's going to be sort of the um the map of of today's talk. But just to start start off with, I just wanted to just talk just briefly about um AMP AMP is actually a pretty big program here at johns Hopkins. Um It's you know, I would say in quick summary it's an inter professional program um we support hospitals, unit staff and providers. Um really the goal is to change culture of patient and mobility and and we provide tools and support resources to design and implement structured quality improvement. Um and it sort of has these different areas related to addressing barriers, sustainability and engaging clinicians uh functional measures in the hospital quality improvement, patient engagement as well as using data for for culture, change. Um the really the areas that I'm gonna be talking about in this talk really have to do with our research work as well as some of the tools and resources that are relevant for AMP. Um and and perhaps maybe some of you guys have also attended our conferences, we're going to be presenting at the upcoming I see rehabilitation conference and I see dr dale needham joining here um as well as in the spring as well. So let me just start with the typical patient story. Um This might be a patient that you can relate to. This. This maybe you know your grandparents or parents for that matter where you know at their baseline. You know they they're often pretty mobile, there might be pretty independent. Um And then what what happens is that patients and and this is a story that happens you know is very common. Is they spend most of their time in bed in the hospital setting. And and the results of that is that throughout the hospitalization particularly if they're in the hospital for at least a few days. Um And especially if they're elderly for example or they have other comorbidities and so forth that they experience pretty significant functional decline. And this functional decline is associated with lots of uh um clinical complications, poor outcomes um etcetera. And and this is really at a central core of what we're trying to address in our amP or activity mobility promotion program. This um just to kind of highlight this with like a recent paper that was just published this summer. Um This uh this was a study, it was in about 21 patients patients who were initially in the I. C. U. What they did was they put a little activity tracker um in this case they used this device called the tree activity um and they measure the number of the mean or the average number of daily steps um during the three days prior to ICU discharge. And then they measured it three days prior to hospital discharge. And then they measured it um for the first few days after hospital discharge and then 4 to 6 days. So they measured it, they averaged at those different time points across all the patients. Um and this is what they showed. Um So the first two bar graphs also, so the Y axis is showing their activity levels or steps per day. You can see that um uh you know, they have pretty low mobility in the ICU and and and and right before hospital discharge. And then there's this dramatic jump from home day one to day three, you know, and and so the question, you know, um the discussion of the paper is like, you know, did this, did these patients make this dramatic recovery when they went home or you know, is there something, you know, is this just kind of highlighting again this kind of culture of the hospital that patients probably have the capability to be more mobile, but but we're doing things or the way the environment is set up and so forth, that patients aren't really that active. So, anyway, this this is just another example of showing how how little activity kind of compared for example to home that patients are participating in. And we like to use the this term, the immobility harm um because really mobility um is related to all these different clinical outcomes that that the hospital really cares about. So things like falls readmissions de BTS, pressure injuries, delirium surgical outcomes, uh discharge disposition. If if we want to address all of these different outcomes, we really need to be thinking about patient activity and mobility and and for this talk, what I'm gonna particularly be focusing on is is falls. Now, how do we how do we do this? So so a lot of times in in clinical care um you know there there's there's been recommendations that have been provided um so for example, patients with heart failure and they need to be in a beta blocker. You know if you're going to go see a patient in the room, You need to be washing your hands. And there's a lot of science that goes behind that. But often what we find is that the science doesn't always translate into clinical practice. And so um there's different models of quality improvement or implementation science that have been developed to try to execute this and this is one um It's called the trip model um that we that was actually developed out of the Armstrong Institute of Quality and Safety here at johNS Hopkins that has these key components including aspects for example putting it within the context of the, you know, the problem within the context of the institution, summarizing evidence, addressing barriers, measurement and then at the local level, doing a cyclical process of engaging staff education, execution and evaluation. I'm not gonna for this talk, I'm not gonna really be going into the full trip model, but what I did want to focus on is the measurement component. And while that's particularly important for quality of improvement, there is this, you know, a well known saying and quality improvement that without measurement you can't you can't do que I. Um And and part of the reason also for in terms if we're thinking about patient function is that um we really, I think the ideal ways that we really want to be thinking about mobility um and activity as a vital sign like like as if it was blood pressure for example, we want to be able to track that over time. We want to be able to to see variants from what were, you know, what our goals are and so forth. Um And then the other concept to is that data drives culture. So being able to for example give feedback um to to uh clinicians who are working directly with patients creating accountability and things like that, that really those are important components of culture. Um And and data is a critical aspect of that. So how do we measure