Good afternoon and welcome to the last session of the best of the best of research projects. We're doing general Rehabilitation today in our session. You're the best of the best Abstracts that were submitted you're all eligible for the presidential citation Which is a prize of free tuition for next year's meeting in beautiful sunny, San Diego So do good or well So our first presenter today will be Hajan Lee and She'll talk about modulatory effect of motor imagery and cancer survivors Doc come on up. Let me get rid of this screen and we could find your stuff And you're probably better at this Thank you Hello, hi So, my name is Hwajin Lee. I'm one of the residents from the Rutgers, New Jersey Medical School in New Jersey and My project is on the modulatory effect of motor imagery in cancer survivors I would like to thank two of my research mentors. Dr. Sally and dr. Yoi And they're both at the cancer foundation So just a little bit of introduction with earlier diagnosis of cancer and Development of new treatments the number of cancer survivors has been increasing With many living with disabilities such as fatigue and weakness diminishing their quality of life So high-intensity arm strength training is essential to gain on muscle strength Especially for cancer survivors as well however people with cancer Often have persistent fatigue and weakness to even perform basic exercises or functional activities Let alone perform conventional high-intensity muscle training So it's very important to develop therapies that suitable for cancer survivors to maintain improve their daily function and quality of life Evidence in recent years has shown that Training with high effort or intended muscle contraction With no or little physical exercise can significantly strengthen a muscle by increasing brain to muscle command Which helps to improve motor unit recruitment and activation level Motor imagery training can be a safe cost-effective and perhaps fatigue immune method for cancer survivors to improve strength So the objective of this study was to test a hypothesis that motor imagery of high effort With actual minimal strength exercise can stimulate the same area of the brain as high effort exercise with that without imagery in cancer patients So this research project was part of a larger study a single-blinded longitudinal study Setting was the outpatient rehabilitation center and major cancer center participants included total of 23 breast cancer survivors and eight non cancer control subjects from data collected over a five-year period from year 2015 to 2019 and Again data in this study was part of a larger clinical trial data Where cancer participants were randomized into one of three training groups? So we had the high effort motor imagery training combined with physical training group Low effort physical training and control group who received no training and healthy control group participants were enrolled in only data collection sessions and For this particular study we focused on the fMRI based measure of brain activation That was acquired during two conditions motor imagery task where participants imagined doing a maximum voluntary contraction task and A motor task where participants physically perform the task itself And behavioral data was acquired during the experiment using an MRI for sensor Not sure if I can Scroll down here Okay Yeah, I'm not sure if I can make it a little smaller there we go um So the behavioral data again was acquired using the experiment using the MRI compatible for sensor the for sensor data was post process to obtain the timing events of each movement trial and ANOVA was used to compare activation between the two groups and within cancer and healthy control groups and fMRI data was processed Using a software called the FSL Toolbox using general linear models which were created to examine the relationship between each of the experiment tasks and the brain activation So the result for this study showed that for both the cancer and healthy group There was overlap of brain activation between the high effort and motor imagery task in the pre motor motor cerebellum and Supplementary motor regions so you can see on the image there I'll kind of try to focus in on the image. So it's better to see Let me see if I can adjust this a little bit there we go So the red areas that you see on the image there is the area of the brain activation while the patient was performing the task So you can see there's some overlap of the red area or the brain activated area when you compare the high effort physical task and the motor imagery task for both the cancer group and the healthy group and Just on a side note on the cancer group. You see more activated areas or more areas. That's colored red And most likely that's because the cancer group they have to put in more Effort or more use of their brain when they're performing the task compared to the healthy group So again to go back to our results here When we look at the between group comparison that did not show statistical difference Either due to low sample size, especially in the healthy group or having the lowest statistical power Or there is no difference in imagery and physical task effect between the groups and this would require further study using a larger sample size also, the F test showed difference between the high effort physical task and the motor imagery task However, the limitation in the F test is that it does not explain the directionality of the difference in Other words, it does not indicate whether the task induced produced a higher brain activity or lower It just shows that there was a difference. So there's the limitation there in post training of combined motor imagery and physical training the change in activation was within the cerebellum and regions in the In the attention network, but this change was not significant with multiple corrections And this is again likely due to low sample size So unfortunately for this research project The biggest limitation was the low sample size But this is an ongoing the larger study that is it's continuously ongoing So we are continuously adding on more to participants. So hopefully in the future We'll have a larger sample size to get a better result so in conclusion for both the high effort physical task and motor imagery task For cancer patients and for healthy controls the overlapping activity of brain on fMRI can be seen in multiple areas including the frontal lobe and the cerebellum and Hopefully with further study and more evidence We can use the motor imagery training as a potential neutral neuromodular approach in cancer survivors Thank you Thanks, dr. Lee. Nice job Any questions for dr. Lee Things were that great Description of your research projects. I am a cancer rehab physician. So the question I have for you is you know While this study didn't necessarily show a statistical difference between the two groups and your let's say your larger study does What do you think are the clinical implications for this and