Morning, everyone. First of all, thank you for getting up so early. I hope everybody's had a good conference so far. I spent last night catching up with old friends in a loud environment and really strained my voice. So hopefully, we'll get through this and I'll still have the voice. Really excited this morning to tell you a little bit of what the Academy has been doing with our registry and how we can take the information from a registry and really use that data to help improve patient care. What we're gonna talk a little bit, I'll start by giving a little overview of the registry and then my colleagues will talk about their individual experiences at their own organizations and then we'll kind of wrap it up a little bit with what's next from the Academy as far as quality and research. None of our faculty have any conflicts of interest. Dr. Wong is the incoming registry steering committee chair. I don't really think that's a conflict but just to let you know. So what's our challenge in our day-to-day clinical practice? Well, we know that the care continuum, a lot of times, it's inefficient, ineffective. We don't always have good ways to measure our outcomes. We have stressed clinicians. We know empirically or at least we think we know what works but we really can't always show the data to prove the clinical results. Often, rehabilitation is devalued by insurance companies, other providers, and there's really little national aggregate data or research available. You know, we, certain back pain issues, we give an epidural, we think that's the right treatment but we don't always have that data to support it. So we started back in 2016 and took a journey on trying to create a mechanism to collect data, to really use big data to be able to solve some of these challenges. And that leads us to the solution, the AAPMNR Registry, which is a single repository of data that will track real-world care, help us define rehabilitation practice, move rehabilitation forward, manage patient populations. It will be a method to help benchmark, excuse me, benchmark your practice to other providers, similar providers, and hopefully, we think it will improve patient outcomes. It's the first registry dedicated to PMNR and it's the only registry that's really focused on long-term outcomes, outcome data collection. So far, we've had some very prestigious organizations who were gracious enough to be some of our early adopters and were helping us work out some of the early bugs. So we've had Alina, Brooks, Shirley, Ryan, Shepard, Russ, just to name a few, Vanderbilt. So we really appreciate their stepping up to the plate and being some of the, lack of a better term, guinea pigs. So how's the registry work? Basically, there's two areas of clinical focus right as of now, ischemic stroke and low back pain. And what we do is we have two different types of data that get entered into the registry. We have clinical data directly from the EHRs to the registry, ideally pulling in directly so it's somewhat seamless, somewhat. And then we have patient-reported data to registry. And that's what really makes us unique because when you look at other registries, none that we were aware of utilize patient-reported data. And ultimately, that's really our goal. What's the outcome in the patient's perspective? So we combine that clinical data and the patient-reported data that really gives us this rich patient-centered outcome data set. And then we continually review the data, capture both clinical and patient-reported. We look at 10 different domain areas within the EHR from patient demographics and counter details, practice or clinician details, the actual diagnosis, coverage details, medication details, observation details, order details, and procedure details and referral details. We have 92 total data elements that are collected for both the low back and the ischemic stroke population. Of those 92 elements, 39 are required in the low back population, 43 in the ischemic stroke. Highly recommended another 30 plus for each data. And then around 20 optional data points for both the low back and ischemic stroke domains. Then we have the patient-reported outcome data elements. We use PROMIS domains of physical function, anxiety, depression, fatigue, sleep disturbance, ability to participate in social roles and activities, pain interference, and pain intensity. In addition, there's additional questions for both ischemic stroke and low back. We have work status or the return to work, blood thinner medications, complications, patient satisfaction, readmissions, medication adherence, recreational drugs. For ischemic stroke, we also look at alcohol use. And then when we started, we had exclusion criteria of prior back surgery, cancer diagnosis, and workers' compensation. Because as we were first starting to collect the data, we really wanted very clean data and see how it would work. But we've opened it up more now to include prior surgeries and workers' compensation because that is part of the total care and part of the analysis we want to be able to see. So how it works from just a workflow standpoint for the patient-reported outcome, we'll have a baseline survey, ideally completed at the first visit. If for some reason it's not completed at the first visit, the patient will receive an email to complete the survey within seven days of their visit. It's a private link totally dedicated to that individual patient. And then the patient will complete patient-reported outcomes at six-week intervals, three months, six months, and 12 months, which gives us that chronological look at a patient's treatment and recovery. Same thing for ischemic stroke, except the baseline survey would be at the time of discharge process. But again, if that does not happen, they will receive an email to complete the survey after discharge. And then we look in the ischemic stroke population at 30 days, 90 days, and six months. So now, kind of the exciting part, this may be the geeky engineer in me, but we've given lectures on the registry at several academy meetings, and up until only recently, it was our theory of how the registry would work. You know, this was the ideal and what we're working towards. Now we've actually collected data and you're gonna be able to see the data at some of our participating organizations in just a minute, but I wanna kind of give you this little bit of an overview. You know, we can break it down in a lot of different ways. We can look at the demographic data by race, by ethnicity, age, sex. How does our patient population