Chapters Transcript Video Shared Decision Making in OAB from step-therapy to treat to target Giulia Ippolito, MD, MS Good morning, everyone. Uh, we're gonna go ahead and get started. Hi, Julia. Uh, so I have the pleasure of welcoming, um, Doctor Eppolito, um, for our Zoom grand rounds today. Uh, so Doctor Eppolito, um, completed her medical school at Texas A&M and went on to do a residency at the University of Minnesota and then has been at Michigan ever since, um, and completed her fellowship there. Um, I've known Doctor Ipolito for a long time, uh, Her and I have served on the GSM AUAUO guidelines together. We also serve on the CERN network together as well, the Sufu Research Network, and, uh, she really is a thought leader and um spearheads a lot of, um, our health, um, sciences, our health services research and, uh, she's going to be talking today about shared decision making in OAB. So thank you so much, Dr. Balio, and it's a pleasure to have you. Thank you so much. I'm really excited to be here. I'm gonna share my screen and just make sure it's all set up. One second here. And then let me make sure it's on present to you. One second. Can you guys see the actual slide or my presenter view? We see your presenter view. OK, but. Always happens. Let me reshare the correct screen. How's that? Perfect. All right, so thank you so much. This is really exciting um to be able to join you guys for grand rounds and tell you a little bit about my research story. Um, I've been really fortunate to have a lot of mentorship and be funded through training grants and Um, this is really the story of my K-12 journey, um, and leading into my RO1, which, um, I'm really excited to see was recently funded, looking at, um, OAB more closely. Um, what I hope, uh, to talk about is where we are with OAB right now, where the current guidelines are, what shared decision making is, um, how it has a role in overactive bladder, and then introduce a concept that I'm calling treat to target therapy for overactive bladder. So, I'm sure many of you, um, if not all of you are very familiar with overactive bladder, so this is a syndrome that's defined by urinary urgency, um, uh, frequency and can be associated with urinary urgency. Continent. Um, the impact of OAB is widespread. Um, it is often can be debilitating and very um stigmatizing, limiting people's quality of life and also, um, their ability to interact in the community. Prior to the 2024 AUA guidelines, our therapies for OAB were a tiered system, so we were um escalating care um after failure of first line, second line, or third line therapies. Um, and patients navigated um this clinical roadmap through these therapies. Um, and I think what's important with this roadmap is it Um, to me, and, and some interviews that we'll go over, um, gave this thought that there was an end to therapy at the end of the road, and, um, that might not be aligned with OEB as a chronic condition. So the AUA uh and Supo put out a new set of guidelines which um really changed the paradigm of how we look at overactive bladder. Um, and what I want to focus on is the um therapies that we now offer for OAB rather than being a tiered system. Uh, can be, um, uh, used, uh, together or escalated without having to go through 1st, 2nd, or third line therapy depending on patients preferences and values. Um, and so one of the, um, early statements in the guideline is on shared decision making and um recommends that clinicians should engage in shared decision making for OAB to make informed decisions about the treatments, um, or no treatment for their symptoms. And so first I'd like to talk, before I talk about what OAB is, I'd like to talk about why um the statement exists and why is OA is shared decision making appropriate in overactive bladder. Um, so, overactive bladder is a preference sensitive condition. So there are multiple treatment options. There's what we call clinical equipoise. So this is relatively similar outcome data between the options, um, within each category of treatment. And the choice is largely based on patients' preferences and values. And so in these situations where there's preference sensitive conditions, and there's no clear right answer on what the first um or how the therapy should be delivered in what order, um, shared decision making is very important because shared decision making is a preference, is a patient centered way to find the right decision when the medical evidence has this um. Uh, um, equipoise. And so let's look at all the therapies and sort of look a little bit at the data of like what that means, um. So our non-invasive therapies are things like pelvic floor physical therapy, bladder training, urge suppression, um, and then fluid management and biofeedback. Um, and, uh, there, these are what we call low medical risk profile. So there's really not high adverse events that are gonna, um, be caused by this. Um, they equally all require patients to be engaged and committed, so that's, um, largely one of the biggest barriers to these therapies. Um, and the outcome data is mixed, so there's no clear support of one option over the other that's where to start. So in these low stakes preference sensitive decisions where there's not one clear winner of what to do first, um, that puts us, um, again, uh, points us towards shared decision making as what to decide. So our next um category of therapies are pharmacotherapies, um, and there's been uh a large amount of evidence looking at the different, different pharmacotherapies for overactive bladder, so we know that we have anticholinergics