administering rehabilitation care for this population? Yeah, so I think that's definitely a good question in terms of clinical application a lot of the cancer patients they suffer from fatigue and and just generalized weakness and that can be a big barrier in terms of patients compliance or Adherence to their physical therapy. So I think incorporating An approach like this where Patients may experience less fatigue. They may be more prone to sticking with their physical therapy routine So I believe that that would be the clinical implication and hopefully with that we will still see some Increase in muscle strength While still limiting their arm fatigue level coming from the physical therapy participation Dr. Lee, it's a very interesting pilot study Out of the you had eight the end was eight. Yes, so Did they all have similar treatment for the breast cancer? You like where they're all like her to positive be our negative I mean, do they all have similar? Yeah, so We had non sort of healthy controls for this specific But you're just taking all breast cancer survivors as a lump or you're taking people that had the same treatment or similar treatments So we did not control for exactly the like the type of chemotherapy or a type of treatment that the patients received this was patients just sort of a overall patients who had a breast cancer diagnosis and had Chemotherapy or radiation, but there was no I Guess no control in terms of so it's an all-comers any difference in time out from treatment to the study Being done for them or is it again sort of an all-comers thing? It was sort of all comprehensive, unfortunately But it would be helpful to have more I think in the future when we have more sample size I think you definitely be beneficial to see Patients who are maybe like one year out of radiation or chemotherapy and have more control there in terms of patient participation I Hate to ask such a basic question without having you know, really reviewed your poster But could you describe in more basic terms? What a motor imagery task? I mean, are they just imagining moving or are they moving? Yeah going through range of motion or yes Yeah, so that's a good question so we had a group of patients Basically, it's based on hand grip which is connected to a monitor So for patients who had who had done an actual physical task, so that would be the high-effort physical task they were physically grabbed onto a bar which has a sensor on it attached to a monitor and When they had when they do a high-intensity Hand grip there. You can see a bar that goes up on a monitor so they can visualize it For patients who had a motor imagery task They would not physically actually hand do a hand grip on that bar But they would imagine themselves doing it while watching the bar going up So they still had that full engagement of their brain So they will be focusing on that but they will not be physically actively doing the exercise So, um, yeah, I guess I should have been pretty clear on the on the poster But um, that's sort of more of the methods of how that was done in this study It's hard when you're trying to fit Finite space and then you have to figure out what's the important part and what's not important to fit But it's definitely interesting start we're gonna go to from here So I'm hoping to do more Background research in terms of why more of the cerebellum area was activated Because I feel like that area is more what we focus on in terms of fine movements and coordination But how there has been prior papers that came out I believe there was a paper back in like 1990s that showed that maybe there was more attention that work involvement of the cerebellum, so I was hoping to do more background research on that and for Upcoming project for my senior project. I hope to focus more on that cerebellum It's one of the graduation I guess requirement for our residency program, but I was hoping to more of a focus Focus more on the cerebellum aspect of the brain Okay That's good. And just not all bench studies translate into clinical measures So this is one of them that may actually be pretty close to what we see clinically as compared to some of the things that we see with mRNA and whatnot and people talking about eating chocolate and living longer, so So it's interesting, yeah, I would encourage you to keep keep going in this direction, thank you. Oh another question fantastic How are you sure that the Subjects who are doing the imaging task weren't doing an isometric contraction Yeah, so Again, this was part of a much larger study So another portion of this study was to look at EMG results as well And they actually had I believe surface EMG leads on their hand and that would support that they the patients did not actually actively contract and they also had a Post activity survey as well To sort of see how if the patients felt comfortable Performing this if the patients felt like they were hand gripping on the bar while they were doing so there were some Subjective and objective portion of this But that's sort of outside the scope of this particular study, but that was the other section of this larger clinical trial study that we were doing at the Kessler Foundation Yeah, fine wire EMG hurts a lot Have the wires placed in you as a person who had them placed in there in fellowship Thanks, dr. Lee he did a good job. Thank you so much forward to next year. All right next is So Osa comparing patient reported and objective measures of physical function among cancer survivors a correlation analysis All right. So thanks for having me. I'm Sonal Oza. I'm currently a cancer rehab physician at Emory University, and this data was obtained during my time at the University of Utah. So this is a retrospective study comparing patient-reported and objective measures of function among cancer survivors, and we did a correlation analysis. So it's well-described that cancer survivors experience functional limitations anywhere from 20 to 90 to 100 percent, depending on the diagnosis. This is clinically significant because functional limitations, a decline in physical function is associated with decreased quality of life, but also with cancer mortality and morbidity across cancer diagnoses. And so there's been a great push to implement measures to capture physical function in the clinical setting, and this has been done through patient-reported outcomes and objective measures of physical function. However, there isn't a current standardized set of recommended measures to obtain in the cancer population, and additionally, the relationship between these measures is not well-known or described in the cancer population. As we know, physical function is a multidimensional construct, and are these measures capturing similar or distinct aspects of physical function? So we performed a retrospective study of a convenient sample of patients presenting to an outpatient cancer rehabilitation clinic, and the measures were obtained part of standard clinical practice. We administered the PROMIS Cancer