Health Information Systems webinar series

(DESCRIPTION) A video conference with Tara Dull and Kathy Harkness. (SPEECH) Good afternoon. And welcome to our community webinar on October 4th. It's hard to believe that we're already in October. I actually had a schedule an appointment for December already. So welcome to our fall community webinar. And we're excited to have Golden Valley with us today. Before I get started and pass it over to our speakers today, I'm just going to go over a couple of housekeeping items. We are utilizing the ON24 platform, which is a web-based platform. So we recommend using Google Chrome, close out a VPN, multiple tabs. That'll help with your bandwidth. There is no dial in number. So check your speaker settings and do a quick refresh. If you are experiencing any audio issues, definitely check your settings. We do have several engagement tools you should be able to see-- the media player, where you can see our beautiful speakers that you can adjust. We also have a closed caption if you need it. There is a presentation area. Because this is mostly a discussion, you can actually close out of that if you want to. There is a resources section. So feel free to download those resources. There is a certificate of attendance for today. And so if you do want to utilize that to request CEUs at an association, you can. But that is just a certificate to say that you attended today. Please ask questions. We'll get to as many as we can at the end, utilizing the Q&A section. And again, we'll get to as many as we can at the end, and then complete the survey at the end. We ask you to let us know how we did. And that's always going to help us for the future. And so again, we're really excited to be hosting Golden Valley Memorial with us today. And I'm going to go ahead and pass it over to Kathy Harkness as our moderator today who will introduce our speakers. So Kathy. Thank you, Lisa and welcome to everyone. I'm really excited that we have a great attendance. And I think you're really going to enjoy listening to these great ladies from Golden Valley Memorial Healthcare. They've had a great journey. They started their program during a time when most of us-- most of us were locked down. And struggling with a lot of the crisis that was going on during COVID, but they had this planned, and they were working on getting their program going. And during that time they reached out and were looking for vendors, and we get the privilege of being their partner with CDI Collaborate, which they went live in 2022. So I welcome them today and excited about learning about a little bit more and sharing with you about what they are. So I'm going to turn it over to Leila and Tara. And Leila, will you'll give us a quick introduction of who you are and your role there and then I'll turn it over to Tara? Hi, I'm Leila Houck. I have been with Golden Valley for going on eight years now. I've held a couple of different roles. I've done chart analysis, at first, with our inpatient and observations. From there, I went to utilization review. And from utilization review, I jumped on the CDI bandwagon. And we're so grateful she did. Hey, good afternoon, everyone. I'm Tara Dull. I've been at Golden Valley for 12 years now. I'm our Executive Director of Compliance, Privacy, and Data Integrity. So our health informatics, case management compliance, chargemaster, coding and release of information, population health as well, all of those teams report up through me including CDI. OK. Tara, do you mind just giving a little background about Golden Valley, just about your community status and just give us a little, not huge, don't have to go into the P&L, but just can go into just give us a quick overview. Absolutely. We are a rural facility. We are about an hour Southeast of Kansas City, Missouri. And we are licensed for 56 beds. We've got four remote clinics or four total clinics, sorry, including three remote clinics, as well as multiple specialties. We're also part of an accountable care organization or ACO. So we're very much on that value-based care journey here at Golden Valley as well. Well, let's talk about the other journey that we're here to talk about is that CDI program, so give us, I said that you started this during COVID and in 2020, and talk to us about what you were looking for in a program and the journey of pre-Collaborate. Absolutely. Yeah. We knew it was something that we wanted to pursue. I, like many of you, I'm sure received that lovely pepper report from CMS, right, and get a lot of insights regarding NCC capture or lack thereof, in our case, some of it was certainly lack thereof. So we knew we needed a program and in a rural community, it can certainly be hard to recruit folks. And so I was confident we could build our program from the ground up and with internal resources. So I twisted Leila's arm and she graciously agreed to join me on this journey and build our program. And so we did an analysis to see what kind of query opportunities there would be in potential return on investment as well as what that case MCS index shift could potentially look like. And so once we did that, then we had started our process totally manually, if you will. So we launched CDI in Microsoft Excel. So Leila tracked all of her reviews in Excel. We tracked our key performance indicators and financial metrics in Excel. Our query process for our providers was totally manual. We're not necessarily proud of it, but it worked. And it's what we had available at the time to be able to prove that return on investment to administration and senior leadership, so that we could justify the cost of software. So once we got to that point, then we were ready to select a vendor partner to help make our processes more efficient and help us be more productive, so that we could spend more time doing CDI and doing the work as opposed to trying to track our efforts and track down providers. So at that point, we signed with-- Collaborate, we were-- of all the vendors we looked at, the software was the most-- I really felt like it met you where you're at, if you will. So if you had a CDIS with 10 years of experience, great. And if you had someone, that perhaps didn't have any experience, in our case, and we were really going to be kind of learning from the ground up, we really felt like Collaborate stood apart from other vendors with that. So it sounds like you did look at a lot of vendors. I mean, there are-- we have a lot of competitors out there in the field. When you were doing that, was there a specific thing, an element about CDI Collaborate or about CDI tools that struck you? Yes. There were several. One of the big things that I liked about Collaborate was the ability to have clinical insights, right? So if we did potentially have someone without as much CDI experience or maybe they didn't have as much clinical experience, to get those clinical insights and say, hey, you've got a patient with this BUN and this creatinine level, you might want to look at this. This could potentially be indicative of this clinical indication or this clinical condition. That was something that was very meaningful to me, because I was under the impression that we likely wouldn't be fortunate enough to hire someone with 10 years of experience. Just at a rural facility, that can be hard to do. So that's certainly one thing that was very impactful to me as well as the NLU piece, so the natural language understanding and that embedded artificial intelligence to be able to use those insights and have your queries generate from that artificial intelligence. We've seen great success with that. I think our average capture, our query capture from that NLU is about 70%. Yeah. So that means for the viewers that about 70% of the queries that you send are generated from that natural language understanding insights that you're able to do. So let's talk about queries. Like many others and back in the day, I would put-- I would print out the queries and put them in red folders or put them in the chart, and they'd be different colors, so that we could get the provider's attention and such. Talk to me about when you were in this tool or any tool, really, how has your query process changed? Yeah. So I will definitely let Leila speak to our current state. She's most equipped for that. But I can definitely tell you what our previous state was prior to Collaborate, and we generated a query in Microsoft Word, Leila probably had hundreds of documents saved for her query library. And she has the ability to work remotely. So she would print the query on site. And then we would have a staff member that would truly go and track down the provider, sometimes, they're waiting while the provider is on lunch break or the provider says, hey, come back and catch me at 3:00. Whatever the case may be. So there was definitely room for improvement there. Current state, I'll let Leila speak to. Currently, yes, it has been a game changer. Now, I can draft the queries in Collaborate. I send them, it goes through Meditech, it goes on to the patient's chart. The doctor sees it. It goes to his workload. He can answer it. The query response rate has been phenomenal, usually, it's that day or it may be the next day. And because of that, the nurses can see that because it's on the patient's chart. Case management can see that. It's just a lot less headache, really. You don't have all these paper queries that you're trying to keep track of. Everything is in the system. You get an alert when the doctor has responded. So getting those queries, noting them, noting the effect that it has on the account, it's just been a game changer. You're not logging all of that into an Excel spreadsheet. In addition that, interface with our EHR has provided us a 100% physician response rate. I would say, probably for the last year, I don't know that we've had one query go unresponded to, which is nice as well. Wow. That says a lot for your physician, team as well. So technology is great. I mean, I'm not going to say that we don't appreciate the kudos there and the partnership with you around that, but boy, what a great team you have there for your providers that are able to do that. I just want to tell you that for our viewers also, they do have Meditech Expand. So that we talk about integration and things. For those of you out there that have that, that's kind of interested in our medical or our EMR integrations and things. So not that we don't integrate others, but I just want to call it out there. So it sounds like obviously, we wanted to get you some automation, we wanted to create some efficiencies. Query delivery is so big, but are there other-- were there other goals that in your program that you were looking CDI Collaborate, to work on as well? Absolutely. So some other goals that we were really looking for were productivity. So as you can imagine, if we're doing everything in Excel what I alluded to earlier, we spent a lot of time trying to track our work and keep track of our work as opposed to being able to spending more time actually doing the work. So that was really a big one for us. And the productivity and our coverage has really driven everything. Certainly the more charts we can review, then the more financial impact to have, more queries we're able to generate. Our case mix continues to climb. So really, that productivity and efficiency that we gained, it was really tenfold for us, because we saw it in every other key performance indicator that we were tracking, if that makes sense. You being able to cut out, the paper has been a big thing for us. One thing I don't think that I've mentioned that was also very important to us in the vendor selection process and that we probably truly just take for granted today is to being able to work in one platform. So we're able to launch Collaborate through the EHR, and then we can view everything within Collaborate. So Leila has the ability to see the medical record. She can search for medical record terms within Collaborate. So she's not having to toggle back and forth between the EHR and multiple other vendors. Of course, it launches seamlessly with our encoder as well and the other three reference tools that are available. So really, that seamless integration between the products makes-- helps drive that efficiency as well. And I think Leila's current productivity, I think she spends an average of 16 to 17 minutes per encounter. Great. Now your coverage and your population, is it just Medicare or all payers? We primarily focus on Medicare, certainly depending on what census allows for, then there is the potential to expand out into other payers. But Medicare is certainly our payer of focus. That's one of the questions that we get from previous webinars conversations. So I just want to clarify that for the group. So let's talk more about the NLU for a moment. How did you select, because our library is growing even as we speak, our library consists of multiple-- I think over 25 different clinical conditions, but then has other clinical triggers around it. So we can set up multiple clinical triggers around sepsis. There's lots of different ways to look for sepsis, a lot of different ways to clinically validate sepsis. So we have about 13 or 14 different sepsis trigger opportunities. So going through that vast library, how did you-- how did you take that on and deciding? And for those maybe need to back up here, we talk about natural language understanding, and that artificial intelligence engine that we call in our business catalyst, that's our brain, that we use to develop these clinical triggers and look for these clinical conditions. That is what we deploy into CDI Collaborate to help you search and tee up opportunities within your patient population. So when we turn on these clinical triggers, we might turn on anywhere from 10 to 20 to 30, depending on what the need is and what the opportunity is at your organization. And so sometimes, making that selection, I mean, there are some obvious ones, right, that pneumonia, those ones are pretty much the foundation. I can probably go to any hospital in the country and say, hey, I bet your top five queries are-- heart failure, type and acuity and type of anemia. And we could probably call those out. But when you're trying to make that selection through this vast library, how did you-- how did you go about doing that? Sure. So probably my biggest piece of advice would be to rely on your solutions advisor and your implementation specialists. They are very, very familiar with that NLU library and how it works. So we certainly relied on their insights. They asked a lot of probing questions or said, hey, you might consider this, or do you have any patients like this in your population. And Leila and I were shaking our heads, yes, saying, oh gosh, we wouldn't have thought of that. So I would definitely tell you to rely on them. They are the experts. They're going to get you through it. But there were certainly some conditions that Leila had mentioned, as being a thorn in her side, sepsis, as you mentioned, Kathy, as well as others, CHF, respiratory failure, I'm sure. No one is stranger or no one's a stranger to those queries for certain. So that's really what we started with are those conditions that we knew that we were querying for and with a lot of volume and frequency, and then relying on our implementation team to really guide us. The NLU is so powerful until you can see it. And they pull that library up for you the first time, it really is overwhelming. It's also a thing of beauty. And when I say it's overwhelming, it's because there are so many choices and options, and you do have the ability to just continue fine tuning it and making tweaks until you get it working for you. And we certainly have it working for us, which we're grateful for. And that really speaks to, again, I hate to just keep repeating our productivity and efficiency, but that 70% NLU query generation means 70% of the queries that Leila generates she has had the help of NLU if you will, to suggest those clinical conditions that need some additional attention. And just, because I say it's growing, it's growing. Basically, like customers like you who say, there's a condition that we're dealing with here, and maybe other customers might have that condition and the opportunity, but maybe it's not built in our library. So with those, we do have the ability to create new chemical triggers for you, just so you know that that's why it's growing, because there's always a different way to look at things. There's always different guidelines that come out. Gosh, malnutrition, everybody is-- that's one of those things where it's like, well, I have this criteria, but I really also want to look at this too. So we know in the field that our lives are in a constant state of flux and change in the sense that rules change all the time, unfortunately. It's a lot of job security. And I love that you called our library a beautiful thing, overwhelming and beautiful. I mean, I'm going to have to be using that from now on. So thank you. I'd love to hear that. Because I think it's the same thing. I mean, that's why I asked the question. I'm like, oh, my gosh, how did you choose? And I think the other thing too is it's not-- you don't-- once you choose, and you can either tweak it to make it right or a better fit or you can say, you know what? I don't think we need that one. Let's do something else. And that's why-- I just talk about that-- so yeah, we're going to talk about that solution advisor role a little bit later. But that's the beauty-- that's the other beautiful thing about that is we do have a person that is for the life of the contract that can help you be flexible in those decisions. So nothing is-- nothing is in stone at that point. Well, taking on a new product and implementing a product can be very daunting. And so let's talk about that implementation process. What did you do to help prepare? And how would you describe that time? Absolutely. I'll tell you what. We got homework every week on our weekly project calls that Collaborate led. They give you a project manager as part of that implementation, which we were incredibly grateful for. And so really, we just focused on making sure we had our homework done each week, if you will. And we're putting out on that progress. And there are a lot of options along the way. So our first option was gosh, we've got queries available in the Collaborate, query library. Do you have some of your own internally that you'd like to use, a hybrid mix of the two, that sort of thing? So we made a lot of decisions like that, if you will, kind of around surrounding our initial build and launch and what that was going to be like and then worked on those weekly project calls. We did, as I mentioned earlier, create multiple different interfaces because we've got our entire EHR coming in to Collaborate, as well as interfaces on the provider side to be able to query electronically within our EHR through Collaborate. So a lot of our interface work of course, was done by our IT team, as well as Collaborates IT resources. And then we trained, we had training here on site. And then we were ready for GoLive. Leila and talk about it and look back, we had an EHR conversion not too long before we implemented Collaborate as well. And it truly is the smoothest GoLive that we ever-- that we've ever participated in. There weren't any showstoppers or red flags, if you will. Each day got a little better. Leila certainly had to start her work from scratch, and we had talked about that. What is the cut over look like? And I finally looked at her one day, and we just kind of agreed, you have to start somewhere. You just have to cut over, and so we're like we're not going to keep worrying about closing out what we were tracking in Excel. We can speak to why we've got a gap in our metrics here for a few weeks. And it is what it is. We're moving on to better things. What would you like to add Leila? Just the-- oh, gosh. I don't know. I mean, it was very seamless. I mean literally, we went live at midnight. I logged in the next morning. And I mean, it was all right there. Everything that I usually had to sort through or print out or try to organize an Excel, I mean, it was all right there. It was, like I said, it was seamless. And we had support the day of GoLive as well. So we had a project call. Set up for the day of GoLive, so that we could talk through any hiccups or perhaps anything that we had forgotten in training or maybe we just needed a little bit of positive reinforcement that we were on the right track and continue to have calls subsequently after GoLive. And you had training on-site. So you had an in-person training, because I know, post-COVID, a lot of people have-- we can do it off-site or we can do it on-site. So that is some-- that is your call not necessarily ours, because we will always be on-site if requested, but that was what you found that very helpful to have somebody on-site with you. We did. And we appreciated the option of, certainly with COVID facilities, have different policies regarding that. But just for Leila's level of comfort, as well as my personal preference, if you're going to be spending that much time training, sometimes it just seem like it's more beneficial or it just flows better. The virtual environment, if you're virtual for eight hours day in and day out, training it can sometimes be a little cumbersome. So we really enjoyed the on-site training. And if we did it again, that's certainly what we would opt for. Yeah. I think the other thing that you mentioned that's really important, and I do a lot of sales conversations about our product, I always talk about we talk about implementation and I say dedicated resources. And that's from our side and yours. Because that's where-- if we have a dedicated, especially IT, and all organizations, no matter the size, their IT resources are stretched. And so we always talk about that-- I try to always talk about that trying to find one or two people that we can reach out to and say, we're setting up this file or we need this interface tested or whatever, having that dedicated resource is probably a big game changer. So it sounds like you had that and, and that's-- I can't say I can't understate the importance of that. The other thing is we talked about interfaces, we do get a DRG feedback for reconciliation. So that's another one of those things that we set up and test. So I don't know if we talked about reconciliation, but that's one of the key pieces to making sure that the reporting and the analytics and all the information coming back and that you're able to leverage and show return on your investment for the product. So very important. So let's get to that special person that has really made a difference. And I think is one of the things that really made us stand out. So this is, when we were acquired by 3M several years ago, we had adoption specialists for our product, our clinical solution products. And CDI Collaborate is one of them. 3M has seen that that has been a game changer for customer satisfaction. And so they are rolling out that adoption specialist role in other products that they have, and we call them solutionvisors now. So you hear us talk about adoption specialist versus solution advisors. It's a new title. So please those listening, don't get confused. It's the same person. But CDI Collaborate, for their product, for the life of the contract, you are assigned a solution visor to help you with the training, with the everyday questions, with the monthly meetings, with their reporting. And I'm going to introduce her. At this time, she is on the line. Laura Neale, if you'd like to introduce herself and just talk about your role for a second and maybe I probably stole some of your thunder by talking about what you've done. So I apologize for that. (DESCRIPTION) Text, Laura Neale. (SPEECH) So