and beta 3 agonists. Um, and, uh, in a 2012 and review of all the anticholinergics, they found that there's really similar efficacy between um the medications, but what's distinguishing between them is their adverse event effects profiles, um, and then, um, between the two available beta 3 agonists, um, they are, um, Um, both superior to placebo and importantly they're comparable in their effect to anticholinergic. So, um, again, we're at this, um, uh. At a point where there's really no clear medication that is the one that we should start on for our patients. Um, they, there was even what's called a network analysis done, so this is where you Usually a meta-analysis compares maybe an intervention in a control, but a network meta-analysis can take from multiple different arms and put all that data together, where there might not be a head to head comparison between two interventions or an intervention and control. Um, and that also found no difference, um, in medical efficacy between, um, the different drug formulations that have been studied head to head. So again, we're at this point where decisions between specific agents are really largely based on access and patients preferences and values regarding adverse effects. There's no medical evidence driving you towards one medication over another. Um, and then finally we're looking at our minimally invasive therapies. So these are, um, sacral neuromodulation, um, neuromodulation with posterior tibial nerve stimulation. I know there's, um, implantable, uh, posterior tibial nerve stimulation now, and that is really not studied yet, that there's not um a lot of evidence that we can pull. And finally, chemo generation. So the Rosetta trial compared sacral nerve modulation to chemogeneration, and, um, while there might have been a statistical difference, there really wasn't a clinical difference in the number of urgency incontinence episodes per day. And then again, another network meta-analysis in 2020. So there hasn't been a head to head trial between, um, each of the modalities, but this is a way that we can look at that data head to head. Um, found that Um, sacral neuromodulation had uh better incontinence episodes, um, compared to PTNS, but they all had um similar improvement in in effects compared to placebo. Um, and like I said, we don't really know where the implantable nerve stimulations are going to land, um, in this, in the spectrum, but so far the data as I understand it, um, and I've reviewed it, um, seems to put it on, on the same spectrum as um PTNS or uh sacral nerve modulation. Um, so for minimally invasive therapies, it's really the distinct, uh, distinction in the delivery and the risk profiles rather than again the clinical efficacy that distinguishes these therapies from each other. So now that we sort of have thought um of why is shared decision making appropriate in when we're making this decision, these decisions for our patients coming to the office with overactive bladder, I wanted to talk about what is over what is shared decision making and how can we support it in our clinical care. So shared decision making is a bi-directional process and it balances patients' preferences and values with the best available medical evidence, um. And you use that evidence to answer the clinical question or come to a, a treatment, um, uh, therapy plan with your patients. Um, but in order to do this, and I'll get into this later, and where I sort of got stuck earlier in my career, sorry, when I was um looking at this as a, a resident and fellow is that Um, we don't really have great data on the patient's preferences of values or the best medical evidence. Um, so, uh, we have a lot of limitations of being able to individually, um, provide care or predict outcomes for patients, um, which is really different as compared to if we look at all the calculators that are available for perhaps, um, guiding care for people with a malignancy. Um, and we'll talk about that a little bit later, but you know, the evidence on shared decision making, why you should do shared decision making is pretty clear. Um, shared decision making can decrease decisional regret, um, and decisional conflict among patients. And importantly, especially in a condition um that's chronic and it requires patient engagement and um therapy, it can improve patient compliance and empowers them to be part of their care, um, improves their satisfaction with their treatment and their trust and confidence and knowledge. And it's also guideline concurrent care as we just talked about um that uh our AE guidelines recommend shared decision making. There is a lot of models for shared decision making um that are available. This is one from um the AHRQ that um is highlighted in the AUA white paper on shared decision making, and essentially the um things that are important to remember about how we do shared decision making is you have to let the patient know that there's a decision to be made and that there's multiple options. And there's no clear right answer, because if, if it's sort of like if you don't know that you need to make a decision until the end, um, you might not be listening with the kind of ears that you listen with when, um, you're being asked to make a decision. And I think a lot of, um, Patients come into the clinical um sphere, uh, maybe relying on the clinician to largely make the decision, and so they might not be digesting information that you're giving them in the same way as they would if you stay up front. We're gonna be making a decision together. I really want you to think