uh function um uh in in in the hospital. And there's definitely lots of different measures for those who are therapists um you know, for for example for assessing, you know, balance. There's you know, there's there's lots of different measures. Um but just from like a global perspective of what we're trying to really do or measure for all patients in the hospital setting, we think about kind of two concepts. Um and and and we we think that that really provides an important signal in terms of understanding about patients overall function and their activities and so forth. So the first concept is the idea of capacity and that's how much help that does the patient require to do a particular task, You know, are they independent doing a task that they need? The assistance of two people to do a task, something like that. And and and and and one concept that's important to keep in mind is that somebody might be capable of being mobile, but they may not actually do that. If you, if you think of yourself, for example going to the gym, you might be capable of doing 10 push ups, but you know, if you're just not feeling it that day or for whatever reason, you may not actually do that. So, so the amount performed is the actual performance and and we use a different measure for that. And that's sort of the idea of how much overtime did you do? Did you achieve some type of mobility milestone, for example, you know, were you able to walk outside of your room or or you know, something that could be measured, let's say with a device like a Fitbit, for example, like how many steps did you uh um walk today? So that that's an example of of performance. So the tools that we use here at johNS Hopkins um the first one um which we've actually been using for for a number of years now is called the johNS Hopkins highest level of mobility. Um It was actually developed off of ICU scale um but it's an ordinary scale that goes from 1 to 8 and it essentially has these different milestones, so there's different levels of bed activity um and kind of progression with that. Then it's this idea of kind of transferring for example to a chair or to a commode um and then and then doing a more upright position of standing and then and then we have different distances of of ambulance asian and and these are based on on key milestones that are you know based on the on the literature. Um so it goes from one through eight and what we found is that the scale is very easy to use and educate staff. Um It records again what the patients actually do um the standard of care that we use for example, uh both at johNS Hopkins Hospital and and and across the system that we advocate for Is that nurses document this twice daily and on each visit for physical therapy and occupational therapy and and kind of going back to that point that I was mentioning before about using this as a vital sign. Um this is actually a when we first initiated or use this tool into clinical practice. This is one of our early quality improvement projects. This is an example of a patient who was actually developing critical cervical in their neck, their stenosis and actually developing weakness and you can see that they came to the hospital, they are able to walk about 10 steps and then over time you can see that their mobility level was going down and this patient ultimately ended up needing surgery. So, so again, kind of towards that goal of of of using mobility as a vital sign um that gets towards that the um so that's so that's performance. Now in terms of capacity, the tool that and and many of you are probably very familiar with this is the an impact tool, but just you know, in brief explanation essentially what it does is it asks how much help from another person does the patient um require for these kind of six key tasks that are relevant for the hospital setting. Things like moving from lying on your back to sitting, standing up from a chair walking in the hospital room, um and this can be done both by patient report also by a clinician proxy. So, um it's not a performance test. You can actually, it's it's more of an assessment and you essentially on all of these questions you ask the or you assess how much help from another person is that patient required, Do they need total assistance all the way to the fact that the patient is completely independent with doing that task and and the and the frequency that uh that again that we recommend within the health system is that nurses are doing this once daily um and that physical therapy is doing this on each visit and I just put a little asterisks by this because I am focused on on basic mobility, but I also want to recognize that there is an uh impact activity scale um that's also being done by nurses as well as occupational therapist. But for the purposes of this talk I'm going to just be focusing on the mobility aspect. So so um uh maybe some of you are familiar um but these two tools, the highest level of mobility and the impact are actually trying to attract an epic um and this is actually a screen and you can see uh you know sort of in that vision of as a vital sign. This is an example of of a patient which I'll talk about uh and a little bit um and when we when we have worked with nursing because because nurses have been really um although AMP is like as I mentioned before is a is an inter professional multidisciplinary approach to to really promoting mobility, nurses play a really creepy key part of that. And so we've done studies to really look at the psycho metrics in terms of you know can nurses really do this well. Um And so one of the first studies that we published um this is a few years back now um we had nurses going into patients rooms, we had therapists observing those mobility sessions. And we were interested in looking at the test retest and inter rater reliability. We found that um uh their uh that the psycho metrics were excellent. We also had nurses comparing the amP Ac and H. L. M. Score or the JH LM scores with other validated metrics such as the grip strength cats, a dl two minute walk tests and the five times sit to stand. Um And so all those other measures are have been recognized with measures of physical functioning. Um And we were able to show strong construct validity with with reasonable correlations suggesting that these measures also do measure various aspects of of physical functioning. So so in you know, a brief summary what we