yes, I am Laura Neale. I am an RN and a CDI specialist. I have worked as a CDI specialist and also as a leader. So I have that experience that I use in my role at 3M as a solution advisor. So my role is to help customers utilize their application in the most optimal way. And my goal is to ensure that the best outcomes-- we get the best outcomes based on the determined success criteria which Tara has spoken very well to their goals, and also in a short time with CDI Collaborate, how they've met those goals. So the communication, the monthly meetings, and then emails in between and the incredible insight that Leila has, has really helped us to tune their NLU to work well for them and to help meet those goals like has already been called out. And it's really been a pleasure working with Tara and Leila and celebrating their successes with them. Yeah. So Tara and Leila, now you get to talk about Laura. So describe how this feature, you've already said that you were surprised that this was an asset that you had for the life of the contract and that this was a continued relationship. It wasn't just, hey, we're going to install this, and I hope it works for you. And here's a 1-800 number. Let us know how you're doing. If you have a problem, call this number. That's never been our philosophy. We definitely are-- we stand behind our products, and we put talented people. All of our solution advisors with CDI Collaborate product are CDI specialist credentials, some inpatient, some outpatient, some both. So we're very, very proud of the team that we have procured. And we are able to supply that to our customers. So I'm going to stop talking. And you guys can talk about how wonderful they are. So your solutions advisor, and as is the case for us, we get Laura, and we're so grateful. Truly makes all the difference. So especially if you happen to be a smaller facility with limited resources, like Kathy mentioned earlier, to have somebody else to bounce ideas off of and really to have the knowledge of the NLU, as well as how Collaborate works, if we do see something that's a little different in a metric between the three of us putting our heads together, we can usually pinpoint exactly, you know what it is OK, we need to make this tweak here or we just didn't realize that Collaborate worked this way. So to be able to have additional suggestions and insights and even I've asked before, hey, what's the benchmark for this? What should we be shooting for on a physician disagree rate or whatever the case may be? It's been very, very helpful. And like Kathy mentioned, after our GoLive, I fully expected we would walk away at some point. That's when vendors are hands off or is typically the case with my experience anytime you've got a vendor partner. And so it really takes the meaning of a vendor partner to an entirely new level, because you truly do have a partner there at your fingertips literally working alongside you elbow to elbow. And it really makes all the difference. I'll let Leila share her personal insights. I guess the only other thing I would add too is that you have access to your solutions advisor by far more than just that monthly call. Laura usually responds same day via email, that sort of thing and is also great about communicating like product enhancements and insights, things that are coming down the pike that we might want to be aware of as well. Like Tara said earlier, we started the CDI program literally from the ground up in 2020. And then in 2022, we went live. Even though Excel was a pain, being somebody new in the CDI world and just learning, I just started getting comfortable in Excel and then we switched to the new program. But having Laura there afterwards after the GoLive, there was some hiccups, there were some different things that I had to do differently in Collaborate that the past two years I had done in Excel. And she was great. We could do like a call or we could do screen shares and help work through that flow. And just if I provide feedback, another thing in NLU, if you see something that maybe that they're picking up, if it's-- then there's maybe a discrepancy, you can provide feedback in the NLU a certain subject like respiratory failure or sepsis, and then they take that feedback. Laura looks at that feedback, and that also helps tweak then the NLU specific to your facility. And like I said, Tara said, just having the access to her has really helped the workflow and getting comfortable then with the Collaborate. Yeah. I think I even remember a time or two when we were discussing probably pretty soon after our GoLive about something that just hadn't quite clicked. And I think Laura said to Leila, well, how did you track this in Excel? And then Laura was able to translate what we're tracking in Excel and how we're not able to find that in Collaborate. So like I said, I think our experience has been fantastic. And Laura, as a solutions advisor, really, really makes the difference, I think, in your ability to leverage the product to get it to work for you. So that you-- how I said earlier, you're going to be as successful as the effort that you're willing to put in. And I can tell you that Laura puts in just as much effort as we do. Well, what's nice is just in the reporting, when you talk about that, it's like how did you track this before, and how are we looking at it. We do have self-service reporting in the Jasper reports. But so again, compared to other vendors, that might be like, oh, well, you can download it, you can look at it. But what's nice with the monthly meetings is she does look at those reports, and she's looking for those changes, those-- does this look right? What happened? Did you have a surgeon go on vacation this week? Those kind of questions about that. So I think that's something that I am very proud to mention during customer calls to say, we're going to walk with you through a lot of this. Now, when you talk about-- we do have consulting services that take it to a different level beyond and our solution advisors do work with our consultants that customers have purchased those services when you talk about program analysis, deep level benchmarking, things like that. So I just want to call that out that they are there to do that one on one. But if there's more than what they're focusing on, we do have those other services. One of the other questions I do see in the chat is wanting to know if you have 360. And no, they do not. So that is our other product that 3M has-- is our 3M 360 Encompass, inpatient and outpatient CDI encoding. No, this is CDI Collaborate. This is a community-based, primarily CDI program. They do not have computer-assisted coding. This is a different product. But what is-- when we talk about NLU, for those of you who either are on the phone that are familiar with 360, the NLU that we brought with us from the model side is now available in 3M 360. So all of the different libraries and the content conversations that we've had does apply to 3M 360, if you have that-- and so if you're wondering about that, we can please reach out in the chat, and we can get somebody to talk to you about how that NLU functionality that they're talking about is available in our other products. OK, so we're doing well on time, but I do want to talk about, let's look back just a second. What things did you-- what other things did you learn about this process and the selection and the implementation, in addition to Laura's role, what would you have to share with others about your journey? Gosh. Probably the biggest thing would just to be patient, I would say at GoLive, we were probably back to baseline within 60 to 90 days. I'm probably sooner. I'd have to go back and look at our data, and then we really started seeing the gains as far as efficiency and productivity, and just to pick your battles, I think I said earlier, Rome wasn't built in a day. And so we're not going to have perfection in every single metric that we're tracking and in every process at GoLive. And that's OK. Let's pick an area and focus on it and hardwire it and then move on to something else. So that would probably be my biggest advice. Do you have anything, Leila, you want to add? Not really, other than just be patient. It is a learning curve, then you're learning a new program. But just don't get frustrated. I have a tendency to get frustrated with myself. But it just took a little time and just use your resources that you have. Use your solution specialist. It just-- I would say, like Tara said, just the time, give it time and use the resources that you guys provide. Yeah. And we talked a little bit in preparation for this. And Leila, you shared with us that within the first 90 days was probably the hardest-- Yes. Time. And then you set it about six months, it was like, don't ever take this away from me. Right. It was-- right. It was like, OK, we're good now. This is going to work and this is strong. And I think that's why when you all keep saying be patient. It's very hard, especially change. Change is hard, whether it's good change or bad change. I mean, we all know that. But I think those are the time frames that you said that the first 90 days you're like, oh, my gosh. Everything is changing. What's new today? That's a long time. But by the six month mark, you were feeling pretty confident that this was going to be going to be OK. Like what's good? And more than anything, the software works as designed and as you built it right on day one. Really what Leila had mentioned when we were discussing this prior to hopping on today's webinar was really it's us working out our habits that we've created. So up to GoLive, and you just can't break all your old habits in a day. And then at the same time you know you've also got to learn to create some new ones. So that's really the biggest thing I think is just your processes internally that you've lined out and just making sure that you're walking those to account for them within your new operating system in that software. But yeah, I mean, by six months, Leila had it for sure. And what's great, another thing that's great about it is you can tailor the program and the software to your facility. You can set up your workflows, how you like them your work groups, how you sort things, how it sorts patients out. That's all at your fingertips. And if you want to something different one day and the other, it's all right there. And like I said, it's really tailored. You can tailor it to your specific needs and that has been really nice. So basically, the prioritization is based on opportunity, based on the NLU opportunities. The NLU, natural language understanding, which is built on natural language processing, so unless people understand NLP, we call it NLU, because we take in more concepts than basic level processing. So it's a lot more expanded to take in other things like, what medications are being administered, what lab studies, things that are a little bit more around clinical context or clinical concepts. That's why we call the NLU. But by using that, that is something, again, that takes that understanding of, what is it looking for? What is the engine looking for? And is it like what I was looking for? It'll tell you what it found, and it'll also tell you what it didn't find based on those clinical triggers. So there's a lot of getting you to understand as well. And it's a learning curve. When you were talking, reminded me of-- I used to work with the customer that was having a lot of change and had a lot of new staff and habits and getting through this implementation. And they had-- I don't know where they got it, but they had a huge cut out of Elsa from Frozen. And on their board it said, let it go. And that was their theme song, because you just sometimes, you just have to let it go and embrace what's coming. So I always think about that when people say it was kind of a joke, but it was a nice visual of, OK, I probably need to let this go, and it's going to be OK. I've got on white knuckles to this flow sheet, this paper flow sheet that I've been using forever. So we understand that and Laura's been there. And we've all been there. And so we're happy to get you to the other side. What would you consider your biggest win at this point? I think our biggest win is truly just having a solid program that runs itself. It does not require any as a manager, executive administrator, whatever your role is, of whatever has the most fires is what gets all of your attention. It's what's draining your time and resources. And I can definitely tell you that our CDI program is not one that drains any of my time or resources. And there are times where I truly don't hear about from Laura or Leila about any concerns, whatsoever. And we're just sharing wins on our monthly call. And that's great too. And I'd love to share those wins across our organization. So for me, as an executive director, that's truly a measure of success. And then as well as all of our key performance indicators that we track in our coverage rate productivity, efficiency, our financial impact, case mix, CC, MCC rates due to query. All of those have just, I would say, many of them have probably quadrupled since we went go-- since we went live or when we were on manual-based processes. So and again, we haven't increased our staffing. We still just-- we still just have Leila as our one CDIS. And so we have really, really been pleased with the product. It's very user-friendly, and it is certainly a tool that we rely on daily. Leila, anything to add to that? No, Tara really did a great job of summing it up. I mean, for me, as a CDI specialist, just the amount of productivity. Like I said, I'm actually looking at charts. I'm not inputting everything into Excel, their account number, their medical record number, all that. Just the productivity and then the financial impact that it's had. Moving forward, as a partner with us, what's next for your program? So one thing that we have in our strategic plan this year is outpatient CDI. So we're definitely currently analyzing and looking at the feasibility of that and where we would want to focus our efforts as well as what the return on investment would look like for an outpatient CDI program and certainly looking to 3M as a partner with Collaborate and seeing what additional resources we might want to invest in. We also are looking at our artificial speech recognition software that our providers use daily. They have asked us to just really look at all vendor solutions and see what's out there, and there's certainly some benefits, if we are able to use Fluency Direct in conjunction with Collaborate, just to enhance our CDI and documentation efforts simultaneously. Yeah. And there's a great question that there from Laurie. With the system-generated query, do the providers ever get query fatigue? So I think there's a little bit of confusion when we say it is automated and it helps with the NLU. These are traditional written queries. These are not anything that-- Leila does not review, build, and generate. So Leila is composing these with the assistance of our NLU evidence that is found in the record. So what Tara just described with having the Fluency Director Speech Tool, it is a speech-to-speech technology also houses the vehicles to which we can send our queries. So we can do electronic queries to the Fluency Direct toolbar, which gives them another way for the providers to interact when they're composing a note. They can actually use that toolbar as a vehicle to open electronic queries from Leila. So that's what would be the connection, and we actually, the brand name is CDI Engage One. It is offered with 360, but it is also offered with CDI Collaborate. So that next iteration of being able to generate-- to create queries in Collaborate and send them electronically, not to the Message Center in Meditech, but actually sending it to a toolbar where providers would be-- where there would be actually composing their notes and actually doing the work. So I just wanted-- these are not automated queries, they are touched by Leila. She's actually composing them, generating them, using her templates or our templates that we've shared with them, if they didn't if they wanted to use ours. But we do have some automated type of opportunities. We call them nudge. So just a little bit different. So I just wanted to-- since we were talking about Fluency Direct. In the next steps, I thought I would just go ahead and answer that question and do some clarifications for the group out there. So at this time, I'm going to thank you all for your participation. We still have another few more minutes. So if there are any questions from the audience that I didn't get to, I mean, I know it's a little-- I did it a little differently, but I was able to see them pop up, and I thought it made sense to address them while we were having the conversation. And then ladies, if there's anything else you'd like to add at this time? Is there anything that we didn't cover today that you were just dying to get out there? Or anything that I had forgotten to talk about? I can't think of anything. No, just to what you said, Kathy, they are not system-generated queries. And it truly is Leila that's creating that query seamlessly through the interface. So she's able to query and collaborate, use all the tools. I love the word that you said, as far as it's a nudge. That's exactly what I would say that NLU is truly a help. So but you get to pick. You can take it or leave it. You don't have to use it. And so that query is then sent by the CDIS through the EHR. The provider receives it in the EHR. And then once the provider has responded, then it's right there in the medical record for everyone to view. So Lisa, I think we've gone through all of our topics for today. Are there any other questions out in the audience that I can't see that you might want to know? No. You covered everything. This was a really great conversation and great webinar. Definitely thanking our speakers. I think it's really great to hear the actual users. And I think it's always good to hear that we're not alone in our sometimes our struggles but also our wins as well. And so I think it's just a really great perspective that they were able share. Real quick. One question, actually is I was starting to say that, did come in and I think this would be a good one. Do you want to take that Kathy? Sure. The question is from April. She says, is there a benefit with this system for medical billing and coding? So Tara and Leila, I mean, do you guys want to-- I mean, I have an answer but I'll let you guys share from your perspective? Yeah, absolutely. So certainly, we know the CDI benefit that we're all familiar with, as it relates to medical billing and coding is that there's an impact for our DRG right and those CC and MCC capture rates, which can make a difference in your reimbursement. So from a financial perspective, there is certainly a benefit. But also, with medical billing and coding, we use 3M for an encoder as well. And so all of those resources are embedded together seamlessly. And the communication between the CDIS and the coder, which Leila can probably speak to, is seamless. And we're also able to get our charts coded and finalized more timely, just due to being able to query electronically. Leila's not having to filter an Excel, scroll through rows and a spreadsheet, and say, well, gosh, what is this account on hold for? That sort of thing. We've got all that available to us right there in Collaborate. So if you think about CDI, and it's truly keeping like your rev cycle flywheel spinning, if you will, Collaborate is what's helping keep that spinning, so that we can get those charts finalized more timely. My take on that question, just to follow up. I think that was on my mind too about the whole communication and visibility and all of that. But also, yeah, CDI collaborate was built specifically for CGI specialists to be working cases and looking at reviews concurrently. With that said, when you turn it on, it reasons over the cases that come in into the patients that come into the system, whether Leila opens it or not. So I mean, it's going to be there. So if, let's say Leila didn't get to that case for whatever reason, she took a day off, have mercy, just give her a day off, the coder can-- if they go to code the record, and they have a question, the coder has a couple different ways we can set this up for customers. The coders can have access to the tool, and they can look it up themselves. They could go in there. They could see all of the findings from the NLU. They can create queries. So we can have coders. We can have that work in the tool. It is not a coding platform, which means that the CDI Coding Summary is a separate instance of 3M Co-finder, Encoder, then what the final coder would do. So they don't share code. But we do send, if the coder doesn't have any questions and the final code it we have an interface that goes back into Collaborate, so that when you're reconciling, you can see that final DRG populate, and it compares to what Leila was her working her final working DRG, and then that's what the reports are generated on, and we show impact of her work compared to what was final coded and such. But coders can do concurrent coding. We do have a concurrent coding workflow that some organizations use, that they can assign a concurrent coder, that they can look at it, give feedback and interact. But the two systems between the coder and the CDI are separate, and they don't necessarily share codes. That product is 360 Encompass. So that's a whole different product, if you want that level of integration. But in this service, this is where it's separated. But we do have customers that do have coders that work within Collaborate, but the final coding is in another system. Awesome. No. That was great. I'm glad that question came in, because I think it was a great-- Great question. Yeah, great question. So yeah, thank you for that. So again, today was really great. The perspective was awesome and being able to bring both Golden Valley together but also having Laura join us as well to give that perspective too. So thank you all very much. So as a reminder, check out the resources. We do have a few other on-demand webinars available that we've recently had. So definitely take a look at those. Those are again our on-demand webinars on our website open to anyone, as well as a case study. And then if you would like to learn a little bit more about that, the CDI Collaborate, there is a fact sheet as well. And if you are interested in learning more within the portal, there is a Learn More button. So if you would like to learn more, go ahead and let us know there. And so again, please fill out that survey for us. Let us know how we did. And again, we appreciate our speakers today. And we look forward to hosting you all again. So thank you, and definitely, thank you to Kathy for being another-- doing another great moderator for us for webinars. So thank you. [INTERPOSING VOICES] Go ahead. I said one more comment. AHIMA is coming up. So any of you listening, if you're going to be at AHIMA this Sunday, Monday and Tuesday, please come by the 3M booth and come talk to me. I will be there and would love to meet current customers, new customers, anybody on the call that wants to come by and say hi. Yeah. That is an-- yeah. That is great. I'm so glad you mentioned it. We were just talking about how excited we're going to be to see each other in just a few short days. So yeah, if you are attending AHIMA, safe travels. We're at booth 447. We have a couple tech demos. Kathy is actually going to be presenting. So yeah, definitely stop by and see us. We'd love to meet you. So all right. Well, everyone, have a great rest of your Wednesday. And thank you again. We look forward to having you on our next webinar. Thank you. Bye. (DESCRIPTION) Kathy waves.