through these options. Um, after you just, um, explain and compare options, you have what's called an informed patient. So, um, it's a little bit unfair to tell people you have to make a decision, but not and ask for their preferences before you tell them what the options are because their preferences might change once they hear what the options are. So after that, you ask them their preferences, and you come to a decision, and it's also very important, especially in overactive bladder, which again is a chronic condition, um, to have follow-up about how, how did this strategy work and do you have any changes in your preferences? So it's really important and what um we found uh in uh some work I did as a fellow, is that eliciting patient preferences is the center of um shared decision making, but it's the most commonly misstep um of shared decision making. So one other model that I like to highlight is what's called the three talk model, and this, um, I think makes it more simple with some uh really easy cues. So, you know, let's work together to make this decision, let's compare options and tell me more. Tell me what matters most to you for this decision. And that last um phrase I think is really um I've started using it, or since I've started using it in my clinical practice, it's always really interesting how people answer that question. So if you ask people what matters most to them for the decision, um, and what their main goal is, um, treatment, it might really not have anything to do with with what their chief complaint is. Um, and so I, I really like doing that in my clinic. It's important to know what shared decision making is not about, so it's not about just letting the patient do whatever they want. Um, it's not about wasting time. If it's really clear what's best for a patient, there certainly are those scenarios in, um, overactive bladder care, especially when you're looking at people with comorbid conditions that um might take some options off the table, um, or add add options, um. It's not just saying, here are the options make your decision, um, and it's not just being nice to the patient, right? This is um really supposed to be a conversation uh with boundaries based on the medical evidence. So now that we have this um framework of how we do shared decision making and why it's important and overactive care, where does it fit into our algorithm? And um, we wrote a paper a few years ago looking at this and, and the main push back on shared decision making and the thing that people really say is the largest barrier, um, is it takes a lot of time. Um, that has, there's been a lot of studies that like time. Uh, encounters with shared decision making and um the, the time difference between encounters is actually not significant, um, and it can almost sometimes be found to save time because it reduces questions or follow-up visits. Um, but there's a a way to do shared decision making that's called everyday shared decision making. And the premise behind this is that rather than saying these are all the options available, the clinician um makes the diagnosis, does the history and physical, and then Um, developed an individualized estimate of benefits and harms based on the patient characteristics. Um, we don't really have the data to do that in overactive bladder. There's not a lot of individualized tools, but we do have some our clinical judgment that can help us. So for example, if there is someone that has um both overactive bladder and um accidental bowel leakage or fecal incontinence, I might lead with, because you have these two conditions. You know, I'm recommending this option, but I also want to talk about these other options. Um, so that's step number 2, leading with the discussion of options with the initial recommendation. So that sort of streamline the streamlines the process. It's not that you're not telling people what's all the, what the options are, but you are giving them a clinical recommendations to begin with. Um, One thing that I that can support shared decision making and sometimes gets um conflated with, uh, shared decision making or decision aids. So decision aids are tools that are meant to support the, um, decision. They're not made to replace. So, um, some thoughts are like, well, can people Just do this worksheet before they come to my clinic, and then we'll just talk about their decision. That's not really the paradigm that we suggest for decision aids. They're really supposed to be in a supportive role. Um, this is just a QR code to one that we made, um, at the University of Michigan, um, that I usually on my first encounter with a patient with overactive bladder, we'll provide them so they can start thinking through their preferences and values and look at options. OK, so we have um talked about why shared decision making is applicable to overactive bladder and how to do shared decision making, and now we're at this point of, well, how can we facilitate that individualized patient centered care, the individualized outcome estimates um through shared decision making. So again, to facilitate shared decision making, we need to understand preference and values and individualized outcomes. Sorry, my animations came out there. So for the preference and values, I'll be talking about a mixed method study that we did as part of my um K-12 here, and then for individual outcomes, um, I'll be presenting some research we've been doing leveraging the electronic medical record. Uh, OK. So, um, This was a study that we did, and there were two fabulous medical students who are now urology