Function Brief 3D Profile. It's a PROMIS short form that consists of three domains, physical function, fatigue, and social participation, and we also had been administering or having patients perform the hand-grip strength through a dynamometer, and then also having them do the 30-second sit-to-stand test. And so we analyzed this data, and our sample was made up of about 63% female, 37% male. Our total cohort was 226 participants. This was done in Utah, and so our racial demographics were, you know, 89% white or Caucasian, and the cancer type of the top four most common cancers in this sample were breast, gynonc, multiple myeloma, and prostate, and then 45% were other. And among this population, among 30 by 9% had received or were insured through Medicare, about half through commercial insurance, and 6% were insured through Medicaid. And about 26% had received systemic therapy in the past six months. Systemic therapy included chemotherapy, targeted therapy, and immunotherapy. And so looking at our results here, bringing your attention to the bottom left table first, and so looking at the PROMIS scores, they're measured in T-scores. So a T-score of 50 is the general population mean, so non-cancer, just general population. And so for PROMIS physical function in our sample, participants scored at 44. For PROMIS fatigue, they scored at 58, and then for PROMIS social participation, they scored 44. The average hand grip strength in our sample was 24 kilograms. And to give you reference, normative values for a male who's about 60 years old, which was the average age in our sample, is about 40 to 45. For female is about 20 to 25. And the 30 seconds to stand was 12. And then the range across men and women for, again, an age of 60 is about 12 to 17. So give some context for those findings. We then developed scatter plots and then performed a Pearson correlation between the subjective measures, so those three domains of the PROMIS short form, against the objective measures of physical function, so hand grip and 30 seconds sit to stand. And so we'll go through those findings next. We'll start with the most significant one. And so we identified that there was a moderate correlation between the 30-second sit to stand test and PROMIS physical function score, with an R of 0.57. And so because this was statistically significant and also a moderate correlation, we opted to perform a multivariate analysis. And so we accounted for age, gender, cancer diagnosis, and receipt of systemic therapy. And then this moderately strong correlation still held through for all cancer diagnoses except for multiple myeloma. And you can see the estimate confidence interval over there. Looking at the other measures now, so we'll start with the 30-second sit to stand. It had a mild correlation with social participation, but a negligible correlation with the PROMIS fatigue score. Now looking at right hand grip strength, whether it was physical function, social participation, or fatigue, we identify that there was a negligible correlation between the two measures. So now what do we make of these findings? In conclusion, through this retrospective study in our convenience sample, we identify that patient-reported physical function measured by the PROMIS moderately correlated with the 30-second sit to stand, but not with hand grip strength. Interestingly, hand grip strength did not correlate with any of our patient-reported domains. However, hand grip strength in itself has been identified as an independent predictor of outcomes after cancer treatment, cancer morbidity, and cancer mortality. Fatigue also did not strongly correlate with our objective measures of physical function. Fatigue is one of the most commonly reported symptoms among cancer survivors, one of the most commonly reported barriers to participating in physical activity and exercise. This may have been because of the questions that the PROMIS assesses are really looking at mild intensity daily activities. Can you get on and off a toilet? Can you dress yourself? Perhaps fatigue was not as limiting to carrying out that intensity of activity. It didn't look at ability to carry out moderate to vigorous. Also with fatigue, patients may be able to carry out a short-time task, but they really feel like it's more of keeping up with that endurance. They fatigue as the day goes on. So perhaps they were able to do that 30-second sit to stand, but if we had them do it repeatedly throughout the day, maybe we would have seen a correlation. And social participation mildly correlated with the 30-second sit to stand. The social participation questions of the PROMIS get at, do you feel you have difficulty or limited in doing work-related activities, social-related activities, like family-related activities? So maybe there is some relationship with mobility, hard to say from just these findings. In terms of future directions, the physical function phenotype has yet to be defined in cancer care, but I think it's important to know how measures relate to each other. So we're capturing the full, as best as we can capture that full construct of physical function. Additionally, this population that we looked at were individuals already receiving, you know, coming into rehabilitation care. Future work can assess these measures and the relationship across all comers, you know, with cancer, not just those that have already been identified to have rehabilitation needs. And then also, you know, what's the clinical relevance of this, and can we use this data to then inform interventions and then also predict cancer outcomes in a more granular manner? So thanks for listening. Thank you, Dr. O. It was a nice job. Any questions from the audience? Sure. Thanks so much. This was really interesting. I'm really interested in patient-reported outcome measures. Were any of them completed by proxies, like caregivers or anything like that? look at this specifically in our sample, but if the individual was non-English speaking, it was done with a translator. So the outcome measure itself was not translated into their native language. Gotcha. So the translator was translated. Okay, and one follow-up question. Did any of the participants have any degree of cognitive impairment, either from their cancer, treatments, complications, do you think that might have affected some of the correlation? It certainly can, and so there's certainly limitations to our study. We didn't look at cognitive function. We also didn't look at levels of anxiety, depression, resilience. That has been identified in other studies looking at patient-reported outcomes to impact how patients are scoring. That's a good point. So I don't know if I missed it early on, because it was, did we have an age range? Yes, I did not say that. So our age range was from 18 to 92 years old, but our average was around 60. Ooh, did we notice any differences within the age range? At least in our, when we looked at the correlation between 30-second sit-to-stand and promised physical function, in the