compare to the aggregate data as a whole? We can look at the PROMIS scorecard, and what we have is two of the domains there, the physical function and pain interference, and you can look at it over time. As far as the physical function, which is the one on your left, the higher number is better, and you can see as an aggregate, there's a slight improvement over time. With pain interference, you want a lower number, less pain interference, and you can see as a whole, the aggregate data is going downward. Just another way to look at our aggregate data, PROMIS-T scores, physical function, you can see the little dots on top, the gray dot is at baseline. As a whole, for the low back pain, it was in the moderate category for physical function, and at 12 months, improved to mild. Likewise, pain interference started in moderate, and at 12 months, improved to mild, as a aggregate data of all the patients that have their data has been uploaded into the registry. We can look, the way the registry is set up, we can slice and dice anything we want to look at, basically. So you have this whole data set that's been entered, and we can start using filters to see what we want to look at. This one happens to be for just comparing different races. How does their low back pain, patients with low back pain, how do they compare different races overall in the congregate data? And what's really interesting, you can see on the right, pain intensity is improved much more in the African American population than in the Caucasian on the left. Just different ways we can slice and dice it. Oh, that helps. Beth, it's really little print. Okay. So we can look at it also by other categories. This one happens to be, we looked at low back pain, return to work is on the left, and patients, excuse me, patients who did not return to work are on the left, patients who did return to work on the right. And sort of intuitively, you'd hope the ones that returned to work actually did better, have less pain. And as a whole, the aggregate data is showing that, for example, the pain intensity, the lower right graph is better as time went on on the patients that returned to work. But we can look at it, breaking it down, categories like this, we can look at it with intervention. You know, patients who were age 50 to 60 who received an epidural, how did they compare to ones that received therapy? And starting to look at the big data to be able to compare. We also have the ability, if individual organizations that are participating wanna use the data in their own way, they can do a data, basically a data dump, this is what the spreadsheet is, to then work the data the way they would want to. And with that little bit of an overview, I'm gonna turn it over to Dr. Karinopoulos. Did I put your? Okay, who will talk a little bit about lifespans. Thank you. So I'm gonna try to use this thingy. So welcome and thank you for coming. Alexios Karinopoulos, I'm the Chief Physiatrist at the Brown Medical System. I'm a pain subspecialist, so I don't generally see the stroke patients in the acute, but I see a fair amount of them should they develop pain. So I'm a professor of neurology. Okay, so I'm going to come over here because I can't see either. So this is a little bit of our demographic information that we just discussed. Admittedly, our patient population may be a little skewed. You know, it's Newport, Rhode Island, and it's a little bit, we have a lot of Mr. and Mrs. Thurston Howell III, so this may not be as generalizable to the overall community. But, you know, again, it helps us to look at where our patients are coming from and how we can, am I doing something wrong? No. Oh, you're the, okay, thank you. And you know, how we can relate our data, if and when we pool it to get more data from other data sets. So again, the PROMIS scorecard, this is great from our perspective because it shows us what we are doing internally and how we can potentially intervene and or predict how our continuum of care goes. So you can look at our physical function. The black line is the registry average, so we're a little bit higher than average. Our anxiety levels, depression are consistent with average, by and large. And this is at the 90-day check-in. Unfortunately, our pain interference is, actually, fortunately, our pain interference is lower, as is our pain intensity. So, you know, I think we have great scores in terms of other metrics that are being used to judge us, but this is an internal validator about what we're doing and the fact that our paradigm seems to work. So we do not have a lot of data, and this is also very important. We were an early adopter, but there were a lot of technical difficulties, mostly in integrating with our EMR. Most of that was our own issue with, you know, sort of integrating or working with the AAPM&R. So this is, you know, I'll talk about this in a minute, but this is something to consider if you're thinking of adopting this program, is all the logistics that are required to implement it. So you know, our data is not great, but it gives us an opportunity to figure out how we can get more data and more discrete data. One of the concerns that we have is on the... So this is just one example of a data set. So a 76-year-old female, cerebral infarction, discharged from our IRF in May, history of hypertension, being treated with a number of medications. With these PROMIS scores, we were able to track this patient at 30, 90 days. Although her sleep got worse, she was less active from 30 to 90 days. So there are implications and inferences that can be drawn from that with regard to physical function and sleep. We also see that her pain went down in that same time frame. So should the data points have been different, this could have been a great opportunity for us to say, at 30 days, where is the pain coming from? Is this some neuropathic, nociceptive, nociplastic pain where we can intervene, whether it be through medications or rehabilitation or even psychological management? Would there be an opportunity to look at comorbidities that would have affected sleep? Is there some sleep apnea that wasn't diagnosed? Pain interference and pain intensity, that goes well with our wheelhouse of rehabilitation and other medications that we can use to intervene sooner than later so you prevent some of the chronicity. All right, so what were we looking for when we decided to? you want to contribute, and how we are using the stroke data to change our practice and also to influence other practices around the country. Identify and assess risks