did find is that there was strong agreement between nurses and physical therapists at least when the physical therapist was an observer in terms of these tools and that JH LM and an impact are reliable and valid and valid measures of patient and function when scored by nursing. Um We've actually conducted a recent study um to to sort of further develop this idea. Um So we actually had nurses and physical therapists independently Go in and mobilize patients on the same day at different times of the day. We did this in on 42 patients in a surgical patient population that study that I just showed you was in a neuroscience population. This is in a surgical patient population just to you know, to make sure that that our conclusions are also generalize able to different patient populations. Um And this is a what's called a bland Altman plot. And basically um the way to interpret this and I think the key is if you look at the Y axis, what we're what we're looking at are the differences of scores between nurses and physical therapists. So that if they had perfect agreements then their H. LM scores would be the same. And basically it would go on that that red line that's the depicted on this graph on the X axis. We average the two H. LM scores. And what you can see in this graph is light gray line shows sort of the overall mean differences between nurses and physical therapists. And so the fact that the line is just slightly below zero just means that on average the physical therapists are just mobilizing the patients to a slightly higher level than nurses um do in in in this uh small uh study. But I think what's also important to to recognize is that um particularly when we're thinking about kind of higher mobility levels. Like when we're thinking about ambulance nations. So I which I'm circling in that dark red line, is that actually the nurses are able to achieve the same or even higher H. LM levels for patients who who had less impairment. So you can see that they were able to reach those H. LM levels of 78 and sometimes even exceed the therapist by by A level. But probably in those in those patients that have a lot more impairments where I kind of circle it in that in that orange graph we see that that the therapists are kind of that we see that like negative one um are uh difference between the nurse and the physical therapist suggesting that the physical therapists are are mobilizing those levels a little higher. So um so this is again but important, you know background and and and sort of data to suggest that that nurses can, you know, for the most part mobilized patients pretty similarly to our our you know what we would consider gold standard therapists. Okay so let me so now that we have, you know, I've sort of explained our tools and and and hopefully you know convinced you that that um that nurses for example can can you know record these tools pretty well. Um I wanted to show an example of how we're using these types of measures to address this real issue that affects patients on a regular basis. And and the topic that I'm gonna be talking about is the tension between mobility and falls. So um this uh this this paper actually was published in Jama Internal Medicine in in 2000 and 17, you can see the senior author is Sharon in anyway who's you know, you know, considered uh you know, she's she's really well recognized for her research and clinician in in the field of geriatrics and and basically they wrote this this opinion piece or this viewpoint and they highlighted that there's this inherent tension between preventing falls and pro promoting mobility. So that's kind of the first line that I highlighted and then they go on to say that the hospital culture has strongly prioritized preventing falls with potential unintended consequences for patient mobility, functional ability and well being okay, So that's kind of the that that's that that was sort of the key points that they brought up in this paper. Um but I think there's a couple of gaps. So so first of all this is, you know, I would consider this an opinion article and it wasn't based on any empirical data, I think we can all relate to that, you know, taking care of patients, but but it wasn't based on data and and I think the question that that, you know, we were as in our group, we were thinking about, you know, are there strategies that for example, we're um that we're doing, you know, particularly for example, members of the nursing team to prevent falls that are really truly impacting mobility. So that was our overall question and I'm going to talk about how we kind of are are looking at this. So the so the first the first thing I think to consider is who has the potential to to ambulance. And I and I and I use when we talk about mobility um obviously there's different levels of mobility but um but oftentimes when people are thinking about mobility in the hospital setting, one of the key goals is we kind of want to get patients walking. So so just as a as a you know um to kind of just put a stick in the sand somewhere and kind of identify sort of a mobility level, we're going to focus on ambulance asian or walking. And this is the graph that I showed you guys earlier which shows the H. L. M. Score on top which is again their performance, what patients actually do and the impact score on the bottom this so this was a patient that I did a consult on to determine whether this patient should be going to inpatient rehabilitation and what you can see is that actually this patient they I had a lot of impairments initially. So their you know their impact scores were really at the floor of six at the lowest level. Um And and also their their mobility scores, you know, we're averaging between one and two. So really low level bed mobility but you can see is that an impact score is going up and we can't really say what's causing effect here. But but we see this trend that as the AMP ACT scores or their capability improves, their functional impairments are going down that their actual mobility there H. LM scores are also going up. So this is something that we have seen just in you know in clinical uh seeing patients um a lot. So what we did is we we did another study um we did a prospective study in 213 mobility sessions, this is with the nurses and the nurse's. What they did is they mobilize the patients and they scored the H. L. M. and the impact. And we wanted to see, you know, within the