multivariate, we did have age as a variable. And so that was not found to be a significant covariate, at least in terms of that moderate correlation. So covariate brings me to my next question, even though that's a big SAT word I couldn't use. So did they have comorbidities? Did we see that if somebody had diabetes, obesity, smokers, did we see any change or correlation? We did not because we were a little bit limited in terms of what we were able to readily extract with our EDW, but there was a study that came out, I think earlier this year, that did look at at least presence of musculoskeletal and neurologic comorbidities. So this was retrospective where they were handing them a sheet, they filled out a retrospective form? We, so all of the data was in our EMR, but just in terms of how, at the prior institution, how they're able to extract the data. So now you're in a much larger, diverse city. Much larger, much more diverse. Plans? Are you gonna do anything with this? Yeah, so I am, I'm at least collecting these measures currently in my clinical practice in Atlanta, and so hope for the future is, you know, if we analyze it in that cohort and replicate the design, do we see similar findings? And I think with this, at my current institution, we'll be able to extract more. Well, I think Winship should have like a big data draw already. Yeah, yes, I think we'll be able to extract more for them from the EMR, and there are several, yeah, several things that we're not, we were not able to analyze as covariates. Great. Any other questions for Dr. Olza? Thank you very much for coming and presenting. Sure, thank you. I appreciate it. Thank you. Thank you. Okay, our next presenter is Christina Sarmiento. Let's just make sure I got the first name right. She'll present on smoking and drinking behaviors among adults with childhood versus adult onset disability. Results from the 2020 National Health Interview Survey. Dr. Sarmiento. Thank you so much. Yes, so I shortened that wordy title for this slide here, so thanks for listening. We talk about our project about disparities in smoking and heavy drinking behaviors by disability status and age of disability onset. So just a little bit of background before I dive into the details of our study. So what prompted us to ask this question is that increasing numbers of children with disabilities are living into and thriving in adulthood, and that transition to adulthood is a very vulnerable time for many health outcomes, including adverse health behaviors like smoking and heavy alcohol drinking, which often begin in adolescence. Studies have shown higher prevalence of smoking, illicit drug use among adults with disabilities, but we haven't really looked at the effect of age of disability onset on these adverse health behaviors. And so our concern was that adolescents and young adults with disabilities may not receive age-appropriate screening for some of these health behaviors, and so we wanted to look a little bit more at this relationship. So our objective was to examine the relationship between current cigarette smoking and heavy alcohol drinking and disability status, as well as age of disability onset in adults. So we did this as a secondary data analysis of the 2020 National Health Interview Survey, or the NHIS, or the NHIS, which is a nationally representative annual survey on health and illness in the United States, and includes information on disability as well. So the participants that we included in our analysis were those respondents to the NHIS that were between 22 and 80 years old, which was a little over 28,000 adults, and that included just over 3,400 adults with disabilities, and that's kind of disabilities in a functional manner, so across diagnoses. In addition to looking at just disability status as yes or no, we also looked at age of disability onset, and we were able to do this because in 2020, the National Health Interview Survey included for the first time a question sponsored by the Association for Community Living, which asked to those participants that had a disability, did your disability begin before or after age 22 years? And so we defined childhood onset disability if they said it began before age 22, and adult onset disability if it was after that time. And while 22 might seem old for a childhood onset disability, it does align with the federal definition of developmental disability. The outcomes we looked at were self-reported current cigarette smoking and heavy alcohol consumption, and the NHIS has several questions that it uses to classify alcohol consumption into light, moderate, and heavy drinkers, and previous studies have used this definition as well to look at adverse health effects of heavy alcohol consumption. And then we used all this data and we used adjusted logistic regression to look at the adjusted odds ratios of smoking and heavy drinking based on disability status and then based on age of disability onset. And we adjusted for many demographic factors in that, which I'll tell you a little bit more about as we look at these results. So in this first forest plot, we are looking at the unadjusted with the open white circles. Let me see if I can get our laser pointer here. Great, so this is looking at unadjusted, the blue bars with the open circles, and then those are unadjusted, and then adjusted with the filled in diamonds and then the green bars. And so we looked at odds ratios for current cigarette smoking, current heavy alcohol drinking, and then a secondary outcome that we looked at was whether or not those participants that endorsed current smoking or drinking were advised by a health professional to either quit, cut down, stop that behavior within the past year. And so I'll focus on, for our discussion, the adjusted regression results. And so for our adjusted regression models, we included covariates age, sex, race, ethnicity, income level as a proportion of the federal poverty level, as well as insurance status and relationship status. So if we look at that, we can see that adults with disabilities, which is to the right on this forest plot, is the odds are higher for adults with disabilities, and on the left is odds are lower for adults with disabilities. So adults with disabilities had significantly higher adjusted odds of current cigarette smoking. Similar adjusted odds for current heavy alcohol drinking. And then if we looked at the cessation advice outcomes, they were statistically significantly more likely to have received advice to quit smoking or to cut or stop down on alcohol drinking within the past year. If we look at a similar forest plot, looking at age of disability onset, so here to the right is the odds are higher for those with childhood onset disabilities. And here on the left, odds are lower for those with childhood onset disabilities or higher for those with adult onset disabilities. And so here, when we look at the health behaviors, we see that there's no significant difference between adults with childhood versus adult onset disabilities