(DESCRIPTION) Logo, 3M Science. Applied to life. Text, The approach to better track patient safety in the outpatient setting. Ambulatory Potentially Preventable Complications. A.M.-P.P.C’s. Danielle Bowen Scheurer, MD, M.S.C.R., Chief Quality Officer, Medical University of South Carolina. Sandeep Wadhwa, MD, M.B.A., Global Chief Medical Officer 3M Health Information Systems. Miki Patterson, PhD, N.P., Product Owner, 3M Health Information System. July 19, 2023. (SPEECH) Welcome, everybody to today's webinar-- The Approach to Better Track Patient Safety in the Outpatient Setting. On behalf of Becker's Healthcare, thank you so much for joining us. Before we begin, I'm going to walk through a few quick housekeeping instructions. We're going to begin today's webinar with a presentation and we'll have time at the end of the hour for a question and answer session. You can submit any questions you have throughout the webinar by typing them into the Q&A box you see on your screen. Today's session is being recorded and will be available after the event. You can use the same link you used to log into today's webinar to access our recording. If at any time you don't see your slides moving or are having trouble with the audio, try refreshing your browser. You can also submit any technical questions into the Q&A box. We are here to help at any time. And with that, I'm so excited to introduce our amazing speakers today. We're thrilled to be joined by Dr. Danielle Scheurer, the chief quality officer at the Medical University of South Carolina, Dr. Sandeep Wadhwa, the global chief medical officer at 3M Health Information Systems, and Dr. Miki Patterson, product owner at 3M Health Information Systems. Today, we will dive into patient safety and discuss how ambulatory potentially preventable complications can help track safety in the outpatient setting. (DESCRIPTION) Text, Learning objectives: Recognize impact of potential updates to U.S. News & World Report, U.S.N.W.R. Best Hospitals rankings. Understand the current challenges and opportunities related to patient safety in outpatient settings. Develop a roadmap for identifying and addressing gaps in current outpatient safety practices. (SPEECH) Our speakers will plan to discuss potential updates to the U.S. News and World Report best hospital rankings, discuss the current challenges and opportunities related to patient safety in the outpatient setting, and identify a roadmap for identifying and addressing gaps in current outpatient safety practices. Before we go ahead and dive in, we wanted to ask you all a quick question-- have you or a family member experienced a complication after an outpatient procedure? Kindly respond to the poll question directly on your webinar console. We will wait about a minute or so for responses to roll in. (DESCRIPTION) Text, A. Yes. B. No. (SPEECH) Thank you to all those who submitted their response. We'll wait about 15 more seconds for additional responses. We're going to take a look at our poll results. So very interesting to see it looks like about 40% of people have had a complication after an outpatient procedure while 60% have answered that they have not. But without further ado, I'm going to turn the floor over to Danielle to go ahead and get us started with our presentation today. All right. Well, thank you so much for having us I'm Danielle Scheurer, chief quality officer at MUSC Health in South Carolina. Our flagship hospital is in Charleston, and we have several newer additions of community hospitals throughout the state. Just wanted to give a tiny bit of a background about MUSC Health. And I think our process for reviewing quality and making quality better might make more sense if you have a little bit of background. So we are the only tertiary care academic medical center in the State of South Carolina, and so our mission reflects that. So our mission is to preserve and optimize human life in South Carolina and beyond through our tripartite mission, so we are very much in academic tertiary care grounded health science center. So we work and we care deeply about all three. Our vision is really to lead health innovation for the lives we touch and core values, compassion teamwork, diversity, accountability, and innovation-- many of which we'll touch on today. (DESCRIPTION) Text, M.U.S.C. by the Numbers. The only comprehensive academic health system in South Carolina. (SPEECH) A little busy slide, I'm not going to read all these numbers. But this is sort of our elevator speech, if you will, for MUSC Health by the numbers, just to give you a sense of where we are and where we're operating. Right now, we're seeing about 1.6 million patient encounters a year. We are each year, getting more and more enveloped in accountable care. So working with payers to shift away from fee-for-service into accountable care, we think it's better care for patients and better payment models in general. So currently, we have about 120,000 covered lives and aiming to grow that by the day. All told, our organization employs about 25,000 team members, 1,400 physicians. We have 16 hospitals, 2,700 licensed beds, and this is all in a milieu of 6 training colleges, where we have almost 1,000 trainees and over 200 students. So it's big complicated academic tertiary care health science center. And again, the only one in South Carolina. (DESCRIPTION) Text, M.U.S.C. Impact. A state map appears with regions outlined. Colored dots fill the regions. Blue depicts clinical locations, orange depicts research locations, and green depicts telehealth connected counties. (SPEECH) So this is just a visual depiction based on our tripartite mission. So places that we have clinical presence in blue, research locations in orange, and again, looking to grow that for obvious reasons to get people access to research opportunities throughout the state. And then we also have a very active and growing telehealth network. We're one of two centers of excellence in telehealth in the country. And obviously, that's been a game changer, I think, for us as far as staying connected to, in particular, in the rural areas of South Carolina of which the vast majority of South Carolina is rural, but really allowing access to care to patients who otherwise would never be able to drive to a physical MUSC Health location. So we're very proud of that telehealth network. (DESCRIPTION) A bar chart appears for the Ambulatory surgery center growth. (SPEECH) As you all well know, the growth in ambulatory surgery and ambulatory surgery centers throughout the US has been astronomical. It is a great site of care for a lot of ambulatory procedures. So to get the cost structure down, especially for procedures where the expectation of complications is low and the expectation for admission is low, those are typically better cared for in ambulatory settings. This graph is, obviously, a graph of the US and the specific types of ambulatory surgeries, ophthalmology, orthopedics, et cetera. But it looks very similar to the graph at MUSC Health, and I'm sure many other sites that are experiencing expansive ambulatory growth and flat, if not shrinking, inpatient surgical growth. And so this is an area where we're really excited as a quality team to really dig our teeth in and better understand. This has been a little bit of a blind spot, if you will, in monitoring, measuring, and really understanding what these patients experience as far as preventable complications. And that's why we're so excited about this partnership to better understand, analyze, and improve upon potential complications in the ambulatory space. (DESCRIPTION) Text, A sea of rankings. Leapfrog Hospital Safety Grade. The Joint Commission Hospital Inpatient Quality Reporting. I.Q.R.. Program Hospital Value Based Purchasing Program. Accountable Care Organizations. Best Hospitals U.S. News. Rankings. Hospital Compare. Hospital-Acquired Condition Reduction Program. Quality Payment Program. N.C.Q.A., C.M.S. Hospital Readmissions Reduction Program. Healthgrades. (SPEECH) So MUSC Health, just like all other health centers, are paying more and more attention to what I call a sea of rankings. Many of these rankings have been very focused on inpatient potentially preventable complications, such as hospital-acquired infections or patient safety indicators, et cetera. And really the vast majority of the quality metrics are still very inpatient-focused when the vast majority of care happens outside the walls of the hospital. So again, another good reason to figure out better ways of understanding what patients are experiencing in the ambulatory space and really encouraging some of these rankings programs to get more into the business of ambulatory complications rather than just focusing on the inpatient space. (DESCRIPTION) Text, Current opportunities and challenges. Over 70% procedures performed in outpatient settings. Limited outcomes/safety/quality measurement systems. Share results of a comprehensive quality outcomes system for ambulatory procedures. Trusted Objective Secure Data Methodologies. (SPEECH) Thank you, Danielle. Sandeep Wadhwa from 3M. I'm a chief medical officer. I think I'm at the altitude extreme from Danielle. I'm joining you from Denver, Colorado. Danielle, sometimes I think Charleston can run sea level or under sea level. So between us, we've got the span of the US covered from an altitude perspective. Very excited to share more about our partnership and also our efforts on providing insight on quality outcomes for outpatient procedures, both those that occur at an ASC setting, but also hospital outpatient and the system we're describing here can also look at office-based procedures as well. At 3M, we are focused on building methodologies, whether their payment methodologies or also quality methodologies. And we had been very aware that there was a gap in the market around measuring outcomes after procedures occurring in outpatient settings. And it's tricky to kind of be able to look at care after the day of the procedure and then link it back to the procedure and develop a quality system that-- where a follow-up event is plausibly related to the procedure in question. And so we've been working on this system for the past five years, maybe primarily interested in bringing a safety lens and looking at variation that's impactable or actionable in providing that insight by site and by the proceduralist. I'll also just quickly say we've been thinking about this issue with a very broad definition of procedure. That certainly we want to include the surgical procedures, but there's so much volume that is occurring from medical subspecialties, whether it's cardiology or GI or IR. And so we wanted to cast a wide net on being able to give health systems and providers, as well as payers and regulators a greater view of patient outcomes from this really important site of care. Next slide. Oh, guess I can control the slide. (DESCRIPTION) Text, U.S.N.W.R. updates. Inclusion of 3M Ambulatory Potentially Preventable Complications. Hospital Rankings Shift Emphasis to Objective Data Away from Expert Opinion. A time horizon table appears. (SPEECH) And as Danielle had indicated, we've been engaged or talking to the various public reporting entities. And U.S. News being a leader in that space just announced two weeks ago that they will be including evaluation of outpatient outcomes from facility sites, hospital outpatient departments for orthopedics and urology in this year's rankings of specialties. And they're looking to evaluate some more specialties and procedures for next year moving into OB/GYN, ENT ophthalmology, and some procedures, in particular, prostate, uterine cancer, hip and knee replacements. And so I think there's an opportunity here to kind of both advance insight on where things are going really well from a procedure point of view and spread those learnings across the system to be able to benchmark that, and then also look at where there may be opportunities. There's such a culture of quality and patient safety in inpatient settings. I so recall kind of surgical M&M, morbidity and mortality conferences. And I think there's a whole host of reasons that we don't, I think, have that same infrastructure for outpatient care. And so we're excited about being part of sharing some insights here on complication, safety, event rates, variation, and describe our relationship a bit further. Let me turn it over to Miki to walk through some results. (DESCRIPTION) A table appears titled, Top 20 PSGs 3M A.M.-P.P.C. rates. 19 to 21 Medicare F.F.S. H.O.P.D. (SPEECH) Sure. Hi. What I want to show you, I'm going to put the bottom line up front here so that you understand the data set that we started looking at. So if you look at the slide, I know it's really busy, but this is the top 20 procedure groups out of the 93 procedure groups that we have been looking at. And it's the data from Medicare hospital outpatient from 2019 through 2021, just so you kind of understand where we're coming from. So let's look at the top level. The number one procedure that's being done in hospital outpatients is an upper GI endoscopy procedure. And when you look at the data, you can start to understand how we're developing benchmarks. So there's 2 and a 1/4 million of these procedures that have been done in the past 3 years-- those 3 years. Of that, over 11,000 patients have gone to the emergency room with complications from an upper GI endoscopy procedure. 34,000 were unplanned admissions for complications directly related, or plausibly related, to that upper GI endoscopy procedure. They had the procedure, they went home, and then they returned to either the ED or were admitted. We're also able to surface all the complications that presented in an outpatient setting. When we looked at our benchmarks, we did not really want to utilize outpatient because that, one, may be the best place to treat that patient, and two, we're looking to see if there's a significant complication. So as we did our benchmarking, you can look at this last column, which is the complication rate. So having an upper GI endoscopy procedure, you have a 2.2% chance of having a complication related to that procedure. And to understand a little bit of the variation that we saw when we looked at these procedures and their complications, it made sense. So if you look at a routine cataract procedure, you yourself would think, there's not really a lot of complications that can come from just taking care of something in your eye. So as you look across, 0.21% chance of having a complication related to that procedure. So now look at dialysis shunt. We know that the patients are very sick that have to have dialysis shunt. They have kidney failure. So we know they're already sicker patients and they're having a procedure, which has a higher risk. So you can see this. Only 389,000 of these that were done, but 10,000 went to the ED with a complication from that shunt procedure, 16,000 were admitted. So when you look at that 6.75 rate, complication risk rate, that's how we come up with those rates. And I'll talk a little bit about the logic that goes behind it in a sec. (DESCRIPTION) A bar chart appears with the title, National Medicare risk-adjusted O.P. facility-based procedure complication variation, C.Y. 2021. Distribution of hospital A/E Medicare F.F.S. performance C.Y. 2021. (SPEECH) But because we could look at 100% claims data for all of those years, what we did look at was at a hospital outpatient level, there were some 3,261 of these different hospital outpatients. And they had to have done more than 200 at-risk cases, meaning one of the cases that go into our procedure groups. And we could look at what their actual complication rate was compared to what we expected with those rates that we just showed you. So now we have a way to kind of compare apples to apples across hospital outpatients. And you might see in the middle of this that the highest rate is less than 1. So understand first, that we don't believe all complications are preventable. They're all potentially preventable complications. But we also now have something to aim with. And when we started doing this work with inpatient potentially preventable complications, this is kind of what the curve looked like. But you can see there's longtails to the right and they're little bit to the left. So these hospital systems on the left-hand side are doing better than expected. So there's a lot of difference going on. So when you see variation, this is where you can impact it. So a little bit about what it is. So the definition of an ambulatory potentially preventable complication is a harmful event, such as a laceration accidental or a negative outcome. And I think this is where you're going to see a lot of action-- sepsis infection, hematoma, bleeding. They have to happen after an elective procedure, somebody scheduled this. So this is not emergency room procedures. And it has to be performed in an ambulatory care setting. And what we're really looking at is complications that are clinically plausibly related to that procedure itself. We're not looking natural progression of a disease like, oh, their hemoglobin A1c, their diabetes is out of control. We're not really looking for that in this component. So here's the logic, how did we make this? So we looked at some 2,900 different procedures with what we would thought had a risk. So they had a needle, a scalpel, intubation, some kind of medication, something was done to a patient that could potentially result in a complication. And we started anatomically. We started looking at things that can be done in the fingers and the hands and the elbows and all anatomic. And we looked at the risk of the procedures. So thinking about the hand, if you had a trigger finger done, one little finger, quick surgery, not a lot of complications kind of like the eye. But if you had a total shoulder done and you think about that, that's a longer procedure, there's a lot more blood loss. It's a very complex and higher-risk procedure. And we're able with this component to bubble up to the top those procedures that we have designated as the highest risk if you had more than one procedure done at a time. We can also look at service lines. We can look at facilities. We can look at a lot of different things. But that procedure is now chained out for 30 days and we're going to look for complications, so then there's logic that goes to the complication. So we divided it in some 1,500 different complications into 35 complication groups because we wanted to put like with like. And you'll see that there will be lung things and there'll be bleeding things, but we wanted to make sure that it was clinically plausibly related to that procedure. And we put on timing limits. Do we expect this to really be from that procedure at 20 days? Maybe, maybe not. So we have timing limits built into this software. We also wanted to see where they turned up. So in the software, we have four different complication types. The first is if they showed up in the emergency. The second is if they were an admitted patient. So that was present-- on this complication was present on admission. We also said we can find them in outpatient and we have a sneaky, little logic that's probably not going to be used much and we found a few results to this. But in Medicare, there is a rule that if you had a procedure and you were admitted within 72 hours of that procedure, you can't bill as an outpatient, it gets rolled into the inpatient. And we have logic that can find that. And when we did this, we made our Medicare and