in care of the cohort. Again, why not know, why wait until three months out or six months out and the patient has a lower level of function? Why not be able to intervene or be aware of the information sooner? Again, objective data, using treatment analysis to predict. This is one of the things I'm most excited about. Use of AI, machine learning, using classifiers to really dive deep into some of the data sets, not necessarily from just our data because we don't have enough data. AI requires a lot of data. But can we pool our data with other registries or other research projects, whether they be from the neurology or the neurosurgery literature or what have you, so that we can get a better idea of how we fare and how we can contribute to this overall trend. Provide an identified source population for potential observation study. Let's do a prospective study and a multi-site effort with the academy, with some of our partners, to really collaborate and make sure we keep the momentum up from what we're doing all together. Okay, next. Again, these are some of the positive impacts and we're happy. Of course, the main focus is getting access to latest clinical trends, so we can compare ourselves to what Shirley Ryan is doing, what the others are doing, just to make sure that the PMNR brand is fairly consistent. Potential opportunities, changes with registry and EMR integration, I think that was one of our biggest hurdles. I'm not sure how the others had experienced, but if we are able to identify the potential hiccups before we integrate with new sites, I think that will make things go more smoothly. And because we're dealing with patient data, access to the actual sites requires multi-factor authentication, which people have found challenging, you know, you basically have a Microsoft authenticator app and everybody forgets their password or what have you, how to use it. So I think it's really important when you have a registry and you have enough people who are able to view the data that you have a collective wisdom about where you're going and how you compare to others. But of course, that requires access. Current benchmarking should be expanded. So before in some of the data, we did look at how we compared to others, but I think there's greater opportunity to build on that, and that's what's going to help the profession, the specialty move forward. And this is, of course, early data, phase one, focuses on the inpatient stay itself and outcomes post-discharge. These are a way to use this data more longitudinally in the post-acute care continuum, you know, in our outpatient clinic, you know, at one year, at six months, one year, et cetera, so that we can not only do a better job, but predict potential changes that could happen. I think that's it. Oh, so again, opportunities using this data to leverage with payers. Hopefully it's good data, but we can not only share what we are doing and how our good outcomes, but also show the internal structure. referrals from neurology or neurosurgery or internal medicine, you can show them what you're doing and how we can affect your patients. Identifying patient priorities. So ideally, any type of data registry gives you an opportunity to show patients the data. Patients want to know how they're doing, and I think it's important because it's a validator or it's a reminder about taking their medications or doing their rehab after they get discharged and that this is a sort of a continuum of care that needs to be provided often in these post-stroke patients. Healthcare trends evolve every day. How do we standardize our data and elements and definitions to meet these changes? And eventually, as I mentioned before, how can we use some AI technology, computational modeling to predict outcomes, which is essential if we're going to survive and we're going to face some of the challenges with reimbursement in our community. And I'm handing it over to Aaron. I guess I'll stand over here because he's set the precedent for this. So my name's Aaron, and I'm at Vanderbilt Medical Center. And so if you could forward to the next slide. So I'm one of the interventional spine providers, and there were three of us there that were part of this data collection, and also one pain physician also contributed some of the data. So I'm not going to necessarily belabor. I mean, every location has different demographics. We're in Nashville, but you get this and you can say, okay, I think as a clinician who sees this, I say, okay, so here's just some basic demographic data of what we're typically seeing age groups. So I think that can just be helpful, just have an overall snapshot. You can see in terms of the number of patients we have so far, there's 3,300 or so patients we have in our registry. Sorry, I hate for you to call you out to move the next slide, but I appreciate it. So here is our PROMIS scorecard. So to be very frank, I really wasn't involved with the registry when this started, which I think is actually a good thing because it's almost running itself without feeling overburdened in my clinical practice. I think one of the biggest barriers when something like this comes out, you think, oh, it's going to disrupt my clinic flow. I don't know if I have time to address this, but Dr. Kennedy, our chair, was very passionate about being a part of the registry, which we all agreed and said, okay, great. But I think just now being more involved and obviously being asked to speak on it, I was like, okay, I got to catch up on this information. But this stuff is interesting when I look at it. So looking at all the different measures, so the blue line is for Vanderbilt itself, and then we just have their registry average, which, again, there's not many sites that are part of this registry right now. But again, just to have an idea of what we're tracking here. So can you go to the next slide here? I think this is also interesting just for myself to look at sort of seeing how I'm doing with my peers. And again, not necessarily, we get judged on a lot of things. Like one of them is like prescany, right? How are we doing with patients? We get that data. I think also just seeing a little bit more granular view of how our patients are doing can be helpful. You know, for me, I always say we're continually learning, and what I'm teaching my residents may change in 10 years, and we're doing it all wrong. But I think it just helps us to continue to get better. For me, I'm always continuing to learn from my peers and seeing how I can be a better clinician. And, you know, I think just