context of of of this study, did we see this a similar relationship and indeed indeed we did. So we'll just to orient you to the graph on the X axis is the H. L. M. Score that was scored by the nurse and on the Y axis is the impact score. And you can see this relationship that uh you know as an impact scores are higher or their capabilities better that they're hln scores also go up and um that correlation coefficient was 0.65. Um So again, you know, they're not measuring the exact same thing but they are there there's definitely you know, a relationship and and the p value for this uh correlation coefficient was was clinic was statistically significant as well. So so so one of the key concepts that we I have been thinking about is trying to match mobility milestones with their capacity. And and the other concept that we have been trying to do is is that we want these mobility, my milestones to be achievable. And and again I I think what's important to keep in mind too is is that and and I think there's further work to be done in the in the intensive care unit setting where there's other factors like you know medical stability and things like that. But in the non ICU setting sort of in what we would consider the general med surge floors, neurology etcetera. Um where we've done these studies when when we've looked at this data, What we find is that for example, when we think about a milestone like walking 200 plus feet or an H. L. M. Of eight, what we see is that when patients have an impact score of 20 for that they're able to achieve that milestone, about 90 per you know 90% plus of the time. And that if we looked at a different milestone like just being able to ambulance, like to be able to take 10 plus steps Um that patients are, you know that nurses can can be successful in achieving that milestone when the impact score is you know 18 or higher essentially. And then lastly like an activity such as transferring a patient to a chair that they are able to do that with when the patient has an impact score of of uh 10 or higher. So using those concepts, what we ended up developing is what we call the johns Hopkins mobility goal calculator, which which essentially aligns their capacity with an actual goal. And the idea is that on a daily basis a patient based on their impact score should have a mobility goal and we should try our best to try to achieve that. So that's the general concept and and to kind of again orient you um to the tool and perhaps maybe many of you have seen this on mouse pads. They do live throughout the hospital, but on the sort of to the left here. Um if you can see my cursor, we see different impact scores for example, 6 to 78 to nine, well maybe just focus on the 24 for example, on the top, their uh their goal for that day would be, is to walk an eight or walk 250 ft or more. For example, if their impact scores are 10 to 15, then then a goal should be for for the patient to move to a chair or a commode. And and we actually looked at this um we we studied this within the context of a quasi experimental perspective clinical study. We had a six month baseline time period in a six month intervention time period. And we actually did it at this hospital. So this was on our neurology unit. We had where the patient populations are actually pretty similar between um 12 east and 12 S. So so one unit served as the control unit and and one unit served as the intervention unit. We did education to staff. We educated um uh particularly members of the nursing team. And then we did an analysis of differences and differences comparisons between time frames as well as the control versus intervention unit. And what we found is that using this tool in the intervention using units compared to the uh control unit As well as compared to their baseline. That using this tool resulted in a in a relative increase in patients meeting their goal. So that increased by 18% and that was statistically significant as well as the overall mobility levels. That when we looked at the mean H. L.M. scores that it increased from 5.225.8 which was a 19% relative increase. And those were all statistically significant. So um so I wanted to set the foundation because I want to go back to this question about falls. Um And and so hopefully you know this idea about matching capacity with goals make sense. And now we can kind of get back to the falls question. So so again preventing falls in the hospital is really a key quality and safety goal. Um uh This is something that's recorded in terms of um national databases for nursing care, um levels of falls has uh you know, magnet status depends on on you know, incidents of falls, there's reimbursement that that's based on uh fall rates and things like that. So something that's really focused on quite a bit. And one of one of the strategies actually um and a lot of work actually has been done by nursing leaders here at johns Hopkins has to do with this idea of fall risk assessment. And so a very what's really well recognized right now is that a strategy for fall prevention is to have an accessible and up to date all risk information and prevention plan for all providers and patients and families. And this is the tool that we use here at johns Hopkins. It's called the followers assessment or the j frat and it has these different components um The way that they they determined this is that they had um they, you know, they went, there was a group of of nurse leaders, they went into the literature and they basically determined what we're um recognized as as important um risk points for Falls. And in a consensus meeting, they basically came up with this fall risk assessment tool and they assigned points to it based on what they felt was was kind of waiting these based on what what they felt was was more important. So for example, they felt like Uh you know, having medications was if somebody had a number of of uh medications that were associated with fall like fall risk drugs, sedated procedures that would have a higher weight than somebody's age, for example. So that's the tool. If you get scored higher than 13, you're you're considered high fall risk. And I just wanted to give a little detail. So the mobility component of this tool um looks at whether the patient requires assistance or supervision for mobility, whether they have some kind of unsteady gait or they have some kind of visual uh a visual problems. And so you can get