residents um that were leading this, um, Hannah Cu and um um Casey Brodsky. So what we did were semi-structured interviews with patients and clinicians, um, and also in uh surveys of those, those same people. And then we took the interviews and the surveys, we transcribed them and coded them, um, and then we came up with a thematic analysis to see what it, um, is important to share to decision making, especially around um minimally invasive therapies. So from the patient experience with overactive bladder, um, patients, um, in their interviews really had a lot of desire for knowledge. Um, uh, the knowledge was not specifically to, um, what the clinicians were telling them, but in general, there was a thought of how did I not know that this was a thing, um, uh, until that it was happening to me. They wanted to shared decision making. So, um, patients want to have a discussion of what the options are and especially if there's any new options, they want to be aware of those options. Um, They spent a lot of time in treatment trials, um, and I thought this quote, Don't waste two years of somebody's life suffering while you are testing different medications is really pointed. Um, and unfortunately, that is what happens to a lot of people is that they, um, sort of get stuck, um, in, in a category and can't move beyond that. Um, there's a lot of uncertainty of outcomes, so you never know what you're gonna get, um, through that trial period and, and patients really dislike that. And then, um, one thing that we found is that there was this discordance between expectations and outcomes. So this patient said, you know, just 50% better doesn't seem like good enough to me. And that was really striking to me because 50% better is what is on all of our trials to say that these medications are efficacious. And so then we looked at the clinician side. um, and clinicians find um overactive bladder treatment rewarding and just to specify, these were all um urologists or gyneco uh neuro gynecologists that um Focused on overactive bladder treatment and offered minimally invasive therapies. So these the caveat might be that these um Responses might not be generalizable to uh either primary care physicians or general urologists. But they found it rewarding because of the potential impact on people's quality of life, um, but very frustrating because of the limited treatment efficacy and most importantly, the barriers to access. So it's a game trying to get them there through that medication step. Um, and wanting for more patients specific, um, treatment options. So, um, Uh, the clinicians and the patients had a lot of overlap, but when we asked them what matters most for making their decision, um, the results were very, um, uh, uh, dichotomous. So clinicians, which is the light blue, uh, really, uh, favored safety. So the risk in adverse event profile was what was most commonly said as the what's most important when you're making this decision. Um, interestingly, um, the patients valued efficacy and had, um, less concern about potential risks than adverse. Um, so there's this, um, divergence in what the goal of therapy are and what people are prioritizing in their internal calculations as they're making these decisions. And that's really important, um, not that we have to We don't have to make these um Sentiments aligned, but we just need to be transparent when we're having conversations. So this would look something like um in a conversation, I would say. I'm very concerned about the risks, or I'm prioritizing the risks of these potential therapies. Um, and because of that, my recommendation is this, um, and the patient might say, Well, I want the most efficacious therapy. I want the therapy that's gonna work the most quickest, and I'm willing to undergo risks or take that risk. Even if it means it might be a little bit higher than some, something that's um has a a uh lower risk profile. Um, so when we put everything together and what's called a mixed method integration, um, the main themes that came out of this research were people are frustrated with inaccessibility of OAB treatments. Um, there's this, um, discordant perception of patient education, so the, um, clinicians were saying. That they're doing all of this counseling and we, I think a lot of us spend a lot of time because there's so much to go through now, especially since all the options are available, um, but the patients were really not feeling like they were getting the counseling at a level that they could understand. um. Like we talked about, there's this divergence in the acceptability of expected outcomes, um, so clinicians, um, think 50% better is the sort of, um, benchmark, and patients might not think that 50% better is worth, um, or is, is an appropriate benchmark. And then again, like I alluded to, there's this lack of insight into the other party's decisional priorities. Um, so, um, being transparent in that, and I think the onus of that really falls on the clinician because the patient might not know all of these um things that they have to weigh, um, is really important. So, um, we talked about the preference and values, um, and what's most important when making these decisions, um, we, uh, got to the Patients really wanting to know what the outcomes are going to be as a main driver. And so now the next um step of the, the, the series of, of projects was looking at individual outcomes. And as a health service researcher, there's, um, traditionally people use um like claims data to answer these questions and Um, there's a huge limitation in claims data