national benchmarks based on emergency and inpatient only, but we can report out the outpatient. (DESCRIPTION) Text, Risk adjustment approach. Age. Disability status. Oncology. Correlated with chronic conditions. (SPEECH) So you say, well, what about the risk? Did I take a sicker patient? My patients are always sicker than somebody else. Again, we are not looking at that patient's condition. We did find when we looked at all procedures that age was a factor, so we built that risk factor if you're over 75 or if you are over 85, you had a higher risk of a complication. Makes sense. We also find that under 65 with our Medicare, which is our disability status, they had higher risk of complication. And oncology-- despite thinking that oncology would have a higher risk down the line, what we found is in our 93 different procedure groups, there were only five that oncology had a higher risk. And it's also associated with chronic-- the numbers correlated well with chronic conditions. And I give you the example of that dialysis shunt, that is a chronic condition and they were all put in the same kind of situation. (DESCRIPTION) Text, Ample of complications. Lung issues, bleeding hematoma, infection sepsis, mechanical complications. (SPEECH) So this is just a sample of the complication types that we have for our groups. The lung issues, so we have aspiration pneumonia, pulmonary embolism. They're all falling into procedures. And we have bleeding, such as post-hemorrhagic and acute anemia, venous thrombosis. We also have various classes of infections that will go with specific procedures. And then really importantly, I want to show you that we have mechanical complications. But when we do this logic, a mechanical complication of, say, a muscle skeletal device is only going to be tagged to a procedure with a muscle skeletal device. So I just want to make sure that you understand we really did a lot of homework to make sure that the real procedures with complications really related to those procedures are what's coming to the top. (DESCRIPTION) A pie chart appears. Text, Top 10 complications. (SPEECH) So what we found was kind of surprising to me that out of all the procedures that we ran through, the number one complication coming to the ED or being admitted was septicemia or severe infection. And we did actually not know this. There is nothing that was able to aggregate all of this data before this. So UTI's, pulmonary and chest, other bleeding, so now we actually have some data on this. (DESCRIPTION) A bar chart appears. Text, Data by procedure subgroup P.S.G. 70 upper G.l. endoscopy. Medicare H.O.P.D. 19 to 21. (SPEECH) You say, so what? Well, this component will actually look at those upper GI complications and now tell you what they're coming in for. So you saw how many cases they went to the EDI inpatient, they have a 2.2% risk, of what? Of number one, coming in for a GI or peritoneal significant bleeding. So these patients had a procedure, they went home, and they came back with bleeding, or pulmonary and chest complications, or septicemia. So these are the ones they are coming in for. Now, we actually have some insight into what's going on. (DESCRIPTION) A table appears titled, Top 10 H.O.P.D. ortho P.S.G's. (SPEECH) And as you say the same thing, I can do it by service line. So look at these. These are the top 10 procedures-- orthopedic procedures done in hospital outpatient in the past 3 years. At the top is a total knee arthroplasty. Now, I'm going to show you some information. There's a lot of these cases. We have a 2.8% complication rate. And again, you can say, so what, Miki? Well, I'll tell you. What are they coming in for? They're coming in for anemia. (DESCRIPTION) A bar chart appears. Text, Complications of P.S.G. 13 total knee arthroplasty. Medicare H.O.P.D. 19 to 21. (SPEECH) Now, remember, total knees were in the hospital for two weeks, and then one week, and then three days, and then one day. And why do we keep them? We looked at their labs. We looked at their mobility. We looked at everything. And if they had some bleeding, we would transfuse them. Now, we're able to do a lot more things with trans-cinnamic acid and tourniquets, and all the ways that we're doing these minimally invasive procedures. However, if that patient came back to the emergency department with anemia, it was palpitations, they were fainting, they had fatigue, there were other things, and they would be treated by medicine and that surgeon may never know how many patients came back with anemia-- unless the patient came back, but that's just a little anecdote. (DESCRIPTION) Red arrows point to the three most common complications. (SPEECH) Oh, I'm just going to show you the places that we see here, you can start to look at your procedures. So now that we know they came in for anemia, UTIs, or infections-- what are we going to do? Look at all of these people with infections. What can we do about that? (DESCRIPTION) Text, Prevention: Post hemorrhagic anemia, acute anemia. Best practice. (SPEECH) Well, we can go back and look at our processes. They came in with anemia preoperative. Did we look at them preoperative? What were their medications? There were a lot of ways that we as 3M can start to highlight these areas for improvement. This is actionable data. (DESCRIPTION) A bar chart appears. Text, Driving network performance on complications, site of service. The best performing provider was 70% better than expected and the worst performer was 70% worse than expected. (SPEECH) And the same way of looking across your whole network. When we start to put the facilities in here and find out how many complications are coming in, this is state data. And if you look at it, the middle, the highest, is 0%, it's not zero complications. It's just where we expect them to be. Everybody on the left, the far left, that 70%, that's an orthopedic ASC that did 70% better than expected for outcomes. If you look to the far right, that's an orthopedic ASC that did 70% worse than expected for outcomes. So wouldn't this be good to know because you would concentrate your performance improvement on that right side of this graph. So I think it's just really, really good at showing things that we actually never had insight to before. (DESCRIPTION) Text, M.U.S.C. quality governance structure. Improve health care quality. Improve KPIs vertically and horizontally. Focus on sustained process improvement. Improve internal and external reporting and transparency. (SPEECH) So thank you, Miki. That was great. So you can see there's just an incredible array of data available through the tool. And then the next question is for hospital, or health system, or ambulatory surgery center, or even a group of physician practices, how do you tackle the data when it comes in. This is just an example. So we have been thinking through this. So again, a lot of our quality structures in many of our hospitals and health care systems have really been focused on inpatient data and quality improvement projects on the inpatient side. And what we've tried to do over the course of many, many years is extend our service lines into what we call ICCE, which are Integrated Centers of Clinical Excellence. So we get our ICCE leaders thinking throughout the spectrum of care, not just inpatient. And now they're really craving more ambulatory data, which is as you've just seen, we'll be getting to them soon. So what we've lined up at MUSC health is that each of our ICCE/service lines, extended service lines, have a quality safety manager or director depending on the scope and complexity of the ICCE that do a whole range of activities for that ICCE. So they certainly look at risk safety quality, they do external reporting, they do data analysis, they work with their ICCE leaders, they work with their operational partners. And more and more getting into all of our value-based care to make sure that the care of that patient in that service line/ICCE is appropriate across the spectrum of care. So really thinking about the longitude of care and not just an isolated area. (DESCRIPTION) Text, M.U.S.C. quality governance structure. (SPEECH) So all of our quality leaders report up to a system director, and we give them guidance on what it that is supposed to be in their strategic plan. So they look at system-level key performance indicators, so big things like safety, mortality readmissions, but again, a lot of those current state are focused on inpatient. They also look at very ICCE-specific key performance indicators. So for example, in the peri-op space, there's lots of KPIs around process. So making sure that patients are getting CHG bathing and that their timeouts are appropriate. And all of those kind of bread and butter safety checks that we do both in and outpatient settings and peri-op. Many population-based attributions like for the population in that ICCE of the patients who have diabetes, is their diabetes controlled? A lot of safety-related things-- so event reviews, hospital-acquired conditions, and anything coming through our event reporting system, where we do a common cause analysis, so again, regardless of whether it's in or outpatient. And then across the spectrum across the longitude doing more quality and performance improvement activities based on other key performance indicators that are specific to their ICCE, up to and including external data and registries. (DESCRIPTION) Text, How will we close the gap? (SPEECH) So in the example of ambulatory surgery key performance indicators, so you saw from Miki the breadth of data that is available through the tool. So really working with each one of those ICCE to make sure they understand. Just like Miki explained, when you say complication, what does that mean? How do it's related? What are the time, stamps? How do you risk adjust all of that? And then understanding for patients that are having a complication, understanding what is a typical workflow, ideal workflow, standard operating procedure for that area, and making sure that the team is following those key processes to avoid complications altogether. Things like medication reconciliation, again, CHG bathing, things that should be occurring in the ambulatory surgery space, but may not be occurring reliably. Analyze those process gaps and then put them into an improvement process. So again, looking at complications very specifically by population, making sure that we're getting that data analysis back to the teams that can have an impact. Going through those performance-- key performance indicators every week, making sure that if there are barriers to success, that those are getting escalated up through the QAPI team. And then again, revising workflows, standard work, or standard operating procedures as necessary. And then obviously, incentivizing and motivating our teams for success. So that's how we normally do quality and process improvement, and how we're going to weave this ambulatory potentially preventable complications into our existing infrastructure to close those gaps. (DESCRIPTION) Text, Summary. Promote patient safety in outpatient settings. Great and growing need for ambulatory standards of care and safety. Identify actual versus expected variations. Source comprehensive, trustable, actionable outpatient methodology and data. Foundation to value-based care in rapidly expanding ambulatory care. (SPEECH) Thanks, Danielle. And just kind of in closing, I think just to recap the topics we've been discussing, I hope the audience is hearing how we're approaching this awareness-- situational awareness, historical awareness of patient safety performance in outpatient settings. And just I think this is so important from a patient experience, patient safety perspective. And I think as we're seeing, many of the folks that look at measuring performance from a payer or quality point of view I think are going to be looking more carefully at procedures. And I'll also just kind of comment that a lot of outpatient care is hard to measure from a quality point of view. I practice geriatrics, the time to see impacts for hypertension care and diabetes care are not short. One of the interesting things about procedures is we're getting this ability to look at this 30-day window, which is a lot easier to measure and is a very relevant recovery period and really are able to look at post-acute care, post-event care as being kind of a longitudinal period here, where I think it promotes itself for the quality improvement infrastructure that I think we're much further along on the inpatient side, and now can kind of extend that to where care is overwhelmingly being delivered now in ambulatory and community settings. What's so important, I think, is that there's an ability to see your performance versus benchmarks versus average and best practice benchmarks. And to unpack that further to understand where there's opportunities for leverage or improvement. There's benchmarks available for efficiency and throughput in outpatient settings, and I think we're really excited to be able to bring in a safety quality dimension. And I suspect that this will be a lot of interest to the physicians and providers that may be insights that they weren't aware of before-- 2% complication rate is non-trivial when I'm making referrals. For me to know what my institution and what different proceduralists rates are, I think may even help informed-- help even inform informed consent. And so we're looking to work with additional health systems as we are starting this journey with MUSC. And I'll also say that I think that this will feed forward into a lot of value-based contracts, that there's an interest in how payers pick the providers that they want to work with in a neural network. And I also think as a health system, knowing your total cost of care and potentially preventable events and how you're performing kind of can inform your negotiation positions in these increasingly downside risk relationships. And so as we kind of move from upside sharing only to downside, I think having insights on relative performance become really important. So thank you. I think-- Clare, I think we wanted to save some time for questions. And I'll let you tee those up and we'll distribute them amongst ourselves Yes, perfect. Well, I would just like to thank all of our presenters for an excellent presentation. And yeah, we will now begin our question and answer session. So you can submit any questions you have by typing them into the Q&A chat box on your webinar console. And we'll go ahead and get started with the first few questions here. So just to kick us off, is there any data breakdown between adult and pediatric outpatient-- outpatient outcomes? Danielle, Miki, may I take that one? Yes, go ahead. Yes. So a great question. I'm really glad that came up. We did design the system to be inclusive of pediatric care. (DESCRIPTION) Text, Q and A. Danielle Bowen Scheurer MD, M.S.C.R.. Miki Patterson PhD, N.P. Miki dot Patterson at mmm dot com. Sandeep Wadhwa MD, M.B.A. S Wadhwa at mmm dot com. (SPEECH) And so U.S. News is focusing on Medicare for their rankings as they are when they're other-- when they look at mortality and discharge status, they're working with a Medicare database. But we designed the system to be inclusive of pediatric procedures. And what we are seeing is lower overall procedure volume, but a very similar rate of complications. Miki, you can correct me if I'm wrong, but my recollection is it was a shade below that aggregate 2%. And obviously, the mix of procedures for kids is different. All the newborn-- not all, but many of the newborn males are getting a circumcision, so that dries up volume in terms of births. But just when we look at ear, nose, and throat procedures, both whether it's ears or tonsils, or the frequency of hernia repairs, or GU procedures, we saw a fairly similar complication rate and we just think that will be of a lot of interest to pediatric sites of care, whether those are outpatient or ambulatory, and so the system is inclusive of pediatric cases as well. Miki, anything to add on peds? Yeah. So for this data, I just presented the large volume Medicare data, but we do have national benchmarks. And as Sandeep said, when we looked at, say, the lower GI or lower GU surgeries for youngsters, we saw that they were very similar to adults in almost every single area. Again, they were a little bit lower. The rates of complications for these younger people were lower, but they follow the exact same pattern. So if there was very few complications in the Medicare data, there were very few in the Medicaid and youngster deaths. So yes. So this next question is a bit of a two-parter, but how important are patient partnerships in developing metrics to measure health system's safety outcomes in ambulatory care and overall health, and how will your hospital system engage patients in this process? I can take that one. Incumbent upon quality leaders to develop and put in place metrics that are meaningful to patients. So if a patient-- if we're measuring something that they really-- that has no meaning or impact on them, it doesn't really make any sense at the end of the day to put a whole lot of effort into that. I do think that patients, obviously, feel and care about complication rates. I do also think that most patients do not expect to ever experience complications much less in an ambulatory setting. And again, the sheer volume of ambulatory procedures is so high that the vast majority of patients don't experience a complication. You saw those numbers. So even though the is low because of the volume of people getting those things done, cataracts, as Miki pointed out, upper GI endoscopy, et cetera, that the volume of people experiencing complications would be high. I think most health systems do have patient family advisory councils, where they do get feedback on implementing new procedures. So I think one area of opportunity in this space would be to help or have our patient family advisors help us on the pre-procedural or education side to help patients understand what the risk of the procedures are, again, that for the most part, I think this is pretty new information to both-- would be new information to patients and certainly on the health system side. So to get their help in making the information understandable from a patient perspective. Miki and Sandeep, do you want to add to that? Yeah. So one of the things that I think there is an understanding that if you're going to have an outpatient procedure, it's no big deal. It's I'm going to have a procedure, I'm going to go home. And one of the things that this is showing is what complications could happen. So as the patient goes home and thinks it was no big deal, if you pre-teach them, preoperatively teach them, OK, if you have dizziness, if you have this, if you have that, please call the office. Please be seen right away, because they may have some symptoms of a complication and may delay care. So I think this actually just brings out if patients are coming back with something, you can teach them they'll call your office sooner, or you can take care of that beforehand. I just think having patients aware of this potential of complication and what symptoms to look for I think is going to be helpful. And I'll just build upon those comments that we think there's a real dimension of patient empowerment. A lot of times when patients are making decisions about where to go for a procedure, you rely on word of mouth and kind of using your informal networks. And the more we can kind of contribute to adding