a maximum of six points based based on those questions. And so the domains are age, fall history, elimination, medications, patient care, equipment, mobility and cognition. Okay, so the so um so now let's let's go back to this question of the potential for ambulance shin and this is why it's gonna become important for this analysis. So, one of the things that we wanted to consider and so this um, this algorithm should look familiar. I just presented a few slides ago is that we expect based on, you know, the research that we've done and so forth. You know, these clinical studies that patients, if they have an impact score of 18 to 24 that they should be ambulatory, they should have uh jH LM scores of sixes, sevens and eights for most of the days during their hospitalization. Uh So that's that that's another assumption that we made for this study. So the question that we asked is if you have a higher fall risk assessment is that associated with low mobility. So even if you have, so if you have, for example high fall risk um but you have the capability to walk, Does that somehow impact your ambulance nation status? So so here we have a patient who's coming into the hospital and what we looked at at a basic level is is on admission, you know, or close to admission. They're gonna get a j frat or a fall risk assessment. And then they're also gonna have an amp act or the sort of an assessment about what their capability of moving. And so the question is, what do we expect this patient? Is this is this patient gonna be you know, ambulatory, or are they going to be you know, predominantly laying in bed um or sitting. And and this analysis is based on a retrospective analysis of over 48,000 patients. Um And and again, I think that's, you know, one of the benefits of having sort of the systematic approach to measurement is because we have these measures of fall risk as well as the impact and H. L. M. On all patients. We were able to look for, you know, over to about a two, sorry, over two year period at this data retrospectively on non ICU units to to look at this relationship. So let's go back to you know our grandfather or this person that we were kind of talking about before. And the question is what is you know, what is the what is the what is the likely outcome? And I wish I could see kind of feedback but I'm curious um Let me see if I can kind of what what if what if what do people think? Um So if you have a patient who is pretty ambulatory and they're labeled as high fall risk, is this patient likely going to be ambulatory? Are they likely gonna be non ambulatory? What what's you know even they even though they at baseline they have that capability. What what what what what what what what do people think? I don't know if I can see the chat eric Yeah. Are you asking if they're high fall risk once they're in hospital, are they going to be more likely to be bed bound or more likely to be ambulatory is at the question. That's correct. Okay. I vote for bed bound. Alright. It looks like it's pretty, yeah, it looks like it's pretty, at least the yeah. The um consensus. Is that what what I think they agree with you dale. All right. All right. Okay. So let's look. Okay. So how do we do this? So so what we did is the first thing is the outcome measure um In terms of their ambulatory status in terms of being bed bound or ambulatory. So the way that we did uh um We we characterize a patient as being ambulatory or not was that we calculated their median H. L. M. Over the hospitalization. So if they were ambulatory that means that for for 50% or more of their hospitalization, they had an H. L. M. Score of six or higher. So if you looked at this patient for example who was in the hospital day one through five um you know on day one there were four but then there were eight and then 667. According to this definition this this patient would be considered as an ambulatory patients. So that's that's the measure that we and and and it's either it's dichotomous. So that means you know, was the patient ambulatory or not. So that was our outcome measure. Um And then we started exploring this relationship between fall risk categories. Um And and and this outcome measure. So here we're looking at um patients labeled as high moderate or low fall risk. And this is completely unadjusted. And this is within 48 hours. And what we see is this this stepwise progression. So so on the Y axis I'm showing the percentage of patients of ambulatory, so patients who have who are considered high fall risk within the 1st 48 hours. Um we're about you know slightly over 20% ambulatory during their hospitalization compared to patients who were labeled as were categorized as low fall risk were nearly 60% of the time they were um they were ambulatory and um for the purposes of analysis, We did also look at, you know the relationship of fall risk within 24 hours and we see a very similar relationship, but because we did have a lot more missing data within 24 hours for this analysis, we just used the fall risk score at 48 hours. And and and we felt like we would have a similar interpretation of our results. So in quick summary, fewer patients categorized as high fall risk or moderate fall risk or ambulatory compared to patients with low risk. And then the second question um that that we wanted to look at is is what is the, what is the relationship? Um um Sorry, I'm trying to the sorry, the zoom thing is kind of in the middle, I'll try to read through this. But what is what's the relationship between somebody's mobility potential and and their fall risk categories. So um the maybe uh what what you can focus on are the are the light blue colors. So so here we have three groups of patients, patients who are categorized as low fall risk, moderate fall risk and high fall risk. And then with each group we can see patients who have high impact mobility scores. Again, those patients that we expect to be mobile, They have an impact scores between 18 and 24. Um and we can see that patients who are low fall risk, you know, a significant percent of them have high impact mobility scores. But we also see a significant number of people who are labeled as moderate fall risk and high fall risk who also have those high impact mobility scores. Um And when we look at, you know, the total number of patients that you know that we considered in this analysis, About 60% of patients have have the potential to ambulance