in either smoking or heavy alcohol drinking. We also saw no significant difference between the odds of receiving alcohol cessation advice for those with childhood versus adult onset disability, but we did see that those with childhood onset disability were less likely to receive tobacco cessation advice despite similar odds of smoking. And for this set of regression analyses, we adjusted for the same covariates except for age because age was collinear with age of disability onset, which kind of makes sense. And since that was our main effect we were looking at, we excluded age from that model. So just to summarize all of that, compared to adults without disabilities, adults with disabilities had higher adjusted odds of current smoking and similar adjusted odds of heavy drinking, and these odds did not vary by age of disability onset. So what does this mean and what can we do with this information? So adults with disabilities might be a high-risk group for these unhealthy behaviors, and this is in line with some prior studies that have demonstrated the same, particularly with smoking. And what our research adds is that this is regardless of age of disability onset. And so adolescents might be a crucial time for screening and counseling on these unhealthy behaviors. It's again, prior studies have shown that this is the time period when a lot of these lifelong behaviors get started. And this counseling and screening includes adolescents with disabilities who might be missed in these routine screening efforts. And so that routine screening is really important for everybody, all persons with a disability, including those with childhood onset disabilities. And again, we should begin that in early adolescence and continue that through the life course. Unfortunately, prior research has shown that adults with disabilities receive lower rates of preventive health screening, including screening for smoking and alcohol use, despite emerging and increasing evidence that they're a particularly high-risk group. So thank you. I wanna thank our co-authors and team, Anna Furness, who is our amazing data analyst, and then Drs. Morris, Stransky, and Thompson. Happy to answer any questions. Nice job, Dr. Sarmiento. What's on? Nice job, Dr. Sarmiento. Sorry, Sarmiento. Any questions? Oh, I have to do it myself. So this was interesting. I thought, you know, I kind of, this seems like it parallels a lot the old work comp thing, who's gonna not get back to work with low back pain, the smokers, the hyper-religious, you know. Were there just those two points that we were looking at, was drinking and disability and smoking? That was the focus of these analyses. We did another set that we, I actually, I presented last year on looking at mental health outcomes, depression and anxiety symptoms. And spoiler alert, adults with disabilities have very high rates of mental health issues. And those with childhood-onset disabilities had even higher risk or odds of having mental health systems compared to adults with adult-onset disabilities. What I learned from diving into this data set is it's this incredibly rich source. I think there's lots of opportunities to look at even more differences. And so this is kind of where we started. Is smoking only tobacco or smoking other things, too? So this one was smoking cigarettes, pipes, or vaping. So any kind of tobacco smoking. It specifically excluded, like, smoking marijuana. Yeah, that was, you know, as marijuana gets liberated from state to state to state. And what my mentors used to say, it was like a cheap vacation, especially if you're disabled and you really can't get away. That's why people drink. That's why people do other things. It's escapism of some sort. That's a great point, yeah. That's what I was told when I was a resident. I had some smart mentors. Anybody with any other questions? What's the next part of the study? So where I'd like to go, so while this, I think, helped raise some really interesting points, one of the frustrating things about, like, secondary data analysis is you can't say, well, why is this the case? And how do we do something about it? So where I'm hopefully heading is to do some qualitative work looking at kind of preventive screening overall for individuals with disabilities, particularly childhood onset ones, as I'm really interested in that transition to adult care. And how do we develop, like, patient-facing self-management tools to improve rates of preventive screening among individuals with disabilities? It didn't look like, and I couldn't tell from the bars whether or not that the discussion, the advisement to stop, did that actually do anything? Or was that just wasting physician time? Great question. We're all forced to see a lot of patients these days. Everybody else. It didn't ask, like, did this advice change your behaviors? Presumably, it probably didn't change them dramatically because they only asked those questions to people that were still currently smoking or drinking. So you really couldn't tease out if it actually made them do anything or got them angry and they actually did more smoking than drinking because somebody in, quote, authority told them not to do it. Yeah, great point. Although, you know, it is, the US Preventive Services Task Force, which spells out the recommendations for preventive screening and spells out evidence behind those, does say that screening for alcohol and tobacco use in all adults is a evidence-based recommendation. Just to give you more to do in your very promising career, it would be interesting if there were a way to compare, you know, we think a lot about diversity in health opportunities and such. Are young people who have early-onset disabilities equally likely to get anti-smoking messages in their teen years as non-disabled teens? Or is there some implicit bias that we're not aware of in our own practice that, I don't know, maybe people feel like, oh, you've already got a hard lot in life with a disability, yeah, go ahead and smoke if you want to because I'm not gonna rag you about one more thing. That would be interesting to know. I don't think you can get it from this particular data pool, but your intellectual curiosity on this can make you a wonderful researcher in that area as well. Yeah, I think that's such a great point. I think that's my currently not evidence-based presumption. I think maybe some of the arguments you put forward, I think there's also probably implicit biases related to infantilization of adolescents with developmental disabilities. Oh, you know, all this person with CP, they're not gonna smoke, they're not gonna drink. Yeah, yeah, it's really interesting. If they have a smartphone and the smartphone's listening or they have Facebook and Facebook is listening, then will they get those targeted ads that tell them to smoke, not to smoke, go here, do this, do the other thing, or if they have spasticity, are they gonna take more CBD or something like that to be looser, right? Because even when I was a resident