to have an idea of where I stand, and eventually as this data gets much, dataset hopefully gets much bigger, can also continue to see where am I trending here. And so that's been helpful just to take a look at. Next slide. Again, just also helpful to have this information and this graphic of, you know, what kind of diagnosis am I giving, what kind of interventions, or CPT codes and E&M codes. So I can see that lumbar radiculopathy has been, you know, 42% of the diagnosis I've given to the patients that are part of this registry. So just having that breakdown is helpful. Having BMI, age distribution, just having this just 1,000-foot overview, I think, was helpful to look at. Next slide. So I think this sort of feeds into, you know, a wake-up call for me, but, you know, I have done really nothing to really have my patients be engaged with this up to now in that I'm not going to them and saying, you know, I have patients who come to me and say, I got this email from AAPMNR. What is, you know, should I do this? Is this spam? And I'm like, oh, my gosh, it's so, we're collecting data. I forgot. Yes, you should do it. So, you know, that's, you know, again, I think having physician engagement is going to be very helpful, and I think that I'm an example of someone who hasn't been engaged, but now I realize, like, wow, look at all this data, and we're contributing to this, and we need to be at the forefront of this. And I can still set it and forget it, or I can say, look, just a quick blurb about collecting. And I think once patients realize it's coming from you, they're obviously going to be more willing to fill this out. And I'm curious to see, I wonder how much of this goes into the spam of their email filters. I didn't think about that until coming up here, but, you know, they may be missing also a lot of this, because also they don't know what AAPMNR is. I don't know what the email title is even, to be honest, and they may just be like, what is this? Because we just get spammed so much. So I think just even just a blurb about it with patients can make a big difference. And I do like it. At our site, it is email-based, because, you know, I think it's just much easier than trying to collect data at the site or, you know, having someone call. It's just, it's continuing collecting it, so I really like that. Next slide. So this is a drill down, and I didn't even know you could do this until I was told I was going to come up and talk. So I like this. And so this is the example we found. So this was a 62-year-old female, had right buttock pain. We sent for physical therapy. She was already on Tylenol arthritis, and she was alternating that with an anti-inflammatory. She did physical therapy, tried that, was very faithful to that, came back. Five months later, we decided to proceed with an injection. We did a right sacroiliac joint injection. And then you can see here, you know, sure, like, it's great. Her pain went down at the six-month mark, and it seemed like it was already trending downwards as she was filling this out. So I think the other things that are helpful are, you know, what's she doing outside of that? I mean, we talk about function, being part of society, getting back into activity. So that was helpful to see because those aren't, you know, we ask about those things briefly, but I don't think I document that very well in my note, to be honest. Or just, so having this information, I think, is helpful as we think about going to payers, going to other departments. You know, our institution has a spine surgical registry, and I always know that they use that to really market themselves in a competitive environment. Like Nashville, there's lots of surgeons. So I think having this kind of data can also help us as a field to show that we can be leaders in non-surgical management of a lot of these conditions. So again, I think this was just helpful to have that graphic. So just a few things, what excites me, I think, again, my hope is that we get more institutions to be a part of this registry, and not even just institutions, but practices in general. I think the more data, the better. Again, we do have this sort of homogenous data because this is, a lot of it's from our site. So having it just around the country can help. Again, I really do like how much involvement I can decide to put in. I would really just encourage everyone, don't feel like this is a big burden for you to take on, and you're going to have to continue that, because I'm a living example of someone who's done very minimal, but we're still collecting data. And so having some baseline demographics, and most importantly, collecting patient report outcomes to our patient population, how we practice as spine physiatrists, I think will be very pertinent as we move forward, and thinking about all the different research things we could look at within this of how we practice would be very helpful having our own data set. So next slide. I think one of the things that I was just thinking about, we also get a lot of direct referrals from surgical colleagues for injections, and so I think that would be interesting to see some of that data as well. Again, just having that ease of connection, reducing the amount of steps we have to access this data will always be helpful. Again, getting the Authenticator app downloaded, and doing that, I understand it's patient data, so we have to be secure, but also just once you get into it, it's much easier. It's like riding a bicycle. The more you do it, it's easy to access. And then I think it will be very interesting at some point when we can just look at the data and say, oh, this is when an intervention was done, or this is when this medication was prescribed. Then you can really see that granular data really quickly, and I think that will make a really big impact on how we practice when we can just quickly see that information. So we can even see when we decided to make the decision to do an injection on a patient. Is that usually after two visits? Is that after six visits? I mean, just having that data, I think, would be very interesting to look at. I think that is my last slide, so I'm going to hand it over to Dr. Huang. So I'm going to start off with, so we're the last site for the presentation. So we actually collect data for both low back pain and ischemic stroke, and we're going to start with our low back pain cohort first. And so this is our demographic data. It's kind of interesting, because I was looking at this, and actually, it's very similar to Erin Yang's site. If you kind