who are labeled as or categorized as moderate or high fall risk. Um and and you know, that's that's over 30,000 patients in the sample. So, so that's that's quite a that's quite a number. So now in terms of our actual statistical analysis, what we did is we we went through a process of multi variable variable modeling. So we considered the outcome measure as our as non ambulatory status. And what we wanted to do is we wanted to consider all these other factors that might be explaining their non ambulatory status. So we considered the different aspects of the J frat. So their age, elimination status, equipment, follow, history, medication, cognition. We also looked at their AMP act score. What they're ambulatory ambulatory potential was their length of stay there comorbidities. We also looked at just basic demographics like um like race and gender and so forth. But what are the key question that we were asking is at the end of the day, even when we account for all these factors, the fact that a patient is put into a category of low fall risk, moderate moderate fall risk or high fall risk. What is that association with that outcome of non ambulatory status? Um So um after we control for all those items. And so uh this is this is our our main finding that we have from our analysis and I'm just focusing really on those patients who have that high mobility capability. So again, these are patients who have that impact scores of 18 to 24 when we compare patients who are at high fall risk versus low fall risk. So that's that middle row right there there are odds of being non ambulatory is uh is slightly over to. Um And and again these all these p values are statistically significant. And I showed that 95% confidence interval. So that's suggesting that about they're more, they're about twice as likely to be non ambulatory patients who is um low fall risk even though they have the capability to to be ambulatory. And that's uh and we saw that relationship also uh consistent for patients when we compared more moderate versus low and even high versus moderate. So we really see that staff wise relationship. So again in summary, so highly capable patients to ambulance late categorized as moderate to high fall risk for falls were about 1.5 to 2 times more likely to be non ambulatory than those categorized as low fall risk for falls. So the other thing that we've been looking at is what is what's the mechanism like what what's you know, is there something about the impact of fall prevention interventions on mobility? And one of the things that members of the nursing team do is that when they score the J frat, they also select some interventions that they that they want to implement for the for the patient and that's part of their workflow. And there's a number of those interventions that impact mobility. So those are things like bed and chair, alarms, restraint sitters etcetera. And what we've looked at is on a daily basis. We looked at all the interventions that could potentially impact somebody's mobility. And we and we basically quantify them that they have 11 intervention per day or maybe they got one intervention every few days and so forth. And so um uh this table, what it's showing is uh we're looking at patients with different J frat risks. So we have at the top we have patients with low low risk, moderate risk and high risk and then um patients who are ambulatory or non ambulatory and and what what this table is showing is basically the rate at which the mobility of interventions are being selected by nursing to try to alleviate the patient's fall risk. So, so at, you know, for patients with low fall risk. Um very, very essentially no interventions that potentially impact mobility are being selected by nurses per patient for patients that are at moderate fall risk. Um you know, there are a few interventions that are being selected by nursing, but particularly when we see that patients who are non ambulatory, they're getting more interventions that impact their mobility. And then for patients who are at high fall risk. Um you know, first of all, we're seeing lots of interventions being selected by nursing that could uh impact mobility, but particularly for patients who are non ambulatory, we see the highest rates uh being selected for these patients. So this could be a, you know, uh possible explanation about why patients are less uh ambulatory. And and um these differences were also uh statistically significant. And it's something that we're currently working on, on writing up for publication. So, so in summary, having more interventions that impact mobility is associated with lower mobility. And patients categorized as high fall risk have more interventions that can impact patient mobility. That's that's I would say, kind of a big theme of what we're seeing with this type of analysis. So, one of the questions that we're asking is that Can patients, you know, depending with if they have moderate or high followers, can they be mobilized safely to achieve mobility goals. And so we actually have a study right now, this is on a surgery unit. We have 100 and 85 mobility sessions now um with with very engaged uh members of the nursing team and they've been doing, they've been mobilizing patients using the gold calculator, which I showed before and we've been looking at, you know, different barriers to mobility, you know why patients can't achieve their goals etcetera. And one of the things that we've been keeping track of is what patients fall risk is. So so in this case here this is mobility events where patients were labeled as moderate fall risk. And In terms of the frequency that patients actually met their mobility goal, we can see that these members of the nursing team were able to to achieve their goal and again this is during routine care, but these were considered uh you know, I would consider them champions of of mobility on their units, but they were able to achieve that 97% of the time, even though the patient was labeled as moderate. And even for patients who are labeled as high fall risk When they used that goal calculator, they were able to achieve the goal at 93%. So you know, a little bit of a little bit less but but definitely, you know, to a really significant um level. So why is this important? So we think that this concept of using um you know, somebody's capability um and and thinking about that within the context of somebody's fall