for functional neurology conditions because it does not capture um the outcomes that are patient relevant. Um, uh, we can have surrogate outcomes, um, but my goal in this is to try and leverage the EMR to develop a Computational phenotype and I'll go into that. Um, and also to develop an ability to get outcomes that are patient, um, relevant. So just a little detour here, what is a computational phenotype? Um, when we look at the electronic medical record, there is a um wide difference in how the data is input into the record. So on the far left side, there's what we call structured data, and this is very similar to claims data, so it is um Essentially, um, discrete points that can be easily mined, um, if you can think about it, it's like, well, if you do an output of this, it will be like an Excel file. Um, the codes are very discrete, um, and the, the data is, um, can be easily gotten by uh exporting or like like for example for claims. Um, semi-structured data is things like problemless or medical diagnosis, so, um, these are contained in the, um, medical, uh, in the EMR in a way that is able to be accessed, so we can pull them into our notes, we can export them, but people can do like free text, so they can add things to the um problem list and Um, there might not be, it's not as structured as like the diagnosis code or the procedure code. And then there's unstructured data and this is things like your notes or the, um, uh, radiology report. Um, so a computational phenotype, um, takes in all of this data, so they, you know, the Starting with the structured data and going to the unstructured data, and tries to overlap it in a way that we can identify the true cases of people with a certain condition, um, because if you rely only on one set of data, You might be systematically missing people that for some reason don't have that structured data, for example. Um, or what happens a lot is you might be miscategorizing people that are labeled in a certain way because someone is doing all that labeling, um. Uh, uh, that really shouldn't belong in your cohort. And so for this part of my K-12, we um were able to build a custom data set from the EHR and this um data set was primarily structured data and so we use structured data to pull um the initial cohort, um, so things like uh diagnosis code. Um, we're used to just put in a wide net to get everyone that had, um, potentially, uh, overactive bladder. And then we created a computational phenotype and this took into consideration um if like office visits, um, clinical notes, medications, um, to make sure that we had identified true cases of people with overactive bladder. Um, and as part of this, we did like a 10% sample to make sure that um we could identify the people. Appropriately. And I think that's one of the strengths of using uh the electronic medical record, especially if you're at an institution where you can gain access to the patient's charts, is that you can check your work um to make sure that you are actually, what you think is true is actually true. And then we uh use this to make um outcome, to look at outcomes for overactive bladder care. Um, what I would love to see is a nomogram for OAB outcomes, but we're not there yet. We're still on that first row. So this was a project by our former fellow Brian Jang, who's out in Boston now, um. Uh, and what we looked at is men and women with refractory, idiopathic overactive bladder who had undergone minimally invasive therapy. And we looked at the outcomes of how many people require um additional therapy, either medication therapy. Um, Uh, or minimally invasive therapy, um, within, uh, or after they get their initial minimally invasive therapy. So what is the longevity or success rate, long term success rate of overactive bladder? We're defining success as not needing a different therapy and maybe there's a flaw in that we'll talk about that a little bit later. And so we what we did is we looked at time to event analysis through Kappameyer. Um, we looked at Cox's proportional hazard models to see whether we could identify variables that are related or associated with um need for more therapy. And finally we applied um a machine learning model called a random forest to see if that could include more data than the Cox um hazard and come up with um or help us determine which um patient level predictors um are associated with need of more therapy. So we had just over 1000 patients over about a 10 year timeline, and um 45% of these were patients with PTNS, uh, 35 had chemogeneration and 19, 20% had um sacral neuromodulation. Um, of the PTNS patients, only about 40% went on to have maintenance session. For sacral neuromodulation, we limited these to people who actually had a successful trial and an implant, so we weren't looking at, like, did you have a trial and it failed? We wanted to say, once you have your established um Uh, once you've had a successful trial, um, what does the future hold for you? And then, um, for chemo denervation, um, we looked at anyone who had had Botox, or sorry, um, chemo denervation as their first um choice. Um, the demographics make clinical sense, so, uh, people choosing PTNS were older, um, and, um, there was a higher proportion of people who chose PTNS as their first line therapy. That makes a lot of sense. Um, and then our population was majority white, and there was an underrepresentation of black or African American compared to the national population, which is around 12%. So first we looked at time to median additional therapy. So, um, we had a follow-up time that was 15 months