of look at sort of those demographics, we have predominantly Caucasian, or white, and then as far as the black, it's 13%, and then just the other races are 18. And then if you can kind of see, it's actually interesting, also at Vanderbilt's site, our larger percentage of female versus male is the patient population, which I thought was kind of interesting. It's actually talking about Dr. Rowe, our musculoskeletal chief, and we have more female physiatrists, and so who knows? Maybe it's that, plus our women's health program that we've integrated that's probably causing some of this sort of predominance of female versus male in our site here. So it's kind of interesting just looking at this kind of data, and obviously, in terms of age, you know, about half being 65 and older for the low back pain patients. Next slide. So it's kind of interesting to be able to see all this. And as we go through all these slides, what's interesting is that this is all straight up regular dashboards available in the database. So there's no doctoring that's being done in terms of this. You pull up an individual patient for these drill downs, those are individual patient data. And you can get that from the registry now. So it's not like this is made up stuff or we've kind of doctored things. So this is all things you can get now. So and this is just our low back pain scoreboard you can kind of see here. Blue is our site, black is registry average. You can kind of see we do have a little bit of improvement in physical function, anxiety, depression, kind of probably kind of filter about the same. A little bit of reduction in pain and interference and pain intensity across the sites for our low back pain cohort. Next slide. And this is sort of our aggregate data collection. We try and collect low back pain patient reported outcomes, the PROMIS tool, plus the additional questions in the office. We've had a lot of struggles with our front desk staff processes. So we've gone through many times where we had actually had the pause in terms of actually delivering the baseline questionnaires in clinic due to some staffing issues. So we had a lot of challenges really getting that up and running due to some staffing issues on our end to try and ideally get these baselines done in the office. Otherwise they're going to get sent an email to have the baselines collected. So you can see our rates are lower here. Also we're having some issues with some patients that actually aren't probably they're getting coded as low back pain as a comorbidity but not the primary. So we're trying to sort of weed some of those patients out of the database. So we're kind of reconfiguring some of our we're going to reconfigure some of our data exports so that those patients aren't actually included. So we've actually found that in some of our database. So we're trying to get some of that data cleaned up. And so this is why the percentages kind of look lower. But you can kind of see we also have if you think about it they're pretty low in terms of just some of the response rates. So we are also kind of in that low response rate overall for the actual patient reported outcomes being included in the registry. Next slide. As far as just this is a drill down of a specific patient again this is this is data you can pull up right away on a patient if you look at a patient at the individual level. So this gentleman we did a little bit of a chart chart review in terms of just to kind of get a sense of what was the initial reason why they saw the clinician and sort of had this recurrent low back pain episode. So he had an exacerbation of some low back pain, L4 or 5 radiculitis. He was actually given a round of mechanical diagnosis and treatment therapy, use of NSAIDs and was seen and followed up two months later and at that point reported an 80% improvement and they advised if he started if he had no further improvement that they could do an epidural steroid injection. But he actually did not pursue that, actually seemed to have improvement in just physical functioning, anxiety and fatigue, slight worsening in anxiety and depression. Next slide. But you can kind of see the pain scores kind of went down. So pain interference, pain intensity went down. So I think that's why the patient didn't follow up to actually get the epidural injection. The MDT treatment in just his home exercise program alone was able to warrant that he didn't need to go back and actually see the physiatrist in clinic again for his complaint. So it actually kind of shows you this story on the back end that, yeah, they actually got better just with therapy, did not require an injection. At that point, because pain interference and pain intensity was reduced, they didn't come back and seek further treatment. So it's kind of an interesting sort of individual snapshot of a specific patient. Next slide. So that's more of the low back pain data, and now we're going to go into ischemic stroke. And this is kind of interesting and more of a split here. You can kind of see as far as the race, we see almost, you know, African American or black being 42 percent, white being 39. You kind of see the male-female distribution is actually evened out in stroke. So we're seeing a very distinct kind of different population in the stroke world over here. And then obviously it makes sense, 56 percent or 65 in order for stroke. So we're seeing an older population, which does make sense for the stroke individuals. Next slide. And as far as the PROMIS scorecard, you can see it's kind of a nice upshot with their physical functioning, anxiety and depression kind of filtering down, although maybe a little bit of a bounce back at the six-month mark. Pain interference, pain intensity, there's a little bit of oscillation overall does seem to go down over time. Next slide. And so our response rates here, we do try and collect them baseline. We actually do have a pretty good collection rate. This is actually a little bit misrepresentative, because we actually have higher collection rates as far as the appropriate stroke patients. Again, we also have this issue where our coders are going in and adding every code under the sun, and so we have to go back and we're going to need to go back and sort of limit to the first five diagnoses, because once our coders get to it, they're coding everything, you know, on the inpatient side. So we've got to get some of that kind of filtered out. But as far as this, in terms of just our follow-ups, you know, unfortunately we're not all of them are putting in e-mail addresses