risk is important. And so members of the AMP team um and I'll give a shout out for example to Annette Lovisa um who's been working with um the whole health system and developing a policy for fall assessment, prevention and management policy but I just wanted to highlight one of the key uh parts of that um policy now that that's now uh I uh my understanding is either pretty close to being implemented or or or or anyway it's it's it's very close I think I think is my understanding but is this idea to set mobility goals and address barriers impacting ability to achieve those mobility goals. So this again this idea of setting mobility goals has actually made its way into the policy which I think is is really an important way of of addressing the immobility harm um in relation to patients with fall risk. The other thing that I think is important to keep in mind that's relevant for AMP is that we you know, we talk about promoting promoting mobility but I think what's also important is this concept of safe mobility and so something that we've been also partner, another group that we've been partnering really closely with is the safe patient handling um group. And so the so hopefully you know this algorithm is looking really familiar to you guys now, you can see the impact score on the left and these mobility goals, but in addition to that we're also suggesting different types of equipment that can be used on the patient um to really help facilitate that because we also not only want to avoid patient injury, but we also want to avoid staff injuries. So that's that's another goal that we've been working on, related to all this work. And I think the other, the other point, you know why this is important, this is, you know, some some more recent analysis that we've been looking at is just who is the question of, is is who who are the patients who are at risk of falling? Um The way that the j fred is designed is that essentially you get penalized if you have more mobility impairments, you you kind of get penalized and the assumption is that you would have a higher fall risk. Um But when we when we look at the relationship between a measure like the impact score, which is here on the X axis and the probability of falling, we actually see a different relationship. We see this in verse you relationship. Um and this is data again from from over 40,000 patients of followers and on followers. And what it seems like is really that there's this kind of this this area where patients are at the highest risk of falling that are sort of in the middle. So if you're at the high end, your impact scores really high, you tend to have less injurious falls and also patients at the really low end, they also tend to have injurious falls. So so just giving this feedback actually to to nursing leadership has been really helpful um for them to understand about who who are patients that we might need to think about. You know for example using equipment or or so forth different strategies to mobilize them safely. And I think in terms of the overall goal of promoting activity and mobility promotion, if we kind of go back to this point, you know this you know our our grandfather or this elderly person that I kind of presented in this presentation, a lot of patients actually get when they come to the hospital have this ability to uh ambulance or to be mobile. But because of immobility them staying there, they're kind of coming down, you know lower in this curve. But but through this process of immobility is particularly over, let's say a prolonged hospitalization, we might actually be moving up moving them up this risk risk curve and that's that's a really important reason but why we need to be paying attention to that because we don't want them to be living high on this curve board where they might be at the highest risk for harm related to um uh fall. So just in conclusion um and then I'll have a little bit of time for questions um just wanted to kind of highlight some some some goals for the future in terms of uh in terms of our our work um this is obviously related to falls work but but I think as our amp group, what we're really trying to do is we're trying to think about how function fits into lots of different goals of of the hospital setting. Um So so I really a core, you know, ideas at the center of every care plan for a patient um that uh in every handoff activity and mobility needs to be prioritized. Um We're we've been working with all these different hospital systems to adopt a system wide interdisciplinary functional assessment strategy and and we advocate for the tools that we you know that I presented today. Um we really want uh these tools and and amP um these initiatives to really uh to be a patient's centered uh safety quality priority no different than than these other types of harms that that we often hear about um like hand hygiene or medication reconciliation and and as I mentioned before, we we really look for opportunities to kind of integrate um into all these different initiatives because because mobility for example, as like in the first one surgical pathways, mobility is really important for all these different aspects or or function I should say. And in activities is really important for all these different aspects of improving patient outcomes. So so the so the the better that we can integrate ourselves into the workflows into ST strategic priorities etcetera. Um I think we can really have a positive impact on patient outcomes um and then um you know we we continue to deploy our tools across the health system. We also are partnering um with with external groups um so we've been partnering with for example with new york age friendly health systems. Um you'll probably hear more about the learn uh system. We have actually a learned scholar who's right now at Hopkins, looking at the relationship between cognition um and some of these outcomes, we have some projects also at Intermountain boston Medical Center and we also have a multi center study looking at implementation of AMP at these three hospitals in Delaware colorado and Louisiana. So I think more to come about um also from an implementation science aspect about how we really deploy AMP and culture change across different institutions um beyond johns Hopkins. And then lastly we're also looking to deploy some of these tools to two other partners such as the V. A. And the D. O. D. So I think lots of