for the PTNS and between 20 and 30 months for the sacral neuromodulation and chemogeneration. And within that Kapayer, the median time to needing another therapy for PTNS was 10 months. Um, for SNM was quite long, uh 53 months, and for chemoration was Uh, about 20 months, so just less than 2 years. And then we looked at at one year, what can we tell people that, what can they expect as far as needing an additional therapy. Um, and so we looked at this in two ways. We looked at any additional therapy, so Uh, another that would include another minimally invasive therapy or uh just some medication. Um, so the rates of this are really quite high. So PTNS is about 60%, which makes sense, might need an additional therapy, um, sacral nerve modulation about 20%, and chemoderation about 40% at one year. Um, and then we looked over at 3 years. So what can people expect? And at 3 years, um, uh, except for the, uh, sacral neuromodulation group, um, the majority of people are needing an adjunct or additional therapy, um, to their, um, minimally invasive therapy. So our next question was, uh, are there variables that can be, that are associated with needing an additional therapy? And so we use a Cox model to do this, and we, I limited this to women because of two reasons. One, I don't think you can put men and women in the same model because there's a lot of differences in why they're, they're underlying OID, um, uh, clinically, um, and I don't think it really makes Uh, and then when we tried to do separate models for men and women, they're just not enough, um, events in the men to make a model. So we limited this to women only. Um, so for PTNS we didn't find any factors that were associated with with needing additional therapy. For sacral marrow modulation, um, Uh, multiple indications is protective. So this is, um, the patient who has fecal incontinence and urgency incontinence. Um, those patients with dual incontinence are half as likely to need an additional therapy, um, than, uh, patients who, um, only have one only have urgency incontinence. Um, we found that those people with a higher BMI. Uh, had about double the um rate of needing or double the risk of needing an additional therapy. Um, for the chemo nervation group, the, um, people who have recurrent UTIs and we were able to define this by um having a culture or having a diagnosis, so it's like a it was a um composite definition, um. And it was uh 3 UTIs in a, in a year, uh, in any year of the follow-up after their therapy. Those who had more than 3 UTIs, um, in 1 year were about double the rate of having to um have an additional therapy. And then we looked at the random survival um for models, and I'll just take like a brief minute to do a primer on this. So, um, this is a machine learning methods. Um, it's useful for um exploring scenarios where there's like less known about survival data, so we thought that that would be like Especially applicable to this, because I, I'm not aware that other people have published survival data, um, or looked at survival, as we, you know, here I'm saying survival, but what I mean is time to, uh, without an additional therapy. Um, and random force, uh, models provide what's called a variable importance measure, which is, um, the change in the prediction error of a new test case if this variable was not. Um, available given that that same variable was in the original forest. Um, It gets a little bit complicated to explain, um, but essentially the variable importance, um, is within each model and really shouldn't be compared between. So you shouldn't compare the numbers across and think that that has like weights or relevance. It's really just in the same. Um, uh, categories. So we did 3 different models. So for PTNS, um, for all the models, the top two predictors were their BMI and their age. Um, and we're still looking at which direction that goes, um, uh, as far as, uh, is it younger or older, or is it higher or lower BMI? So I'm still trying to sort that out. Um, and then after that, uh, pretty consistently was their comorbidities and their number of visits to urology, um, which makes a lot of sense. If you're having more visits, you're probably more likely, there might be a little confounding because you're more likely to get something done or be given a new therapy if you're coming in. Um, interestingly, alcohol use came up in this, um, uh, in a at menopause is, uh, and we go usually down to 10. I truncated here. So there might be Other things to explore that we're not really putting in our models, um, that the machine learning models, um, is, is helping us understand. So, uh, at the end of that, and those two projects, um, some takeaways that I've, uh, that have helped me move on to the next step is, um, Patients really want knowledge and they want to participate in their decision making, and there are a lot of opportunities to improve, and those themes that, um, we did in the qualitative work have really set the foundation of where we should be, where I plan to focus my attention and my research, um. The things that um are important to me are aligning treatment goals between clinicians and patients, um trying to um eliminate some of the uncertainty of outcomes, so decrease that time spent in trials, um, and also decrease that discordance between expectations and outcomes. So having a, um, Clear goal that everyone agrees on of what uh success, how we define success for overactive bladder treatment, and I don't think that that is routinely done in clinical practice. Um, I also found that we can leverage the, um, medical