a lens of some objective data to help inform decision making, I think that's a good thing. And just to Danielle's point, I think this notion of patient partnerships and voice of the patient is something that we're also welcoming in terms of feedback and areas for improvement and focus and evolution. So really appreciate the question. Are each of the complication types weighted the same or differently when evaluating providers? Miki, do you want me to-- Sure. The system kind of looks at four complications, four potential complication categories-- office-based, emergency room, inpatient follow up, and kind of direct admits. When you have-- within 72 hours, if you're directly admitted that we capture-- we have that as a measure. The office visit and direct admit are not part of our benchmark norms. We think those are important areas for understanding and unpacking why-- what's the cause and reason for those settings for follow up care. So we are in our norms file kind of looking at emergency room visits and inpatient as kind of our-- they're disaggregated but one can add them together as we did to kind of show that complication rate. And I think payers or quality regulators may have a different weighting system, but our benchmark is kind of looking at the emergency room and inpatient rates as being the two key events that are part of our norms file. Miki, anything to add on that. No, I think that just understanding that the complications that belong to that procedure are what is being followed, and it's not weighted differently by the procedures, except that these are the complications from those particular procedures and that's the complication rate that we're benchmarking. And it works for ASCs, it works for any-- it works for anybody's data that wants to go into that. And I think that's one of the things I wanted to make sure people understood that you can put whatever data in here, and it will bring out your specific information. Just Miki, I'll make one more quick comment, which is we are using indirect rate standardization. So if the only procedure you were doing was a cataract, we would not-- the software would weight the complication rate for that volume accordingly so that that center is not looking exceptional against a center that predominantly was doing hips. We're looking at the rates within the procedure type and weighting accordingly. And so if there is a higher complication rate with those cataracts that the weighting would bear that out in a comparison file. So what role has CMS played in monitoring complications secondary to ambulatory surgery? I got that. Danielle, you mind if Miki goes to bat on that? OK. So one of the things that we looked at is what are the CMS regulations for procedures, and what are the regulations for ASCs. And you'll see that they have-- a lot of them are procedure-based, meaning do they come back and follow up for, do they have vision improved for cataracts or colonoscopies? But they don't have anything comprehensive as yet. So I think this is going to be one of those eye openers and first steps of finally seeing that you can look into these in a really, really rich way. Yeah. I would add that Outpatient Quality Reporting, OQR, and ASQR differ. And that what we're talking about-- allowing here is a drill down by these procedure groups rather than an aggregate value. And that we're also looking at follow-up care that is a potential complication, not just all ER visits within a certain period. Danielle, any other insights about how you're using OQR CMS measures? Yeah. No, I agree. I think the-- I think the difference between current state and what we have here is just the level of specificity, and I think the data is much more credible because it goes to great lengths to link what complications would be related to the procedure as opposed to just overall things like readmission rates and things like that. So I think the level of specificity with the data lends a lot more credibility, and therefore, I think, we'll get more traction in process improvement from key stakeholders because of the credibility of the data. We have to win the hearts and minds of the providers. So the first reaction is this doesn't apply to me. Right. Every time. Moving on to our next question-- how do we tie together outcomes from procedures performed in the outpatient and inpatient settings? Miki, I can jump in on that. So two ways, one is we work with the health systems facility and physician billing data or out of the data warehouse, the clinical data warehouse, and are able to link together the procedure with the follow up care that's in the system. And as wide a net as we can cast, we can kind of connect the dots between events in time and space. And so that's kind of a path 1. Path 2 is brothers and sisters, when we can get payer data, we jump on it because the payers are able to see out-of-network care, and so the system also can operate with payer data. And many of our clients have risk sharing arrangements, and with that comes these payer claims data assets. And so those are the two ways that we're trying-- that we link the subsequent care to the incident precipitating procedure. And I believe we have time for one more question. What process does MUSC use to ensure the OPDs under the med center follow the PNPs. Can you repeat that? Yes. What process does MUSC use to ensure that OPDs under the med center follow the PNPs? So Danielle-- There were some acronyms there, I'm not-- I think the OPD would be the outpatient department, kind of the hospital outpatient department. OK, yes. So the facility. And then, Claire, that second acronym I could guess, but what was the second acronym, Claire? Yes. PNPs, I believe it's referring to regulatory agency regulation. So I think that's where our ICCE structure comes in handy. So we expect consistency in care despite location of care. We do run into this. Sometimes different clinics have different regulatory oversight. So we just take a single standard approach to what is the right care paradigm and what are the regulatory expectations regardless of if we're actually getting a site visit or not. So I think the ICCE infrastructure helps us standardize that across sites regardless of who's actually paying us a visit, Thank you for all of those great answers, and thank you to our audience for the great questions. Unfortunately, that's all the time we have for today. But I would just like to thank Dr. Scheurer, Dr. Wadhwa, and Dr. Patterson for an excellent presentation, and I'd like to thank 3M Health Information Systems for sponsoring today's webinar. To learn more about the content presented today, please check out the resources section on your webinar console and fill out the post-webinar survey. Thank you to everyone for joining us today, and we hope you all have a wonderful afternoon. Thank you. Thank you.

(DESCRIPTION) Presentation. Text, on 24 for a better webinar experience. Screenshot, a webinar interface with different boxes labeled slides, speaker bio, survey, live stream, resources, Q&A. Logo, 3M, science, applied to life. (SPEECH) Good afternoon. And welcome to our webinar where we're going to be going over our 3M M*Modal User Management system. Before we get started, I'm just going to go over a couple housekeeping items. We are utilizing the On24 platform. It (DESCRIPTION) Slide, housekeeping. (SPEECH) is a web based platform. So we do encourage you to use Google Chrome, close out a VPN or multiple tabs that will help with your bandwidth because this is a web based platform. We do not have a dial in number. So check your speaker settings. If you are having any audio issues, if you do a quick refresh, that usually fixes any of those. You can see that there are a lot of engagement tools. We have the media player where our speaker is going to be actually doing a demo of this platform in a bit. You have the presentation area. But then also we have the Q&A section. So if you do have questions, we encourage you to ask questions throughout. And then we also have a survey at the end. So we appreciate you letting us know how we did. And if you're interested in learning more, there's also a button there. So, again, in that Q&A section, please feel free to ask as many questions as you want and we'll get to as many as we can at the end. We do anticipate this being about a 45-minute webinar. And so after our speaker is done, like I said, just put those questions in the Q&A. So (DESCRIPTION) Slide, meet our speaker. Katie Canonge, product manager, 3M, M Modal fluency direct business. (SPEECH) about our speaker, we have Katie joining us today. She is our product manager with the Fluency Direct Business line. If you'd like to learn more about her, her information is in that meet the speaker section. After the webinar, we will be sending out some more information. And I'll also go over a couple other things at the end. So I'm going to go ahead and pass it over to Katie to get things started. Hi, everybody. Thanks so much for joining our webinar for M*Modal user management. I'm really excited to be here today with you to talk about our new administrative tool. (DESCRIPTION) Slide, what is 3M MUM? A bullet point list. Screenshot, an interface with a list of users and columns of data. (SPEECH) So I'm going to use our management or MUM is our version 2.0 of our Fluent Direct Administration tool that we have today. Just like FDA, it's a web based application that's used for user management and configuration. While FDA was focused on FD user management, MUM is focused on management across many different 3M M*Modal solutions. (DESCRIPTION) Slide, why change? A bullet point list. (SPEECH) So the reasons why we're changing to this new administrative tool, there are many. But the main one is for the pure reason, that when we first created FDA, it was really with Lindsay direct in mind. Since then, we've created a lot of new products here at 3M M*Modal. And with that, there was a need to consolidate user management across those solutions. For our internal resources, this new tool is great because it allows them to have better flexibility and control over the user management and configuration for our customers to really make them more effective and efficient with their workflows for our end users. From a development perspective, FDA had a lot of limitations. And with that, we were unable to bring a lot of new features that we really wanted to the application to give you guys some more advantages. And so with MUM, now, we have the ability to create new features for troubleshooting, user management, and even software distribution. (DESCRIPTION) Slide, see it in action. A bullet point list. (SPEECH) So today, I'm actually going to have us jump into the MUM application. So you guys can get a look and feel for all the new features that are coming your way soon. So I'm going to go ahead and share my screen here. And hopefully everybody can see. All right. So this is M*Modal User Management or MUM as we like to call it. You'll (DESCRIPTION) Slide. Screen share, an interface with a table of users and columns of data. Along the left is a side bar with the text Academy Hospital, users, groups, reporting, log out. There is a table in the main window with the text, manage users. There is a button to add a user and a button for bulk actions, as well as a search bar. (SPEECH) notice on the left-hand side, we have a main navigation menu where you can quickly navigate and access different categories within the user management tool. Really briefly, you can see you have organization management, user management, group management, as well as reporting available to you through this menu today. This menu is collapsible, which is really nice because it can give you more real estate on your screen, especially important for admins who might have smaller monitor sizes and really want to be able to dive into the content that they want to see, and not have to deal with a menu. We're going to go ahead and take a look at the organization management or overview. (DESCRIPTION) Selects Academy Hospital from the left sidebar. Text, manage organizations. A table with five columns labeled organization, organization ID, users, groups, edit. Academy Hospital has three items in its drop-down menu, emergency medicine, internal medicine, orthopedic. (SPEECH) Here you'll see a snapshot of what's going on with your organization at a high level. And the bottom left-hand corner here, you'll be able to see the total number of organizations that are created for your particular facility, as well as the total number of users created across all of those individual organizations. Taking a look at the table here. What we've done is we've broken out each organization level. So you can clearly see how many users are added at each level, as well as the total number of groups that are created, as well as a preview of those groups. These also act as a quick navigation buttons to take you into those details, to dive into the specific users or groups as needed to modify. The other thing we have here is a search option, which is really nice if you have a large organization with multiple different levels. It can be really cumbersome and FDA. Just scroll and find what you're looking for. Now, you can quickly just put in what you're looking for into the search option and find it from the menu choices. (DESCRIPTION) Selects the edit icon beside Academy Hospital. (SPEECH) If you have a need to edit a profile at any time, you can simply click on Edit and it will take you into what we call our organizational profile. (DESCRIPTION) There are five tabs labeled summary, basic settings, user settings, administrative settings, downloads. The summary tab is highlighted. (SPEECH) From here, you'll see that we have a summary view that gives you all of the details related to this particular organization level. Users can scroll and then modify as necessary. We also have the additional tabs for quick navigation. You can easily jump to the content that's relevant to you. So really end user preference, but giving you a couple different options. Under organizations and new to MUM is downloads. Also the Downloads section houses fluency direct install files for ease of use and self-service for admins. This is really nice because you no longer need to come to us at 3M M*Modal in order to grab an installer. If it's an existing one, that doesn't need to be modified. It can be really easy to use a search option to find the one that you're looking for or even use a different source in order to find the date of last modified. That might be relevant to what you want. From this table, admins have the ability to either download an installer directly to their workstation or they can copy and create a link that they can send to a technical resource if they're doing fluency direct deployment or something of that nature. Today, this table houses fluency direct information. But over time, the intention is to expand this to additional products. Moving out of the organizational profile, we're going to take a look at user management. (DESCRIPTION) Selects the user's option from the left side. (SPEECH) So one thing you'll notice right off the bat is that all users are available through this table. We no longer have the paginated view where you have to flip through the pages in order to find an additional set of users. This also shows you all users that are created across all of the different organization levels. This is really nice because previously with FDA, you had to click into each org level in order to maybe get an exported list of users and then compile them all together. Now, admins can simply go up to bulk actions and generate a report of all of their users in one easy step. (DESCRIPTION) Clicks on bulk actions and a drop-down menu appears with the option, among others, to export all users. (SPEECH) You'll also notice with the table that we show the total number of users and the bottom left just to give you a quick glimpse of how many users are being displayed. We (DESCRIPTION) There is a button on the bottom right side that reads show limited users. (SPEECH) also have the ability to toggle on and off limited user profiles. This is really nice if you're looking for a fluency for transcription profile that you might want to convert into a full fluency direct profile. Admins can either toggle this option on and off as they see fit. Moving back into the user management table. You'll notice that there are several columns here. And at a high level, you can get a glimpse of what's going on with the user without necessarily needing to dive into their profile. Looking at each of the columns, you can see that there are sort and/or filter options available now to really help you look deeper into your users and find content that is relevant to you. For example, I can apply a filter for a user type. So (DESCRIPTION) Selects the user type column. A menu appears with a list of items with checkboxes and an apply and cancel button. (SPEECH) if I only want to look at physician users and maybe clinical support staff, I can go ahead and check those off, apply. And it will apply the filter for this particular chart. We do keep track of all the filters that are applied across the information. So you can easily see what is there and then clear them as necessary once you want to see all users again. (DESCRIPTION) Text, you have one filter applied. Clear all filters. (SPEECH) Looking at the different columns, there's two that are new to MUM that are really helpful. Groups is the first one. You can now see at a high level the total number of groups that a user has applied to their profile. By hovering over, you can quickly see a preview of those groups. And this is really nice if you're trying to compare multiple users and just get an idea if maybe a user is missing from a group and need to be applied to get those special user settings. The last column is last activity. And this is really nice because it gives you a high level, just quick snapshot if the user has used fluency direct or their profile in general within a certain amount of time. So it'll tell you the date of last use. And this can be really nice if you're trying to review user profiles in order to see are they no longer with the organization. Potentially maybe we want to move this license to a new user. Or even for benefits of follow up adoption, you can use this information to say, hey, user A is struggling. They haven't used it for a while. Let's try to re-engage. With this menu, you'll see in the top right, you have the ability to quickly search. So if I'm looking for a particular user, I can go ahead and type in my results. And it will quickly filter and show me like results to what I typed in. Also, a part of user management, we have our bulk actions. As I showed you before, we have the dropdown right up here in the right of the menu. One nice thing that we have with MUM is that you don't necessarily need to create a CSV file of your users in order to complete bulk actions for them. Now, through the checkmark options on the left, you can go ahead and just check off a user that you want to modify or several. The system will keep track of the users that you've selected. And then from there, you can go up to bulk actions. Quickly see how many users you've selected. And then pick a bulk action that is applicable that you'd like to perform. (DESCRIPTION) Items in the bulk actions list, export users, enable users, disable users, expired passwords, change settings, delete users, batch update users, batch upload custom attributes. (SPEECH) A lot of these are