exciting, exciting things um to come in terms of AMP Um so here's a link to uh to our website. We do have a campaign for you know everybody moves which is um uh you know an effort really internationally to to try to bring awareness on two to reduce the immobility harm and I think with that I'll end the presentation and and take any take any questions. Mhm. Eric I think there's a question from Megan in the chat. Hey eric, I can also just follow up with my voice. First of all, I just wanted to thank you and a net and everybody who's been working on AMP, it's sort of staggering to think about how much harm reduction has happened here at Hopkins in multiple other places where you guys are doing all this training. Um, it is just a phenomenal accomplishment. And the question was as, as an AMP evolves and um the and you guys have sort of progression in your goals. I'm wondering if there's a plan for a workflow for people who are designated as at baseline when they're in acute care. So for example, I'm thinking of folks who have spinal cord injuries or um like a muscular disease genes dystrophy and these folks might have been active in their own ways at home, but when they come to hospital or a designated um, kind of at baseline and then the diffusion for keeping them act, there's a diffusion of responsibility for keeping those folks active. That's just, that's an anecdotal observation. And I could also be wrong about that. Yeah. You know, I don't know if Annette or or samia are on this because I think that that's just from an operational standpoint, I think they would probably be best. And so I'm I'm happy to follow up with that. I would say from a research perspective, we've been, I think we've been focusing on kind of the, you know, The broad patient population, you know, the 70% and and trying to make an impact, you know, kind of trying to move the needle forward and I definitely agree that there's patients I think who are, you know, perhaps at more risk or or um more vulnerable, who need a rehab plan for when they're in the hospital um and and how that's being implemented in particular, I think that's something that I would love to hear. Um you know, our our acute care therapy team um address eric, I'm gonna jump in this is dale uh with a couple of comments primarily because you you said that you're going to go on the road with this and if I was a skeptical audience member, I had maybe a couple of questions to ask, but that's the big picture is, I think this is just absolutely fantastic. I think your presentation is very motivating, really important, a huge body of work. So I'm just gonna nitpick about a couple of smaller things. Um so one thing I wonder about is I presume that you got the data on bed, chair, alarms, restraints and Sydor use from epic. Uh and if yes, just how accurate are those data when you're actually doing modeling with them. Um so so these are, these are, it's part of the, so it's it's actually not based like it's not the device, it's it's what the nurse selects. So part of in the flow sheet that the nurse selects as part of their workflow when they're completing their fall risk assessment is that they actually have to they have to select um uh interventions that they will carry out as part of their care plan um to try to mitigate the fall risk. And that's the data that we we were collecting to for for that for that analysis. Okay, great. So so that that that's really helpful naively. I wasn't sure how those data came about. Um You know, I don't I think for the sake of your presentation you could gloss gloss over that. The other thing that I wonder about from a statistical point of view is um you have a multi variable model that I think includes, I think it's called the h frat score. The fall risk score and then also includes the components of the fall risk score. And I recognize what you're trying to do. I I know the research question you're trying to tease out, you know, it's sent off a kind of an amber flag in my head from a statistical point of view regarding, you know, having a composite score and then all the components of the composite score and the same, the same regression model. You know, the the the easy question is is there a multi culinary t and presumably that was tested for and there's no significant multi culinary t. But then the second question is does that is, and I don't know is there any sort of concern in terms of the statistical model, when when when it's designed like that. Um Yeah, yeah, so so um you know, you you know Elizabeth well and and you know, this is something that we talked about at length um I know that we have tried, you know, we we definitely we we felt that it was important that in order to get to the category, the fall risk category, we felt like it was still these these these components were still um uh important beyond the categorization. And we did look at things like the multi culinary et issue, which was which was not an issue. So so that that was um and and and and we also did sensitivity analysis because you know, even you know sort of the presence or absence of these different impairments or the actual raw score. So we looked at it in a in a bunch of different ways and essentially our results were consistent and similar. And so I think that was a key reason about why we felt confident that that the that the results that we were getting were were were were accurate or true. Great. Great. And I think I I don't imagine that probably wouldn't come up in a presentation, it might come up in peer review of a of a paper and I presume that you had thoughtfully talked about that and that sounds sounds great, thank you so much. Yeah, of course. Thank you. Thank you dale for those comments. Um so I think our time is technically up but I wanted to say again, thank you dr Heuer for presenting your research. I get very excited to see all the work we're doing just to get our patients moving and I know that the group on this call is very invested and excited about that as well. So um again, thank you for your work and for sharing and I'm sure we look forward to hearing more about how this progresses. Um So everybody, I I hope you learned a little something, I'm sure dr Heuer will be happy to take questions via email or other um ways of messaging if you have more and I hope everyone has a great day. Thanks everyone have a nice Wednesday.
Facebook
Twitter
YouTube
LinkedIn
Instagram
E-News
BROADCASTMED