record to explore outcomes for OAB, and we were able to create that computational phenotype. Um, and it's really changed my thoughts around overactive bladder that the need for adjunct therapy is very common. Um, and this is something that I've started to really share with my patients that these are therapies, not cures, and I don't necessarily think that the need for an adjunct means that people have failed or, um, it might be that their underlying characteristics have changed. Um, Uh, or this might just be the, the truth of how this, um, is how overactive bladder could be managed in this escalation, de-escalation, and I'll talk about that in a, in a minute. Um, One of the things that we, that I'm hoping to include in the computational phenotype outcomes is direct measures of efficacy. So we do have patient reported outcome data and I'm hoping to include that. So in this project, we looked at um need for additional therapy, but in the next project, I'd like to flip it to look at how many people have a change or what is the change in patient reported outcomes. And then, where is this going? So, um, I'd like to touch a little bit on what's called treat to target therapy. So treat to target therapy is um early intervention and tight symptom control to optimize long-term outcomes. And this is um something I borrowed from rheumatoid arthritis and um irritable bowel disease. um. And Uh, what? The way that this is done is that targets of cure or of success are set between the patient and the clinician at initial visits. And for rheumatoid arthritis, examples can be both um uh patient reported outcomes, um, symptom scales, or there's also like lab values that they look at, um. And then um the therapy is chosen based on um getting people to their target. Um, so this would look like in LAB that you would say, you know, you're starting with 5 urgency incontinence episodes a day. Um, and, you know, waking up twice a night, you'd like look at all their symptoms and say, what is a target that is, um, acceptable, like, what is your goal? And that would drive our decisions about around therapy. So if someone is starting with 10 and they want to get down to 3, your therapy might look a lot different than someone who's starting with 2 and they wanna get down to 1, right? Um, So the main difference is in that from standard of care is that the initial treatment is based on a severity score that is decided between the clinicians and patients. Uh, uh, sorry, and the treatment goal is decided between clinicians and patients, and it really supports this concept of therapy over cure, that this is a chronic disease model, um, and that it might require escalation and de-escalation over time. So for rheumatoid arthritis, um, they may start with very aggressive therapies, um, But once someone reached their targets, they might de-escalate some therapies if things are going well. And in overactive bladder, clinically, I've seen this fluctuation, um, and so I think it is very applicable or at least should be considered for how we think about overactive bladder care. And so how would this fit back into our shared decision making model is that um we would make the diagnosis of overactive bladder and then before we even start going next, we'd quantify people's severity, which we are already doing that's guideline concordant care of getting patient reported outcome measures. And then I think what's different from what we're doing now is we incorporate that severity into patient-centered targets. And I think by having these targets, then we can start going back to our shared decision making to develop those individualized estimates of, you know, how can we get you to this target, um, and again, um, helps us lead with that discussion of what our recommendations are. And the decision of which therapies and how to combine therapies, so multimodal therapies often very um Common in treat to target therapies, especially upfront, um, and then you can remove therapies if needed. Um, so I think this could be a new way to think about how we address overactive bladder. Um, there's also always like a lot, there's obviously a lot of barriers, and the biggest barrier to treat to target is that we are in a step therapy model and our pairs are really, um, aligned with step therapy, which is essentially the opposite of treat to target therapy. Um, step therapy says that you have to go through the steps in, in a specific order. And so my pay, you know, I, when I thought about this, I thought. If I don't understand step therapy and really um Quantify the consequences, we can't propose alternatives, um, and so the um My my R1 is basically scaling the um computational phenotype using what's called the cornet, which is a nationwide um Database based on the electronic medical record that employs what's called the common data model, so multiple institutions have um the same data that we can get um that um I use for my foundational work. Um, to address, um. The question of who is getting minimally invasive therapies. Importantly, what is the impact of insurance status on utilizing these minimally invasive therapies and, you know, I'm trying both the potential benefits and the potential harms. And then finally to start exploring the acceptability and appropriateness and feasibility of that treat to target approach for overactive bladder. So this is, um, what I'm starting and I will be working on for the next 5 years. Um, so stay tuned. Um, so I'd like to just really thank, uh, my team, especially my mentors, through my training grants and, um, the many, many trainees