aren't new. They exist in FDA. But one that is new is change settings. (DESCRIPTION) New screen. Text, edit users in bulk. (SPEECH) So if I click on that, I can now make some quick changes to these three users if I would like by modifying either user type, medical specialty, and even applying groups if I see fit. Once you're done, you can clear your selections and then move on. (DESCRIPTION) Selects add user button. New screen. Text, add a user. A form for data to be filled in. (SPEECH) And the last thing we have in the user management section is being able to create an individual user. In FDA we had multiple forms that the user had to fill out as part of the user creation process. This could be a bit cumbersome. And so what we've done is added everything into one form. So now, you can quickly see the fields that are required by the asterisks and fill out all of the necessary information before creating the user profile. You can even now add users to groups. Once you've selected their organization level, those groups will be available in the mover. And you can go ahead and add them over once they appear. So I can quickly click on a group. And then it will move over or remove as necessary before creating this user profile. The other thing that's really nice is that you can also create a user in a disabled state through the individual user creation which wasn't possible before. So now, I can toggle this off if I want to make sure that the user doesn't log in to this profile prior to a training session. (DESCRIPTION) Clicks the cancel button. (SPEECH) All right. So we're going to talk about user maintenance now. So I'm going to go ahead and search for my profile. And once you find the user, you want to view. You can go ahead and click on their name. And this will bring up the user profile. Just like in the organization profile, we have a summary tab of all of the information in this particular profile, as well as the individual tabs at the top that you can use for quick navigation. So once we have this page load here, I'll go ahead and dive into some more of the features. (DESCRIPTION) New page. Text, user profile. A loading animation. There are eight tabs labeled summary, basic settings, group membership, dictation history, password settings, audio devices, user settings history, and advanced settings. (SPEECH) So something that is new with M*Modal user management is you now have the ability to send a self password reset to an end user, instead of the necessary steps of having support go and reset a user's profile in order to then have them create a temporary password and then change. So from within here, you can see that the password self reset email is available. As long as the user has an email on file, they can go ahead and send a support resource, or an admin can send an email to the end user. And they can go ahead and quickly modify their password without the need of support, doing some extra steps to create a temporary password, and send that on its way to the end user. Under group management, a lot of this is very similar to what we have today. But you'll notice that we have the group name, as well as the description that are displayed, as well as the location of where the group is created. It's going to refresh here. Think we're encountering some slowness with the webinar. (DESCRIPTION) Loading animation. (SPEECH) All right. So I'm actually going to move back to my presentation. Oh, there we go. All right. So under group membership, you can clearly see the group name, and then the description, as well as the location of that group, which is really nice in terms of just giving you a high level overview of the group. And then from there, if necessary, you can add or remove groups more easily for that particular user. So let's see if this page will load for us. So within the group view, what we'll see is we'll see the mover option that will become available. With this, I can quickly search across organizations that are available here to find a group that I would like to add to the user, and then go ahead and click the plus sign in order to move it to the users profile. So really easy to remove and then add groups as necessary. This can also be really convenient if you have multiple groups available for your organization. This just allows you to quickly find them without having to scroll through a complete list in order to find the appropriate one. Next item that I was hoping to discuss is audio devices. We'll notice once the table populates here, that we're going to see the same content that we see in FDA. But you'll notice that there's now sorts available for some of these columns. You can quickly use to find information as necessary for input devices or playback devices for a user. Now with MUM, admins have more control and can remove input devices as necessary. If you find them for troubleshooting reasons, maybe a user hasn't used that particular workstation in a long time or they don't necessarily use that input device anymore. You can clean up this table. And that's available to some user types. I'm moving on to User Settings history. This will show you at a glance of the last time that there were changes made to the user profile. And so you can quickly see the date and time of what those changes were applied, as well as a quick overview of what those particular settings are. So if I click on the version number here, it'll generate just a quick snapshot of that information for me. And I can just know at a high level that there's 186 commands added and a couple words. The biggest change for MUM is the dictation history. (DESCRIPTION) Two calendars side-by-side of the months June and July 2023. (SPEECH) FDA had the one calendar view with the red highlighting. And so we really tried to expand this particular feature to give you guys more tools that you can use whenever you're troubleshooting issues. Here you'll see a two-month calendar view. And at a high level, you can quickly see where there's user activity without needing to dig into a particular user session. You'll notice that days are highlighted in a blue color. And this auto scales based on the amount of dictation for a particular user. You can easily see days with high and low dictation, as well as when you're glancing at the calendar view. We've also added the total number of audio for a given day directly on that calendar date. So you know without needing to dig any further that this user had 51 seconds of audio on the 29th. We've also added the ability to toggle on sessions with and without audio. And so I can see now if there were days where the user signed in, but didn't use fluency direct to dictate. This can also be really helpful if you have CAPD only users and just want to see end user activity across the days. I'm going to have the ability to toggle this off or on depending on what their preference is. You can also search in the top right for results. This will narrow down individual session logs that match your criteria so you can dig in further in order to complete troubleshooting. Clicking on one of the days in the calendar. What we've done is actually given you the dictation that goes along with the audio for a particular session. No longer are the days where you'll need to click on the individual ISL in order to see the audio and then listen to it. You can review it directly from in this view and also playback the results at the same time. We've also provided a quick dropdown so you can see relevant session details that might be helpful as you're reviewing the audio. In the top left, we have the ability to modify the date range. And so when I click on this, I have quick links that I can expand the view that I want to choose. Maybe I want to look at more than just one day or I can even use the calendar view to really dig into a specific time frame. So I can pick my day and then even the time of which maybe a recognition concern occurred. Taking a look at this view. We also have the filter options in the top right here. This just allows you to filter down the data again for troubleshooting reasons. So you can go ahead and check off any of the filters that you wish to apply. And it will go ahead and narrow down your results further. So if you're looking for a particular FD versions or microphones that are being utilized, you can apply those through these filters. You'll notice now that commands are automatically highlighted. So built in or custom commands will be highlighted in this field of view. As well as you would see the feedback icon, should a user have submitted feedback during one of their sessions. The last item in here that's important is users can now download audio. This is really nice if you're just trying to support best practices for an end user. Maybe they're struggling with having the microphone too far away or maybe they're eating while they're dictating. You can go ahead and send them their audio just to help reinforce that, hey, we're noticing this. We'd like you to maybe try to make these changes to better improve their user experience. In the top right, you can also search the results in order to find like matches. It'll go ahead and narrow the results and highlight them for you so you can see where those instances are. (DESCRIPTION) Searches for the word patient in the search bar. (SPEECH) Just like in FDA, we do have the async results available. You can see that they'll generate here. And you can move through them just like you would of an FDA. You can play back the results and as well as see any related metadata here. We've also added in the ability to hide this information. Should you want a larger field of view. And then the last thing and what is new to MUM are the result annotation logs or what we call the rails. (DESCRIPTION) Clicks on a button in the upper right that says go to RAL. (SPEECH) These are really more in depth logging associated to this particular session. So here, I can find more in-depth information about recording devices, as well as FDA versions, the workstation the user is on, and even metrics related to their recognition, speed, and quality. These can be searched as well or you can copy the results and look at them in another field of view if necessary. Moving out of user management, we're going to talk about group management. (DESCRIPTION) Clicks on the groups tile in the left sidebar. (SPEECH) So here you'll see the groups that have been added across the different organization levels. You can see the group name and then the associated description of that group to give you just high level details of what user settings might be a part of that group. Just like in the user management table, we added in some sort and filter options so you can really dig into more details if you desire. Looking at the table, we also provided quick links so that you can quickly jump into users that are a member of a group and search to see if a user is missing or needs to be added. In the top right, we also added in the ability to search again for ease of use. So you can quickly find a group that you might want to edit in order to modify the settings. This is really nice, again, if you have groups that are maybe like and naming or just many where you're trying to find the correct one. So talking about group management, a lot of times, we talk about impersonation through affluency direct. And so what we will be doing with the MUM application is actually allowing customers to impersonate a user or group profile through their web browser. And so to do that, a user could simply just click the overflow menu and then open up a profile which they wish to view. So we'll go ahead and move over here. So I had just opened up my user profile. (DESCRIPTION) Clicks a second tab in the web browser. In the left sidebar are the tiles dictionary, commands, abbreviations, device button mappings, medical specialty, formatting, settings, sign out. Logo, fluency direct. The commands tab is highlighted. A panel with the text manage commands that takes up most of the screen. (SPEECH) You can see that the UI is very similar to what we have in the desktop application. Just trying to make sure that admins are already familiar and accustomed to this layout. And so we tried to make sure that we transferred that over to this tool as well. You have your main menu on the left hand side, which reflects what you would see in the FD control bar. From here, admins can go ahead and select any of the user settings options in order to actually jump into these specific categories to make edits. Right now, I'm under commands. You can see I can see all of my command groups associated to this larger command, group rather. And once you're in here, you can go ahead and click on one of the entries. Hold on one second while I sign back in. (DESCRIPTION) The fluency direct log in screen. (SPEECH) We're going to stop sharing momentarily. And we're going to drop back down to our presentation. (DESCRIPTION) Slide, see it in action. A bullet point list. (SPEECH) All right. Hopefully, you guys can see this. Sorry for the interruption there. So this is what I was trying to edit was my command. Unfortunately, (DESCRIPTION) A pop up box with the text edit command. Multiple fields with data. A button at the bottom that says save. (SPEECH) logged me out of the application. But you can see that all the steps are provided just like you would see in fluency direct. And you can go ahead and edit these if necessary, which is really nice. Just being able to quickly make modifications without needing to do all the extra steps that we had with FD for impersonation. The other thing I really like about this ability to impersonate through the web browser is that you can actually open up multiple user profiles or groups at one time and compare them. So this can be really nice if you have a user that might be struggling. But they should have similar settings as another user. Here, you can quickly just utilize the multiple tabs in order to see what user A has compared to user B and then make modifications as necessary. So definitely a nice feature. So moving back-- we're going to just share my screen one more time. So moving back into the MUM menu. On the last thing in here is reporting. (DESCRIPTION) Selects the reporting tile from the user management interface in the first tab. (SPEECH) So I'm going to go ahead and just click on that. And then we'll go ahead and open up our fluency analytics tool, which I have here. (DESCRIPTION) A new tab opens. Text, fluency analytics. FD dashboard. Four different panels. The first panel shows the monthly volume trend as a line graph. The second panel shows the active author split as a pie chart. The third panel shows custom versus edit command as a bar graph. And the fourth panel shows the quarterly volume as a table. (SPEECH) So what we've done with MUM is we've actually moved reporting out of the admin tool and into fluency analytics. And so the nice thing about fluency analytics is it's a hub for multiple products and not only fluency direct. So within here, you'll find information related to FFT, fluency align and even haiku canto, as well as fluency direct. You can create reports and schedule them, which is a nice feature if there's a report that you look at often and are generally trying to create parameters and look at data. Now you can create that report and then schedule it to be sent to you on a regular cadence. So eliminating some of the steps involved. In addition with reporting, this allows you to create many different visualizations of the data in order to be able to really look across it and see if there are trends over time and just give you different interpretations. What I'm showing here on my screen is actually a fluency direct dashboard that was created. All this data is de-identified. But you can get an idea. I have a comparing two sites for a particular organization. So I can see trends over time. This can be really nice if you have multiple sites that you're trying to review and trying to see maybe if you want to do some follow up adoption. You can go and do that by looking at the data and seeing, oh, hospital A is struggling compared to hospital B. I think it's maybe good if we follow up with the end users. So these are just some of the reports that were selected for this. But you can customize this to any reports that you already have generated. All right. So that was the last item that I had in the fluency direct menu here. We're going to actually stop sharing and move back to the presentation. (DESCRIPTION) Slide, conclusion. (SPEECH) All right. So I just wanted to thank everybody for joining. I really appreciate you taking time out of your day to be here with us. And I hope you're excited about the new tools that are coming and some of the features that you'll be seeing shortly. I think at this time, Lisa is going to go ahead and do some questions. Yeah, absolutely. That was a great presentation. And some of the questions that have come in are around the recording. So yes, after this is done, we'll go through the process on our end of getting it uploaded to our website, which will be in the next couple of weeks. So we'll definitely send out a communication to you all that the recording is there. So going into some of the questions. We have is, is this version already released? Not generally to the public. We are finalizing our distribution plan for the US right now. And customers will be receiving additional communication from their adoption specialists within the next month about next steps and how we'll be moving forward. In international markets, this will be available either later this year or early next year. Great. A question that came in is, when will FDA sunset? At this time, we don't have any plans to sunset. FDA formally. Once we get all of our customers moved over to MUM, we'll then take a look at where we are and then place a formal announcement out to customers. So it'll be here for a while. OK. So along those lines, when you make those announcements, and when you start moving things over, a question that came in. Will you be then providing training and support documents? Yes. Every customer will receive formalized training. So adoption will be helping to schedule those just to make sure that everybody has what they need before they transition to the new tool. And that will be free of charge. Great. A question interested in using bulk load of users. That also adds them to a user group as well. Is there any comment on that? Yes. So through the user management, as you're updating users, you can specify what groups you want to add them to. So it can be a single group or multiple groups at a time. Similarly, you can also remove groups as you're adding them. So trying to just consolidate that process. Going back to the FDA sunset, would that then be the same URL or would that change? So the FDA URL will still be the same until we likely make it internal. The MUM URL will be different from the one that we have for FDA today. OK. Just a comment from Macy. Thank you very much that this is much more user friendly. I'm just going through the questions. And there's a lot that are similar. So just trying to consolidate. When user logs are enabled to pull debug logs for your users, where do they go or save to? Yeah. So those go to our internal tool, fluency direct manager. And so once those are sent, we have the access internally to go and grab those once they're available. Does this have the import file feature to add, edit, and/or delete users? Yes. Yup. Just like FDA, you can complete all of those functions. OK. And then how long do we store