on the spinal cord unit, people were sneaking in marijuana, so they had less spasticity back in the 80s, and we made a conscious decision not to go after that. Great. Thanks, Dr. Sarmankyo. Thank you so much. You're welcome. Jennifer Sloan. Dr. Sloan is gonna talk about determinants of discharge destination after inpatient rehabilitation for non-traumatic spinal cord injury patients, a single-center retrospective review. All right. Here we go. All right, thank you for having me. I'm Jennifer Sloan. I'm a fourth-year resident at Coral Well Health East, formerly known as Beaumont Royal Oak in Michigan. Wanted to thank Alex Shoemaker and Brandon Trevex, other physicians on this project with me. So we were looking at determinants of discharge after an inpatient rehab stay for non-traumatic spinal cord injury patients. Patients with spinal cord injury are an important population that we treat as rehab physicians, and a main goal that we have for these patients is to become as independent as possible with the goal of reintegration into the community. And most of these patients need a inpatient rehab stay. But unfortunately, not all of them can reach a functional level to be able to go home after an inpatient rehab stay. So the goal of this study was to identify any important determinants that will help us see whether they are more likely to go home or go to SAR after an inpatient rehab stay. So we did this study at Beaumont Royal Oak, which is a tertiary care university hospital. We have a single inpatient rehab facility there. It's about 60 beds. And this was a retrospective cohort study. And we looked at patients' electronical medical records retrospectively. So the variables that we looked at included sex, age, body mass index, the length of stay on the acute care side, the length of stay on our inpatient rehab, the injury type, the injury level, the Charlson comorbidity index. So this is basically a score of how sick the patient is. So it looks at mortality risk and based on their comorbid conditions. And then we also looked at insurance provider. So whether they had Medicare or another insurance. So patients that were included in our study, we had 249 patients that met our inclusion criteria. This would be any patients with a new non-traumatic spinal cord injury. And they had to have either gone home or to a subacute rehab facility after our inpatient rehab stay. So we didn't include any patients that had an old non-traumatic spinal cord injury. And we excluded any patients that went to say an LTAC long-term care facility afterwards or if they had to be readmitted to the acute side, or if they unfortunately expired on the unit. And this was electronic medical records from 2012 to 2022. So the main outcome measures that we looked at was after the inpatient rehab stay, did they go straight home or did they need additional rehab at a skilled nursing facility? So looking at our logistic regression analysis, in figure one, we see all of the variables that we analyzed. And then, again, that would be male versus female, age, BMI, Medicare versus other, our length of stays, CCI. For injury type, we looked at structural versus other, and so structural would be like spondylosis or stenosis versus other included, neoplasm, vascular, infectious. We did not include inflammatory in this because, like MS, transverse myelitis, because it can also affect the brain and I'm not sure if it would be a true spinal cord origin. And then injury level, we compared upper and lower cervical and upper and lower thoracic. So in the first column, we see our unadjusted model, second column, our adjusted model. So adjusting for all other variables. And as you can see, our p-values, 0.005, 0.002 for age and BMI, those were our statistically ... Yeah. Better? It's a little small. Okay. Those were our two significant variables when adjusting for all of our other variables. And then I can go back if you needed to. So then figure two, we have our odds ratio plot. So again, you can see that age and BMI do not cross our line at one. So those are our statistically significant variables. And then they are both to the right of the line. So those were more likely to be discharged to SAR instead of home. All right. And then get out for you. So further with our odds ratio results, we did find that for each additional 10 years in age, the odds that a patient was discharged to a subacute rehab facility after the inpatient rehab stay was about 1.78 times higher. And then the odds for every one unit increase in BMI, the odds that a patient was discharged to a subacute rehab facility was 1.09 times higher. And then as we see, no other significant differences were found in the other variables that we looked at. So in conclusion, I thought this was an interesting study for us to use for a couple of reasons. When we're on the acute side determining who is an appropriate patient for our inpatient rehab, this is something to keep in mind. Age and BMI, we can say those are not surprising factors that may influence a patient to need additional rehab. But it's good to have the data to back that up. We might want to talk to these patients early on and say, okay, yes, we want to bring you to our inpatient rehab facility, but no, you might be at a higher risk of needing additional rehab. And so to have those expectations early on, a big thing that we do as rehab physicians is talking to families and establishing clear goals and expectations. And so to know that these patients are at a higher risk or might need a little bit more attention to be able to get them to get home at a functional level will be helpful. So a couple limitations that we thought of for our study, where the patient was discharged to, we are assuming that that was the appropriate discharge location. It may be that they went home and that wasn't the best place for them to go at the time. And then we also didn't look at the severity of the injury, like we didn't look at ASIA scores or mobility scores. So that could be an interesting thing to look at, or that definitely could influence our study as well. And then this was done many years through the COVID pandemic. And as we all know, there was a lot of hesitancy to go to SAR at the time because of COVID outbreak. And so that could have also altered our data. But thank you for listening. Thanks, Dr. Sloan. Thanks, Dr. Sloan. Any questions? All right, I'll start. I got a bunch. So, nice study. So this is non-traumatic spinal cord injury? Sure. At both cervical and thoracic, no lumbar levels? No, we did not. Okay. And they subdivided them, so I have four levels. So what do you think the average length of stay was before they went either home or to the SNF? Yeah. So our average length of stay is around 10.5 days, I believe. Wow. Even for cervical? So it's a pretty quick stay, and it would be interesting. I trained under the late, great Dr. Murray Freed, one of the founders of the field of spinal cord injury, and he would be astounded that it's that fast, especially with cervical