at the end. That's our primary means of getting a PRO on the follow-ups, so it's resulting in lower percentage success rate for getting the PRO data in the follow-up for a lot of our patients. And again, as what was brought up, the observation is actually filling those out. Is it a caregiver? Is that the actual patient themselves going to fill out the survey? So next slide. So we, again, sort of a drill down of an individual patient, a 72-year-old gentleman who had dysarthria right side of weakness, had his inpatient rehabilitation stay, discharged home and attended sort of a day rehabilitation sort of program. What was interesting, his physical functioning report is kind of staying about the same. Anxiety may be a little worse, sleep disturbance worse. Pain interference, pain intensity did kind of go up. It's interesting, the actual visit in the outpatient setting, he actually reported that he's actually improved physical functioning because he's walking up to a mile and a half a day, which is pretty darn good. But he really didn't complain of pain during the office visit, so I thought that was fascinating. So a little bit more pain interference, pain intensity reported on the survey, but he didn't actually report that during the physician visit. So I thought that was interesting. It probably wasn't. And then I think the relative sort of physical functioning, getting back in the community, like he's feeling like it's still the same, but yet he's walking a mile and a half a day. So it's kind of interesting, sort of that perspective in terms of just seeing the patient's perspective of physical functioning maybe more expanded as they get home. They feel like, well, still not doing quite all the things, but certainly better than what they had done during their inpatient stay. So it's just kind of interesting that we're seeing those. And maybe we saw a little bit of pain interference, pain intensity go up, and some other instances, are they having central post-stroke pain issue, long-term sequelae musculoskeletal, maybe a hemiplegic shoulder or whatnot, or altered gait mechanics. Could those things be playing a factor in sort of increased pain complaints? And maybe that's something where if you're seeing it on your survey, maybe it's like, hey, it looks like you're having more pain on your survey. Where is it? Maybe that's an opportunity for your clinician to be able to drill down and just talk to the patient, because they reported on the survey, but they don't tell you in clinic. So kind of like an interesting thing that this is an additional adjunct where a PRO can provide immense value for targeted questioning and interventions, if appropriate, in these populations. Next slide. So what are the benefits? I mean, obviously, we want to collect patient-reported outcomes data. I mean, most of our sites, we do it anyway, right? We're doing some sort of PRO data, but how can we get it more in a standardized way and be able to monitor this data accumulation? And we like that we have these dashboards that currently exist to explore real-time demographics. What's going on with patients? What's going on with response rates? And using these current dashboards to drive internal data collection processes. Are we doing this? Are we collecting the right data? And how can we modify our processes to kind of enhance the data collection, especially the PRO side? So next slide. So I think, obviously, some of the lessons learned and next steps, PRO collection is pretty challenging. On the inpatient side, it's actually been easier, because we get our nurses, our inpatient nursing staff, to actually help collect that baseline PRO data for ischemic stroke. In the clinic, it's been a little more challenging. We're trying to use iPads and having that as part of the check-in process, doing the iPads, sending them back. And what we wanted to do is we, of course, tried to make it more complicated with us, because we wanted to give clinicians real-time data. So we are trying to actually have them finish their iPad surveys, get those surveys printed so the clinicians can see them. And I think that's been our big challenge, is just that really messes with our clinic processes, you know, unlike Vandy, where they're just going to just get the email survey post-talk, post-visit. But we're trying to get that in the clinician's hands. This is why it's been so challenging for us. It's just, you know, our clinic resources and just trying not to slow down clinics. So this has been our big issue, is trying to how do we do this in the best, most efficient way without burdening the nursing staff, check-in staff so much that it becomes impractical. And as you can see, it's been challenging. So trying to even get that process in place. And then just needing to educate the patients. We actually do give the patients a handout. The AAPMNR does have a standard handout for educational materials that talks about, okay, this is what the survey email is going to look like. You know, it's going to come from AAPMNR. This is what it's going to look like. And therefore, then you give the patients say, ah, okay, I'm going to get the survey. Because, you know, everybody gets the surveys, right? Everywhere you go, McDonald's, Chipotle, you know, your flight's here. You're going to get a survey for feedback, right? How many of those things are you going to get? And how many of the patients are really going to fill those out? So especially if they come from, you know, the organization, right? So I think that's obviously the challenge. Now as far as implementation, you know, what we've done is I've been holding meetings, regular meetings with both teams, ischemic stroke and low back pain, to kind of review what are the challenges you have? How is your data collection rates? I will literally go onto the registry website and go, here's our current collection rates here. You know, we're seeing breakdowns. We can see the accumulated number of surveys per month. So I can say, hey, we only had this many surveys this month, so where are we breaking down in the process? And it kind of makes sense, you know, when we had some clinic challenges, we were not doing a whole lot of survey accumulation. So, or we found some things, hey, we're seeing to have no surveys for the month of this month. What happened? And it's going back to the registry. Do we make sure that we, are they getting matched properly in the back end? So there is some, it's good to be able to meet regularly with your teams to kind of see how we're doing. But, you know, I think that, you