the audio recordings? Audio is stored for 120 days. Depending on how your site was configured, you might have a smaller range of view. OK. Let's see. Is there a feedback view on the dashboard when users submit? So through the dictation history, we have the icon to let you know that they've submitted feedback. And just like if you would be diving into the audio results and FDA, it would share with you some of the information related to that feedback. OK. And a couple more questions have come in about the release. Can you go back to that? I know you said it information would be coming next month. But can you just go over that again? Yeah. So we'll be sending out a formalized announcement through your adoption specialist and be working with each site in order to schedule a training. With their release, since this is a web based application, new releases will be pushed out to everybody at once. So any time we have a new release a MUM, we'll make a formal announcement and then let you know what changes are coming through the release notes. What you can expect as we update the system. But users will get a URL that they'll use in order to log in. OK. And along those lines, will the credentials be the same? Yes. Yup. Your credentials that you have in FDA will be the same as what you'll have for MUM. And does this version have single sign on? At this time, we do not. We are currently working on a solution for that and anticipate it to be ready towards the end of the year. If not, early next year. OK. So the MUM tool-- and I'm not sure if you'll know this question or not. So is replacing the M*Modal admin tool, will this also be replacing other applications like Fluency for Transcription? No. At this time, Fluency for Transcription, I'm assuming they're referring to the management tool for Fluency for Transcription. That will still be there and be utilized. OK. We would like to start automation to create users. We will be using sailpoint. Can we do this in FDA or would we be better to build this with MUM? So we actually have an FD, while API where you can do user automation if you have a meta directory. I'd have to look more into sailpoint specifically to get more details. But potentially, we'd be able to set up the FD web API in order to complete that user creation. OK. Let's see. Like I said, there's a lot of questions. And some of them are similar. So trying to put them together. With a fluency analytics, will we be able to create our own reports? Yes. Admins will be able to go in and create their own reporting, which is really nice. You're not just set to the unique reports that we had in FDA, which was pretty limited. So there's a lot more capabilities with fluency analytics. So going back to piggybacking on the question about the single sign on. Is that something that they can wait until it's available? Or how will that work until the single sign on is developed? Yeah. We can certainly have a conversation about it. Something to bring up with your adoption specialist. And we can discuss further at that point. OK. Will we be able to import and export commands in MUM? At this time, not directly from MUM. You can complete any of the exports that we have in the desktop application through the impersonation that I showed, through the web browser. Moving into the future, we're hoping that we can have something similar for our customers in order to provide more in terms of commands and like a command list. OK. Let's see. Does this change anything with the M*Modal FD API? No. That will remain the same. OK. And so piggybacking on that. Is there anything that will be needed from the clients and customers to make this transition happen? Nope. Really the biggest thing is just switching over to the new URL. Like I mentioned, your sign on will be the same as you used in FDA today. So it should be seamless. I might ask you to reset your password. But just the first thing you have to do and then you're good to go. You'll have your formalized training just so that how to use the tool. But other than that, there shouldn't be anything else that customers are required to do. OK. If I currently add new users in FFTM and manage them there, would I now add users and manage them in MUM? So you would still create users and FFTM. Those will sync over now to MUM as they do today and FDA. And then from there, you can make modifications to users through MUM or Fluency for Transcription manager depending on what you're trying to edit. OK. Can we use FDA and MUM simultaneously? Yes. So any changes that you make in FDA will sync over to MUM. And likewise, any changes that are made in MUM will sync over to FDA. So as we're going through the transition period and customers are moving over, and as you become more comfortable with the tool, you'll be able to complete some actions in FDA. So if you're unsure or need to follow up with us here at 3M M*Modal, but hopefully you'll get everything you need in your training and be able to use MUM exclusively. Question on, will a customer facing audit be added to MU? We are currently working on that now. I know a lot of customers have asked about that. And it's something that we're currently working on in our sprints and hoping to have it to you by Q4 of this year for those customers that are just looking for a more detailed information about admin changes that are occurring at their organization. OK. A question about user management. How do you get access? User management, in general, you would just want to reach out to your fluency direct adoption specialist and you can discuss if it's appropriate that you have access to the tools. And then from there, if necessary, you would have a training. OK. And going back to the question of how long FDA will be available? Can you just go over that real quickly again? Yeah. So FDA will continue to remain available. We do not have a finalized date yet of when it will go away. So at this time, we'll be here for customers as we do the transition period. And once we're ready, we'll send out a formalized notice about when it will no longer be customer facing. Awesome. All right. Well, I am going to go ahead, since we only have a couple minutes left. Understanding there are still a few more questions in the Q&A. But definitely when this does come out, be sure to reach out to your adoption specialist with these questions. Or if you want to reach out to them now with any more questions, we can try to follow up with you with more. Again, we did record this. So we will be adding this to our website in the next couple of weeks. So if there is anyone from your organization that you would like for them to see this as well, we will definitely have that there to be able to review. We also plan on making these update webinars quarterly. So we do plan on having another one in the September, October time frame. And that will be about web extensions. So that will be coming here to you soon. So be on the lookout for that to register. Again, if you have any questions, please feel free to reach out to your adoption specialist. You will be getting an email after with just some information as well. So, again, we really appreciate you joining today. Thank you, Katie, for all of the information. There was a lot here. So we really appreciate your time today. Katie, anything else that you would like to say before we close for the day? (DESCRIPTION) Slide, thank you. (SPEECH) Just thanks again for joining. I really appreciate you spending the time with us this afternoon. Awesome. Well, thank you again. And we will talk to you all again soon. Thank you again for joining us.

(DESCRIPTION) Slide presentation. Text, Welcome! We will start at 1:00 PM ET/10:00 AM PT. Advancing Outpatient Safety with Reliable and Actionable Insights. April 25, 2023. Click "Join Audio" and adjust your settings before we begin. MGMA logo. Daniel Williams' video feed appears in the upper right corner of the screen. He is seated and speaking to camera. (SPEECH) Hi, everyone. This is Daniel Williams, Senior Editor of Industry Content at MGMA. Welcome to today's webinar, Advancing Outpatient Safety with Reliable and Actionable Insights. Thanks so much for joining us. (DESCRIPTION) New slide. Text, Thank you! Logos: 3M, M asterisk Modal, 3 M dot com slash H-I-S. (SPEECH) We would like to thank 3M Health Information Systems for sponsoring today's webinar. You can visit their website at 3m.com/his to learn more about improving patient safety and all things capture to code. Please see the slide for available credits for the live and on-demand experiences. To claim all credit types, you must complete the session evaluation following the webinar. (DESCRIPTION) New slide. Text, CME/CPE Pre-session knowledge assessment: If you plan to apply for CME or CPE credit, please answer the poll question now. (SPEECH) Now, anyone claiming CME or CPE credit, you will need to answer the pre-session knowledge assessment. It's about to pop up on your screen. And I'll give everybody about 30 seconds to submit their answer. (DESCRIPTION) New slide. Text, Housekeeping: Chat, ask a question, additional resources. (SPEECH) We highly encourage interactivity at our digital events. So please use the chat button to talk to your fellow attendees. And look for that Q&A button when you want to ask questions for the presenters. Any additional resources such as the slides and session evaluation can be found in the learning management system where you accessed this program. The education for this session will now begin. (DESCRIPTION) Logo: 3M, Science, Applied to Life. Text, Advancing Outpatient safety with reliable and actionable insights: Ambulatory Potentially Preventable Complications (AM-PPCs). MGMA. Sandeep Wadhwa, MD, MBA. Miki Patterson, PhD. 3M Health Information Systems. April 25, 2023. (SPEECH) In today's webinar, Dr. Sandeep Wadhwa, Global Chief Medical Officer 3M Health Information Systems, and Dr. Mickey Patterson, product owner of 3M's ambulatory potentially preventable complications grouping software will share actionable insights for advancing outpatient safety measures. Dr. Wadhwa, Dr. Patterson, I know you have a lot of great information. So I'm going to turn this over to you now. (DESCRIPTION) Sandeep's video feed replaces Daniel's. (SPEECH) Daniel, thank you for the kind introduction and for the little pop quiz. I was thinking about those questions too. We look forward to seeing folks' pre-post responses on that. Thank you so much for joining us today. We're excited to share with you some advances in how to look at post-procedure care, and particularly looking at related complications, and quality defects, and beginning to get more insight about 30-day outcomes post-procedure. We have a really rich infrastructure for looking at 30-day events on inpatient care and technologies advancing now to be able to look at procedures as well and add that lens to how we improve our ability to care for patients. So Mick, you can go to the next slide. (DESCRIPTION) New slide titled, Learning Objectives. Text, Review the current challenges and opportunities to track and improve patient safety with elective procedures. Recognize the broader implications of safety incidents on shared risk and value-based care (VBC) contracts. Identify a roadmap to identifying gaps in current outpatient safety practices. (SPEECH) I'll kick off these first few slides and then turn it over to Mickey. And then I'll come back to discuss the kind of close this out. It's nice to see folks from across the country. Texas and New Jersey representing so far. Just a couple of learning objectives. And again, I kind of teed up this conversation as beginning to advance our approaches for measuring post-acute events after elective procedures. And we'll take a really broad definition of procedures and kind of expand that definition from surgical procedures to also include procedures that occur in radiology settings or in all of the internal medicine subspecialties to which I belong. Though was a geriatrician, we don't do too many procedures in ambulatory surgical suites. And so the other part that we thought would be really interesting in this conversation was how looking at procedures could inform shared risk and value-based care contracts for a lot of our listeners today as being an area where there's almost a double, triple bottom line in terms of not only if we're able to reduce some complications, does that increase the patient experience? But it also reduces excess costs. And so as more relationships are being formed and interest is being formed, working directly with medical groups doing procedures. We think this is an area of getting more insight really can help improve performance on value-based care contracts. And then lastly is being able to drill down on where there's better than expected performance against benchmarks and where there's opportunities. By the different physicians, many of you may have several different sites where procedures are occurring. And being able to identify by procedure type, by site, and by practitioner starts to give insights on how to address for performance improvement and perhaps even opens up conversations around, are we getting the complexity of the procedure adequately and appropriately reflected in our coding? Next slide. (DESCRIPTION) A new slide shows a photo of a sign in front of a hospital that reads, Outpatient Services Drop Off, Outpatient parking. Text, Have you or a family member experienced a complication after an outpatient procedure. A, Yes. B, No. (SPEECH) So let's start off with a quick question. And part of this is we're just interested in whether you or a family member-- take a broad definition of family, whatever works for you on how you define a family, if it's spouses, kids, parents, grandparents-- have had a complication after an outpatient procedure, whether that's a surgical procedure, or an endoscopy, or any other procedure? We'll give you a second to feed that information back to us. Wowza. Boy, that is-- Mickey and I have spoken to a couple of audiences. That is a higher percentage, isn't it, Mickey, than I think we've been-- typically in our audience I think-- well, Mickey, what's your sense of just our informal feedback so far? Yeah, somewhere in the high 40s or 50s. Yeah. And so I think we're seeing it on the call here with over half the folks. (DESCRIPTION) New slide. Text, Safety Incident impact for elective procedures. Post procedure adverse events: Challenging to systematically analyze. Contribute to impaired patient recovery, excess paint and suffering, negative patient experience. Professional edict to "do no harm." Litigation costs. Reputation. Insurance premiums and inclusion in care networks. VBC opportunity: Excess emergency department (E-D) and inpatient admission for complication care. Staff time (MD, RN, ancillary services, consultants and supplies). Loss of reimbursement penalties. (SPEECH) And I think we can go to the next slide. I think part of the setup for this conversation is there a sense by patients that these procedures, if they're occurring in an outpatient setting, have a lower risk of complications. And we as providers don't see the patients after the discharge that day. And so obviously, the inpatient setting we're watching folks for overnight or even longer. And also, there's literature that patients may underestimate the severity of symptoms or signs post-procedure, that they may not appreciate-- they may go into the procedure thinking that it's low risk. And so these are parts on the patient side what may be factors that end up manifesting themselves later when things arise to a level of involvement. But part of the other challenge is it's tricky to analyze post-procedure events that were kind of separated both in time and space that the post-procedure care may not be at the site where the care was rendered initially, where the procedure was given. Someone may end up at a different ER. And how you kind of link events together and have a view of a post-procedure window across different sites of care, whether it's an office, an ER, an inpatient setting, has been tricky to do. The other reason why we think this is an important topic is it is central to understanding this edict and extending the edict of doing no harm. There may be an initial focus at the day of the procedure and really focusing in on the best practices to ensure the procedure goes smoothly and kind of bringing some more light to the post-acute period to having that expanded definition of excellence on the day of and then also expanding and helping to support excellence for the next-- in our case, we're looking at 30-day windows here as a relevant range. And as previously mentioned, we think getting more insights for your groups and your practices on where you're doing better or not kind of just helps in so many domains in terms of improvement. And where we are particularly interested in is this intersection between the patient experience and also efficient use of healthcare resources. And that post-procedure events lead to kind of excess emergency room, office visits, supply, inpatient resources, which in an increasing environment of value-based care become areas where we can act upon that have that impact of both enhancing the patient experience, as well as being more efficient in how we use total cost of care and total resources. Next slide, Mickey. (DESCRIPTION) New slide titled, Outpatient procedure trends. Text, Orthopedic procedures have shifted to ambulatory centers (ASCs), with hip and knee replacements seeing the greatest percentage of shift. Technological advancements and Medicare coverage for these procedures being performed in ASCs have aided in these shifts. A bar graph titled, Percent of change in procedures performed at ASCs between 2019 and 2021. Spine arthroplasty, 36%. Shoulder replacement, 45%. Spinal fusion, 73%. Hip replacement, 150%. Knee replacement, 168%. (SPEECH) And so this kind of speaks to the trend that has been going on for 20 years. In a prior life a couple of Governors ago, I was the state's Medicaid director. And this line of outpatient spend was stunning to me in terms of just having to plan for a budget. Every year we were planning for 10%, 15% growth on outpatient services. And we saw a 2% decrease year over year in inpatient. And that was 15 years ago practically, right at the Great Recession. And that trend has just accelerated and continued. And we're finding that we're able to do more procedures in places that are convenient to patients, mainly in community settings, and are doing more procedures and more complex procedures in day settings. And what we're particularly interested at 3M is adding a lens of patient safety quality outcomes to that window. Next slide. And (DESCRIPTION) New slide. Text, Current opportunities and challenges. An aerial photo shows three men gathered together, a man in a white coat and stethoscope shaking the hand of a man wearing a tie, standing with a man in blue scrubs holding a tablet. Text, Greater than 70% procedures performed in outpatient setting. Limited outcomes/safety/quality measurement systems. Share results of a comprehensive quality outcomes system for ambulatory procedures. Trusted objective secure data methodologies. (SPEECH) oh, Mickey, here's the answer to our question. So just kind of more data to support that trend, where we're now more than 70% of procedures, surgical procedures are occurring in outpatient settings. And I did not train as a proceduralist. I probably trained as a talker and a diagnostician. But there was such a rich infrastructure around surgical morbidity and