spine. So when I was a resident, and we didn't have cars and phones and this and that and the other thing, cervical spine stayed in the hospital between six and 12 months. Paras, three to six months. Sound about right? Yeah. It's an interesting point. Who's forcing them out so fast? Is it Beaumont? Is it the insurance company? Is it the type of insurance? I mean, I'm astounded that somebody's going to go with a spinal cord injury to an SNF, because they're not going to get the care. Yeah, and we don't have a dedicated spinal cord injury unit, and so we might not be seeing the sickest or most severe spinal cord injury patients. They're usually, honestly, transferred to another facility if they're going to need a very long length of stay. But you're sending them to an SNF. They're not done. Sure. Do you know the rationale for the SNF placement? Are they not tolerating therapy? Are they older? Yeah, so I think it's a combination of our unit. That's our average length of stay, and then also insurance if they're not making significant progress. But our spinal cord injury patients, we do keep longer than usually 10 days. That average was for overall diagnosis, not average for a new spinal cord? Exactly. Yep, yep. It's still fast. So I stopped seeing patients almost two years ago and now work for the insurance company and do the thing, so I'm one of the bad guys now. But I would tell you, spinal cord injury, I don't rush any spinal cord injury out of the hospital. Yeah, we always obviously want to do what's best for the patient. Ever. And there's about 30 physiatrists that work at my company now, and I think we all have a pretty similar view on things. Anybody else with any questions? Oh, thank goodness. What impact did availability of in-home care, a caregiver, have on the discharge site? And length of stay? Definitely. So that's something we really wanted to look at, and we tried to make that a variable when we created the study. What is family support? Because as we know, that's a huge, huge variable here in getting patients home. There wasn't enough documentation that we could use from the electronical medical records for that many patients for 249. And so it was a little vague with our retrospective review what the support was. So if there's a spouse there, are they there all the time, and how much support they do have. But that's one of the big directions that I'd like to take this study is, how does family support influence that? Because I think that's a huge, huge factor. I assume that none of these patients had spinal cord injuries or any kind of neurological diagnoses. What's factored into the Charleston comorbidity score? Yeah. More things like diabetes, hypertension, all of those comorbidities. They didn't have any neurologic diagnoses before, no. And we'll wrap it up with Dr. Sarmiento's questions. Thanks so much. I'll say, at Colorado, at our adult-based hospital, UC Health, our average length of stay, last time I checked, was 10 days overall as well. Spinal cord injuries did tend to stay more like six weeks, but I think that's why we need more physiatrists and insurance companies making these decisions. Yes. Absolutely. But now my question is, I'm curious for your insurance variable, why you chose to do Medicare, yes, no, versus Medicaid, yes, no, or private, public? Sure. Thank you. So, as we all know, the insurance companies do dictate a lot of whether they can go to a subacute rehab following the inpatient rehab stay. And so, unfortunately, that's something that we have to take into account when bringing our patients over. If we can't have them for six weeks, if that's not going to be covered by insurance, then inpatient rehab, you know, it's always the best option for a spinal cord injury patient. But we do have to have those clear expectations where, you know, you might have to go home after a two- to three-week stay, and is that a realistic goal, and do you need a longer stay? So, that's why we did Medicare versus all the other insurances, because most of the time, we can easily get those patients to subacute rehab without a problem. Thanks, Dr. Sloan. Yes. Nice job. Thank you. Appreciate it. Okay. I'm going to use readers. All right. Next, we have Justin Tram, who's going to talk about factors associated with increased readmission following inpatient rehab, post-spinal cord injury, traumatic brain injury, total knee arthroplasty, and total hip arthroplasty. I'm going to put the end in there. Hi, everyone. Thanks for sticking around for the last presentation on the last day. So, I'm Justin Tram. I'm a fourth-year medical student at Albany Medical College in New York. I'd like to thank the rest of my study team, Dr. Sanjeri, Dr. Linzemeyer, Dr. Thiel at Sunnyview Rehab for working on this with me. We don't have any disclosures related to this project. So, rates of readmission back to acute care once a patient has been in inpatient rehab is something that has been well-studied. These rates are normally about 9% for TBI, 20% for SCI, 8.5% for total knee arthroplasty, and sub-10% for total hip arthroplasty as well. And factors that are associated with return to acute care have also been well-delineated too. These factors include older age, as well as presence of comorbid conditions like congestive heart failure, diabetes, and obesity. But the studies are not as prevalent in regards to what causes a return to the acute hospital after a patient has been discharged home from acute rehab. And that's where our study tries to come in, where we attempt to define these clinical and demographic reasons for a 30-day readmission to acute hospital after the patient has been successfully discharged home for these following conditions. And the purpose of the study was to just identify the high-risk patients for future risk stratification. We utilize the National Readmissions Database. This is a national public database sponsored by the U.S. government. And we extracted a cohort of patients who had an ICD-10 condition code of spinal cord injury or traumatic brain injury, or an ICD-10 procedural code of total hip arthroplasty or total knee arthroplasty. And we only examined these patients that had a primary diagnosis or primary procedure code of one of the following. And this is important because if a patient comes in with, for example, post-motor vehicle accident, they could have a number of traumatic injuries. But if a TBI or SCI is not the first primary diagnosis on their chart, they would not be included in our cohort. And that's done so that we have the main condition that's causing the inpatient stay, and we have a more robust cohort for our study. And we only examined patients that were greater than 18 as well. Demographic characteristics that were studied in our study include their age range, 18 to 44, 44 to 64, 64 to 74, 74 plus, their sex, male or female, their race, their payer type, and then their income quartile. Admission time and setting characteristics were also studied, whether the hospital was in an urban or rural location, what