and learners and statisticians that have, um, uh, really helped this, uh, push this research forward. Um, and I'm really happy to take any questions. Thank you so much, it was such interesting research. Um, I think it also brings up so many questions, um, that I, I don't know so much of what you said, maybe you think about so many things. So some of the questions I have. Early in the lecture, you mentioned that one of the patients in your, in your um focus groups mentioned that they wish that testing had been done earlier. I thought we, we were having this discussion with our fellows the other day about like, what do patients want in terms of urodynamics. When they said testing, is that what they were meaning or what did testing mean? I can't, um, I don't recall that specific example, but there was definitely the notion of people wanted to be, um, diagnosed and and be told about the condition earlier. I think there is on that spectrum, like if we're thinking about that treat to target approach and here today we've only talked about therapies, but you could definitely apply it to the to invasive testing to say, You know, your severity is here, your goal is really low. If, you know, the best way for me to get you to your goal is to understand your disease by or not your disease, but your syndrome better by doing that upfront testing. So Um, I really, I, I think that sometimes we feel like, well, we shouldn't do urodynamics until we try some things first. Um, and I'm hopeful that what I've, um, talked about is that really, that is an opportunity to say, what, what matters most to you? Like, what are you trying to get out of this visit? And if the patient is saying, I just want this better as soon as possible, then I think it's absolutely OK to proceed to Aerodynamics very quickly if that's what's needed to get them to their target. Got it. And then I, you know, in your last slide, you were talking about how insurance is such a big, you know, it's gonna be a big focus of your work. Um, it's such a problem here in Florida. Uh, as much as we say you can do anything you like, they're so governed by what the insurance will allow and and cover, um, and, and a lot of them, you have to do step wise therapy no matter what because of insurance. Um, do you guys face that problem? Do is that something you're gonna be exploring in terms of patients' relationships with insurance issues? Yeah, absolutely, that is essentially the main um focus of that grant, so that aim number 2 is hoping to leverage. So absolutely in both ways. One, we definitely have trouble with insurance, um, uh, and, um, I'm hoping that we can quantify the impact of The the delaying care or the impact of how many people are not getting therapies that they might have otherwise chosen, um, because of their insurance status, and that's Um, what I'm planning to do is leverage, um, what's sort of like a natural experiment and how, um, prior off is. Um, enforced, so as you know that private payers use prior authorization, whereas Medicare uses audits and clawbacks, and so that that difference we see always a really No, what I've seen in in our cohort is that there's a big jump in um use of minimally invasive therapies after people turn 65. And I don't think it's because their overactive bladder has changed between when they were 64 and 65. I think it's because how um prior authorization is enforced, um, differently between Medicare and um Uh, private or commercial insurance. And so that is a difference that we can quantify. So if the only difference, if you think that all being the same, there's not much, the only difference that happens within a two year period is that someone changes their insurance status, then we can use that change to quantify like what, what is that difference? You know, how many people would have gotten it earlier if they could have. Um, and I think that my hypothesis is it's gonna be a significant, um, Impact, and I'm hoping um to use this data to maybe um drive policies. That might convince insurers to say, you know, you're actually um probably not saving money because we're stringing people along with expensive medications that have side effects. And so I think in order to change insurers, it's going to be a lot of work by a lot of different people looking at what is the quantifying the impact of these policies. Such great points. Um, it'll be amazing if, uh, we can convince insurances to do something different, no matter what, no, I, I am, one thing that I'm really um interested in exploring, there's a lot of, um, you know, there's music here that is the collaborative that we have, and music is funded by Blue Cross, um, and so I really think that um. After I quantify this, um, or at the same time looking at they're called like um CQIs, so community quality improvement collaboratives, where um researchers and institutions partner with um insurers to provide better quality care. And I absolutely, you know, music has focused on prostate cancer and kidney stones, and now they're doing BPH, but I think that there is a lot of opportunity in overactive bladder to um Uh, provide quality improvement in the care that other people are getting. Absolutely. Um, OK, we do have another mini lecture from a student, so I'm sorry, I kind of monopolized the questions, um. But thank you so much for for being available for this great talk. You know, best of luck with all your research and uh it's, it's really great work. So thank you. Thank you so much for the uh the invitation. Bye everyone. Bye. Awesome. Thank you Published October 30, 2025 Created by