know, as far as next steps, you can see on this slide, two main issues, right? One, PRO collection rates, bad. We need to get those up, right? We have to get those up. And how do we get that more efficient? So they can't, you know, I'm sorry, we're going to probably steal some of Alan's thunder. So I'm going to let him, let him talk. But, but, you know, one of the big things we are trying to work towards getting a better mechanism from the registry itself to send the surveys out other than just email, because this is what's killing people is just email. A lot of people don't communicate that way, especially low back pain, don't even have email, right? You know, so we need to figure out the other way to getting it through text messaging or other ways to get to these patients more easily to do that, right? And then, and then the second is really training the clinicians, educating them to really get by in it. And then the third piece is actually, so I said, I lied, I said two, but it's going to be three. But in terms, we need to get that data back in as, as Aaron Yang was saying, you know, it's like, it's too clunky to log into the registry and do this. It needs to be integrated back into the EHR somehow. So we are, that's our long-term goal that we got to get to that point, because otherwise, if you don't get actionable data back to clinicians, it's hard to get their buy-in, right? Hard to get patient buy-in, right? So, so these are the goals moving forward is trying to get that, you know, get that data back in more easily for clinicians to see it other than just logging in the database. Much like, you know, everybody uses PDMP data. We all love that, right? So for uropoids, you know, trying to get that. So can we get that sort of query into the, into the PDMP, bring it back face up in the, in the EHR, which most people have that right now in their, their electronic health records. Can we get that same sort of mechanism back for, for the registry? Can we go to that pass? That's clearly one of the things we have to try and get to, to get this data easier for you to see an action on it. So it's actionable for you at the patient level, both at an individual patient level, plus at the level of aggregate patient data. So anyway, next up. Okay, cool. I think one of the more fascinating things as we've developed the registry is that no two organizations that participate do exactly the same way. It's not a cookie cutter platform. It really is sort of customized to what's needed, but the data still all pours in. So how to participate. Let's, next slide. If you are interested, and we hope everybody's interested because the more participants we get, the more robust the data is going to be, the more we can utilize that data. You know, we don't just want academic centers participating. We want the, the private practice guys, we want everybody participating so we can see are there trends, are there differences? If you go to an academic center versus a solo practitioner, the more data we get, the larger, the more useful it's going to be. So if you are interested in participating, you can schedule a meeting with the AAPMNR staff. You can contact them at registry at AAPMNR.org or just meet with anybody in the background. Kavitha and Beth are back there who probably know more about a registry than most humans. The problems, once you decide you want to participate, there's actually some challenges. You have contracts, business associate agreement, data use agreement, registry participant agreement, and what's really important is you need to have IT resources at your organization that can help lead the integration efforts. You also need clinical resource champions. You need somebody who really buys into this and helps push and promote moving the registry forward. Next slide. Oh, we are doing a demo tomorrow morning. Do we have the time? Pod three. More information to come. AAPMNR has significant efforts in quality and research initiatives. The registry is just part of that. You can go to the next slide. What's going to happen in 2024? One of the great things about the registry, it is not static. It keeps evolving, keeps changing, and a lot of the changes are all coming forward by the different users, helping sort of what's working, what isn't, what more do we want out of the registry. So right now for 2024, we're looking at where are the new areas that we want to look at collecting data. Perhaps spasticity, cancer rehab, pediatric rehab. Those are all topics we're sort of looking at what will be the next target area. We're looking at patient reported outcome, a web entry tool to make it easier. Dr. Wong kind of alluded to that a little bit about that. If we could build a web entry module, it would allow clinicians to enter a few data points in order to launch the probe for their patients. Next slide. Looking at some building some new modules, we kind of get a lot of data on medication. And so it comes in in some odd formats, NDC and RxNorm formats. So we're looking at mapping that will allow us to easily map the meds to the patient. Looking at building new reports that will allow us to combine the medications and procedure data a little bit better so we can look at the outcomes a little bit more in those areas. Next slide. One of the things that yesterday in the business meeting, the academy board approved a quality and research committee strategic coordinating committee that's going to be led by the one and only Dr. James Sliwa. And the committee will focus on creating an AAPM&R overall quality and research strategy, work on developing a strategic approach for collecting and analyze patient reported outcome data, and hopefully we'll sort of define the standard physiatric standards of care throughout the continuum, giving us benchmarks for physiatry quality of care, enhancing patient safety, facilitating professional development and ongoing education within physiatry, and instilling confidence in patients, providers across the health care system. Really developing those guidance statements relevant to our specialty, which the registry will just be a part of this overall strategy on quality and research. Hopefully the committee will help evaluate the current evidence base for physiatry, advancing awareness, appreciation, and value among all our AAPM&R stakeholders. Next slide. So I will open it up to questions. I'll take the last one first. That's the goal. That's really kind of what we're looking towards in real time to give it to the clinicians. There were so many questions. Let's take them one