region of the United States the hospital was in, the trauma level of the hospital, and when the patient came in, whether this was a week in admission or the month of admission to the acute hospital. And then we also examined different comorbid conditions that the patient had as well on their chart, which we've broken up into specific categories, including musculoskeletal system, endocrine system, circulatory system, skin conditions, as well as nervous system disorders. And once we had all these attributes, we did descriptive statistics and multiple logistic regression to see which factors were associated with increased 30-day readmission after the patient was sent home from acute rehab. So we had a total end of 25,852 patients who fit this criteria. We had almost 10,000 TBI patients, 1,500 SCI patients, 5,700 total hip arthroplasty patients, and 8,600 total knee replacement patients. And our numbers for 30-day readmission after discharge were a little bit lower than return to acute care during the acute rehab stay, which makes sense because patients who are post-rehab discharge home should be in more stable condition than patients in the acute rehab setting. And the factors that were associated that we found with increased 30-day readmission were the usual suspects, things like age 74 plus, Medicare, and male sex. However, interestingly, other demographic factors, such as race or income level, were not significant, and the concomitant diagnoses that were significant in return to acute care, such as congestive heart failure or diabetes, those kinds of things, were not significant in our study. So some of the limitations in our study include the fact that, as previously mentioned, we only specifically included patients with a primary diagnosis of either that specific conditions or the procedural codes mentioned. So the database also was not created with rehab in mind, so it didn't include factors like functional scores, ASIA scores, these types of rehab characteristics that could also be of interest to us. And it did not specify between those who were admitted to the same hospital for acute rehab or those who were admitted to a standalone rehab facility for their acute rehab setting. So future directions for this. The real big takeaway is that we, as physiatrists, really need to advocate for the creation of more public databases that are rehab-generated, that are rehab-focused, to have the characteristics to make for stronger, robust studies that we can use to guide patient outcomes in the future. But hopefully this initial study provides a good benchmark for future studies as well. Thank you. Thanks, Justin. Good job, man. Any questions? I'll lead off. How's that? Can you go back to the readmission numbers where you have total hip replacement, total knee replacement? It's fascinating that total hip replacement readmission is twice as much as total knee. Yeah. So looking at the data, is there a way to sift out and see if there was any rationale why total hip had a 6.7% readmission rate, what the actual diagnoses were? No. So these patients came in for elective surgery, so all that's in the charts is that they had a total hip replacement done for the most part. But the numbers for total hip replacement and total knee replacement were—total knee replacement was actually more surprising for us because their rate of return to acute care for total knee replacement is 8.5% in previous studies. We found it only to be 3.8% here. But as you said, yeah, we don't know for sure at this time. You know, it's different years you look at. The data, things are always different. And each state has their own thing about—and each insurance within each state has their own thing about who they'll allow to go to either an SNF level or an inpatient rehab. And typically with total hip, total knee, it's now done as ambulatory, and you don't even get an overnight stay in the regular hospital. You know, when I was a resident, you got two weeks in the regular hospital, and then you went to rehab for two weeks in Boston. Connecticut, you didn't get to go to anything. You were home in a week. But yeah, this was a national public database, nationally represented. So hopefully these differences in guidelines between different institutions in different states would be— Well, there's hundreds of thousands of hip and knee replacements done every year now. They're done very, very frequently. And so the relatively low percentage now goes to inpatient rehab. They usually have other things going on in their lives where you have multiple comorbidities to even be able to get into a rehab center. Even two knees at once. Two knees at once will get you— What is the MG thing that I have to go through? The inpatient guideline, the McGillicuddy or whatever it's called. You know, the MG guideline will give you, I think, two to three days. I think I can give two days, a third day if there's a problem is what it'll pull. So it's interesting. I mean, there's a lot. You did a good job. You picked a lot of factors and stuff. I don't know if you had any hypotheses in your mind before of what you would expect to be determinant factors. But it looks like male gender and increased age were the key ones. You said payer-type Medicare, but there's a large percentage of Medicare patients already. Were they, what, 50%? I guess I'm surprised that TBI readmissions are a higher percent in general than others. I don't know if you have any speculation of what you thought these numbers might be and how they turned out. Yes. Does it include patients who leave against medical advice and then come back? For TBI? For TBI, right? That's not a variable that is included in this data set, so we didn't look at that. But what was interesting when we were looking through this was our SCI number was actually quite low compared to our other conditions. It's only 1,500 compared to TBI, which is close to 10,000. And that could potentially be due to, as I was saying before, where traumatic SCI patients could have a different injury be coded as their primary condition for why they were admitted to the hospital. So those patients would not be included in our cohort. But that's one of the things that we really want to look into for a secondary analysis before we turn this into a manuscript. Nice. Any other questions? Justin, really good job, man. Thanks for coming. Thanks for coming all the way from Albany. And I have one other question. We have possible one other presenter. Is Shane Werdemann here? Well, thank you very much, Mark Allen. I'm the outgoing chair. This is my last 30 seconds of being chair of the Evidence Committee. I appreciate all of you coming for today's showcase. All the presenters, thank you for doing your jobs. You did a super job, and we look forward to seeing you next year in San Diego. Thank you.

rehabilitation

cancer survivors

spinal cord injuries

traumatic brain injuries

total knee arthroplasty

total hip arthroplasty

return to acute care

discharge destination

readmission

subacute rehab