at a time. Yeah, the first question was a technical question regarding interfacing. And as far as smart on fire or fired, not there yet. So right now, we're doing more sort of data. Some of them are data pushes. Some of them are data pulls in terms of getting them into the registry. Based on the site and the platform, so in terms of the EHR. But fire is the dream to kind of get that back in, right? To kind of, you know, smart on fire, we've been using that. So that's kind of like the basis of a lot of these sort of queries back to PDMP. So can we do that as a future iteration in the registry? It's definitely something we want to explore. So you can do a data dump. You can do literally a data export to Excel or something else right now. So it's kind of manual, right? It's not automatic. But yes, you could at this day say, all right, everyone in the database right now, can I do a data dump? Yeah. You can do that in Excel spreadsheet. Super easy. You can just kind of export that way. But not exactly like a, all right, I'm going to do it within the EHR kind of thing. Like a real-time data dump? Well, you can. Yeah, you can. Well, but yeah, but it's like an on-demand, right? It's like a download of a file, right? It's like a CSV file or something like that, I think, is the way they do it, right? So you can do it that way, but it's a manual request. Oh, patient or, excuse me, therapy or other reported. Yeah, so we haven't quite gone there yet. I mean, obviously, those weren't in our initial data dictionary. So right now, you saw the demographic. There's a more deeper dive in terms of the data elements. And some of those things aren't in there yet, honestly. And that was just because we just wanted to get started with initial demographics and PRO focus for stroke, starting with ischemic stroke and obviously low back pain. So those are kind of your primary things. Do we oh I knew it's six questions sign on there's a lot but we're trying we I think we had three or four So, yeah, as you saw from, yeah, so we have, not quite yet, but basically from the standpoint of meds, that's kind of our, we're trying to tackle meds next in terms of just, you know, what we can do with the meds data to present it in a useful fashion that could be analyzed in a dashboard format. I think if you're trying to do like a more sophisticated analysis, just doing that sort of data file export and then doing it, you know, within other statistical software or other analyses, they certainly have that capability. It's just not intrinsically built into the registry right now as of yet. So we've obviously got some of the more basic elements. You'll be able to do filtering and comparison on. We just don't have that as of yet. I do not foresee our register, we have a steering committee that governs all of our work. I do not foresee the steering committee opening up the governance model for individuals to access the aggregate data unless they are participating in the registry. Now years from now that may change, but that is what at least where we're starting. Aggregate data can be used by sites that are participating in the registry. I do not believe we will open it up to others in the short term until we get to a place where we might need to have those conversations. Participating institutions have access to all the aggregate data already. So everything you see here you have access. Yeah, we're not quite there yet, but that's exactly where we need to go, and if institutions that are in it start asking us those questions, we can start getting there, but maybe we can talk after this and get your name and have more discussions. It'd be great. So we do have, as Kavitha said, a steering committee made up of representatives from each organization, and that would be something that they would drive if there was interest in going down that route. Great question. We've had these conversations, and the real rate-limiting step is EMRs. A lot of the SNFs don't have the sophisticated EMR to be able to pull the data. You know, in an ideal world, that's exactly where we want to be able to show, you know, does an IRF do better than a SNF, or, you know, God forbid it shows the other way around, you know. But data's data, and if it does, you know, then we have to move forward. But that is the goal, but we're limited by the EMRs. the for inpatient definitely I was going to say for the low back for outpatient it's a different animal and that's why we've kind of started and kind of keeping a little separate. I think Dr. Cary Annapolis hit the nail on the head and saying everybody in this room has great ideas and I think purchasing other data sets getting other data sets to integrate with our data is where we need to go. We have started at ground zero and saying we need to get people involved PM&R as a whole is not standardizing data collection is not collecting patient reported outcomes data. Do we need to get there. We would love to. I wish today we said we had a representative sample of the earth population stroke earth population so that we could get there but that is where absolutely where we need to go. Another challenge with with the sniff area is a lot of physiatrists or consultants in those spaces so the connection is really with the corporate and we have started having some of those discussions but it's a little different in terms of the control and power that the physiatrist has in those settings that also shifts the paradigm in how we work with them. Thank you for watching! But that's the whole purpose of if we can collect data. I mean obviously there's limitations and sniffs, but if we could collect collect the data and can show that You know how much therapy intervention led towards a better outcome and take that data the insurance company. It's Wow Jim you you go comment And I would say get involved, you know the Yeah whether whether it's whether it's Whether it's serving on a committee or just networking at this meeting with people who are in leadership roles It's so important to communicate that so that we can then work on issues like that Thanks, everyone. I appreciate your attention, and I can't wait till next year when we show you even more. And we really have to thank the Academy staff because they do a phenomenal job helping us with this. I would talk to Dr. Wong, Mark Wong, he's going to take over this meeting. Thank you.

AAPMNR Registry

patient care

real-world care

low back pain

ischemic stroke

rehabilitation practices

clinical data

patient-reported data

outcome comparison

electronic health records integration