mortality in the inpatient setting. And Mickey is a leader in orthopedics. And and we have such a rich culture of learning from what we did well and where we could have done better on that inpatient setting. And we see this trend of bringing back culture of-- and it exists today. So I'm not trying to imply it doesn't exist today. I think that as data is able to be linked and we're able to apply technology and software, we're able now to get better views of this dominant side of care. And frankly, we have very limited tools for looking at post-acute care. I'll let Mickey kind of describe our work in this area. But we think that there's an opportunity to add to the measurement methodologies out there for outpatient safety. And this is work that 3M has been involved with for 40 years. There's a role for the private sector in maintaining, and updating, and innovating on quality, and safety, and payment. Great that we have government options. They serve a purpose. But we also think that having a private sector option allows for choice, and enhancement, and responsiveness. So Mickey, let me turn the floor over to you to kind of talk a little bit about what we've built here. Sure. (DESCRIPTION) Miki Patterson's video feed replaces Sandeep's. New slide. Text, 3M Ambulatory Potentially Preventable Complications (AM-PPCs) compared to A-S-C-Q-R. 3M AM-PPCs: 3M Ambulatory Potentially Preventable Complications, computerized grouper for billing or claims. A-S-C-Q-R: Ambulatory Surgical Center Quality Reporting, very manual process. A chart compares 3M AM-PPCs and A-S-CQ-R in the following categories: Measure scope, procedures included in measure, procedures excluded from measure, procedure definition, complication defined, and risk adjustment. (SPEECH) So we're going to talk a little bit about the differences between ambulatory potentially preventable complications. I'm going to call them AMPBCs because it's easier. And what kind of quality outcomes are we looking at now for these same complications. And as you look, you know that the ambulatory surgery centers, the quality is really looking at processes. There's the same biggies, wrong site, wrong side, wrong patient. But they also have is the cataracts, improve someone's vision, and did they have normothermia? We go deeper. And what we're going to show you is 1,500 different complications put into groups. We're looking for 30 days, not just how many counts of the occurrences. We're also looking for, where did they show up and what procedures. So when we start to look, I'll give you a little bit more detail when I talk to you about the logic. But we think that the AMPBCs are a much richer way to start to look at our data. (DESCRIPTION) New slide. Text, USNWR evaluating inclusion of elective procedure safety in 2023 ranking. Outpatient outcome measurement: We recently began working with 3M Health Information Systems: Licensed 3M's Ambulatory Potentially Preventable Complications (AM-PCC) Grouping Software. Exploring potential use in specialty rankings and procedure rankings. Initial exploratory work will focus on: ear, nose and throat, gynecology, urology. The first two paragraphs of an article titled, Hospital Rankings Shift Emphasis to Objective Data Away from Expert Opinion. (SPEECH) I did want to say that US News World Reports has actually licensed our grouping software and are considering using it for the 2023 hospital rankings. They did say that they were starting with ENT, GYN, and urology. And expect to hear something from them in the June time frame to be used if they're using this for the July rankings. (DESCRIPTION) New slide. Text, Ambulatory potentially preventable complication. Harmful events (e.g., accidental laceration) or negative outcomes (e.g., sepsis) that develop after an elective procedure was performed in the ambulatory care setting and that may result from processes of care and treatment rather than from natural progression of an underlying illness and are therefore potentially preventable. (SPEECH) So I'm going to talk a little bit of details of, what is an ambulatory potentially preventable complication? It's a harmful event. So it's an accidental laceration, or a negative outcome such as an infection. It has to develop after an elective procedure. So there's no emergency room procedures here. And it had to be performed in an ambulatory care setting. These are key pieces so that we're only looking at scheduled elective procedures. And we're not really looking for a problem that had medical issues from an underlying illness. We're not looking for worse kidney because they're diabetics. We're really looking for a complication that was related to that particular procedure. (DESCRIPTION) New slide. Text, Identifying, locating and quantifying procedural complications. Procedure groups: 93 procedure groups which include more than 2,900 elective procedures. PSGs include similar procedures that also share the same relative risk. A classification hierarchy is applied to select a single and primary procedure group that best classifies outpatient encounters. For example: colonoscopies, upper GI endoscopy, biopsies, cataracts, arthroscopies. (SPEECH) So here's the logic. And there's about three or four little chains of this logic. So I'm going to just give you an overview of this. There were some almost 3,000 procedures. And what we did was we started anatomically looking at these procedures and then grouped them together by the complexity of the procedure. And to give you an example of this, if you had, say, a trigger finger release, that's not really complex. It's pretty routine. But if you were an outpatient and you had a total shoulder replacement done, much more complex. That's in a different procedure group. And we expect different complications. When we did this, we were able to do a hierarchy. So if you had more than one procedure done at the same day, one of those, the highest risk, would bubble to the top. And then we linked it for 30 days out looking for a complication. (DESCRIPTION) Complication groups (AM-PPCs): 35 total complication groups which include greater than 1,500 unique complications. Complications must meet defined timing requirements. (example: 48 hours for infection.) Complications plausibly related to the procedure. Sepsis, UTIs, bleeding, pneumonia, infections, hemorrhage. (SPEECH) The logic for the complication is such that we have some 1,500 different unique complications. And we also group them. So we group them in hematomas, and lung problems, and infection problems. We also had those complications needing to meet timing requirements. And they had to be plausibly related to the procedure. So you wouldn't have a problem with an eyeball related to a cardiac procedure. (DESCRIPTION) Complication setting: Emergency room visits, inpatient admission, outpatient encounters. Available on premise (CGS) or cloud (GPCS) with Medicare and national benchmarks. (SPEECH) And then the most important one is, where did that complication show up? So we are able to look at 100% claims and see that some of these complications showed up in the emergency department. Some were inpatient admissions. And that's not a readmission because they were done outpatient. So it wasn't a readmission. Or the complication showed up in an outpatient encounter. When we did look at these, we also started to think, what are we going to gear this on? So when we looked at benchmarks, we wanted to make sure it was a significant complication. So we excluded in our benchmarks those outpatient encounters. We only looked at the emergency room and inpatient as significant complications. And then this component is available on the grouping software that's either on-premise or in the cloud. When we did make those benchmarks, we started with the Medicare benchmarks. And then we added to national benchmarks. That includes pediatrics. (DESCRIPTION) 3M AM-PPC: Risk adjustment approach. Case adjustment: procedures with higher risk of complication grouped together. Plausible procedure and complication relationship: Clinically relevant focus on complication to procedure relationship limits effect of chronic conditions. Elective procedure focus: Less procedure complexity, care team informed by knowledge of chronic conditions (exclude ED procedures). Research: Chronic conditions reported on claim at time of procedure have few complications. Age, disability status, oncology adjustment: correlated with chronic conditions, frailty, ability to self manage. (SPEECH) And I want to talk a little bit about our risk adjustment approach. So the basic premises that we used was we wanted the procedure to be the risk adjustment because it was elective and because these patients were going to be done in outpatient and expected to go home. That's how we started to categorize. And we make sure the complication is plausibly related to that procedure, whether it was an anesthesia, or a puncture, or things like that. And then we put them into different complexities. So we put the less complex procedures in a group. And when we did do the data-- and I'll show you a little bit about the data after that-- but it made sense. When we thought these were highly complex procedures, they did show that they had higher rates of complications. And we have looked at the chronic conditions. And actually when we started to look at it, if somebody had an oncology diagnosis in five of our procedure groups, they had a higher risk. So we built that adjustment in. We also had an age adjustment. So if you are over 75 or over 85, you had increasing risks. So we put those risk adjustments in. When we did the Medicare data, we also noted under 65 had a higher risk. And these would be the dual status. These would be patients that had disabilities. So it kind of made sense when we went through it. And again, we're always learning. We're doing research. But this is where all of the data has brought us. And I'll give you a little more explanation about that. (DESCRIPTION) A new slide shows a flow chart titled, Knee procedures and example of 3M AM-PPC exclusions. It begins with two branches: In gray, Text, Exclude: Knee procedures done as inpatient or observed for 23 hours and admitted. In Blue: Include: Knee procedures that are elective, performed in ambulatory setting. From the blue box is a row five other boxes: In gray: PSG 3 Knee Arthroscopy, PSG 14 Knee Arthroplasty Revision, PSG 28 Open Knee Fracture and Ligament Repair, PSG 29 Other Knee and Soft Tissue Procedures. In blue: PSG 13 Knee Arthroplasty. An arrow points below the blue box to another blue box: PSG Assigned: 13 Knee Arthroplasty. From that blue box is a row of four boxes. In gray: ED visit day 1 with fever, ED visit day 15 for burn on hand, ED visit day 32 with hematoma and hemorrhage. Each gray box has "excluded" written in red and the reason. The blue box reads: ED visit day 10 with deep vein thrombosis: Included, plausibly related to procedure and meets timing requirements. (SPEECH) But when we did the logic for this, we have this exclusion logic. So when we looked at all of the claims data, the first thing we did was eliminate anything that went inpatient. If they didn't make the 24 hours for observation, they're not considered an ambulatory procedure. So let's use knees for an example. You can see the five groups here. We have knee arthroscopy, which is a very less complex than a total knee, and a revision total knee, and fracture repairs. So if you particularly went into the group of having a total knee joint done, you would be in PSG 13. And we would start to look out from then on for any complications. And we would tie it. Now, we did talk a little bit about that 24-hour window for the fever, for infection of a wound. And we're going to keep that on our back burner. It's excluded at this time, so we can look at more data. But we think it takes more than 24 to 48 hours to develop a wound infection that can be seen post surgery. We also looked at DVTs. So if this patient came back to the ED with a deep venous thrombosis, that's plausibly related to that knee joint surgery. It makes sense. But if you came back and had a burn on your hand, that is not really plausibly related. So that would be excluded in our logic. However, if you did come back 32 days after and had a hematoma or a hemorrhage, we would flag it. It's outside the 30-day window. But it is an event. And that would be a flagged event. (DESCRIPTION) A new slide, titled Complication Groups, shows a long list of AM-PPCs with descriptions, each one with a different number code. (SPEECH) So here's a little bit more definition about what complication groups are. If you look on the left hand column, pneumonia, aspiration pneumonia, pulmonary embolism, these are very common complications. And they're used for inpatient potentially preventable complications. But (DESCRIPTION) A red box outlines two codes in the right-hand column: 101, Post-procedural infections of eye and adnexa. 102, post-procedural complications of eye and adnexa. (SPEECH) if you look on the right hand column, we have started to put in the complications directly tied to procedures, such as a post-procedure infection of the eye and adnexa. This would be correlated with eye procedures. And the same thing with muscle skeletal. So (DESCRIPTION) A new slide shows a chart titled Top procedure '19/'20 FFS Medicare HOPD 3M AM-PPC. It shows, PSG, Description, At-risk procedures, ED complications, IP complications and AM-PPC Rate. (SPEECH) here's a lot of data to show you. But I want to walk you through it so you understand what we came out with. When we looked at all of the Medicare fee-for-service hospital outpatient data, and we looked at the top 20 procedures, top 15 procedures, at the very top the number one procedure done for this group was an upper GI endoscopy procedure. There were over a million and a half of these done in '19 and '20. Of those, almost 7 and 1/2 thousand were seen in the emergency department for complications related to that upper GI procedure. And some 23,000 were unplanned admissions from complications from that upper GI procedure. Hence giving us that complication rate at the end. (DESCRIPTION) On the chart, the 103, Routine Cataract Procedures row is underlined in red. At-risk procedures, 487,922. ED complications, 114. IP complications, 938. AM-PPC rate, 0.2%. (SPEECH) If you look down a little further, you can see 103, the routine cataract procedures. Again, a half a million of these are being done. Very few are coming to the ED. Very few are being admitted. And the risk rate is 0.2%. So that made sense to us that a very low risk procedure would have low risk. Again, upper GI, almost 2%. And when we look at it, if you look down here at number 94, upper genitourinary stent and guidewire, now this is a very complex procedure. These are sick patients. We're putting stents in them. There's 174,000 done. But of those, 5,600 come to the ED. And 9,000 are being admitted. And because we know that's a high risk kind of a procedure, it made sense-- look at the risk of complications, 7.6. So it did pan out when we started to look at these procedures. (DESCRIPTION) New slide titled, Data by PSG 70 upper gastrointestinal edoscopy procedure. A bar chart shows AM-PPC for the procedure. At the top at 0 AM-PPC 70 is, Infection, Inflammation and Clotting Complications. At the bottom is Gastrointestinal and Peritoneal Complications or Significant, dot, dot, dot, at about 10,500 AM-PPC 70. (SPEECH) What we're also able to do is not only look at the complications what comes to the ED and gets admitted. But we know what the complication was. So if you look at this upper GI that we looked at that top layer, when we started to look at the data, the highest percentage were coming in for gastrointestinal or peritoneal complications or significant bleeding. And this was after they had been discharged and gone home from a procedure. So this is not direct admits. So when we look at this, we are now able to see. We can focus on where we need to make improvements. We need to find out what's going on in these particular cases. (DESCRIPTION) New slide titled Complications by complication group. A pie charted titled Complication Frequency Medicare 2020. Septicemia and Severe infections, 20%. Urinary Tract infection, 18%, other pulmonary complications, 14%, Moderate infections, 5%, Hemorrhage and hematoma, 9%, pneumonia and other lung infections, 6%, Infection, inflammation and other complications of devices, implants or grafts except vascular infection, 5%, Major gastrointestinal complications or significant bleeding, 5%, post-hemorrhagic and other acute anemia, 5%. (SPEECH) And another surprising thing when we looked at all of that data for 2020, of all of the outpatient procedures, septicemia and severe infections came up number one. And that's 20% of all complications for all outpatient Medicare procedures for that year. So it does tell us we have a significant problem here. Urinary tract infections, what's happening post-procedure with those urinary retentions? Are they getting infections? Are they having to go to the emergency room or waiting too long and needing to be admitted? Pulmonary complications, hemorrhage hematoma. See this red one here, the mechanical complications? This (DESCRIPTION) The red wedge of the pie chart is labeled, 21,032, 7%. (SPEECH) is actually good to highlight that there may be problems with an implant or a device that we may not know of. So this information is really helpful for us to start to look and give your head a scratch and say, I wonder what's going on there? (DESCRIPTION) New slide. Text, AM-PPC surfaces contamination issue. Vignette: Previously healthy 65-year-old presented to the E-D with symptoms of sepsis two weeks after Upper Gastrointestinal Endoscopy Procedure. E-D doctor thought it odd that this was the fifth person in the past couple months with a similar story. Outcomes: AM-PPC reveals 2x as many complications than expected for Upper GI Endo resulting in sepsis. Root cause analysis: After new endoscopes were purchased for the GI group, Central Sterile never changed to the manufacturer's new sterilization method. What can we learn: High rates of procedure complications relative to a benchmark can serve as a trigger for review (or basis for performance improvement incentive). (SPEECH) And here's an analysis of what this could surface. If you have a 65-year-old presenting to the ED with symptoms of sepsis two weeks after that upper GI endoscopy and having an ED doc think, wow, this is kind of odd. This is like the fifth person that I've seen with this kind of scenario in the past couple of months. So when you looked at it, if you looked at those upper GIs and you saw that they had sepsis relating from that, you could start to do a root cause analysis. What did this find? And this actually happened nationwide a few years ago is that the GI groups were purchasing these new scopes. And they never alerted central sterile that they're new scopes. Central sterile never knew that there was a new manufacturer sterilization method. So this kind of data would help you do that root cause analysis and maybe change the way you attack something, look at something, or just trigger a review to look at the process. (DESCRIPTION) New slide. Text, National Medicare risk adjusted OP facility-based procedure complication variation (CY 2020). A bar graph titled, Distribution of hospital A/E Medicare FFS performance CY2020, showing HOPDs at A/E. The bars form a standard bell curve, the highest center bar going up to just over 250 at just before 1 A/E. (SPEECH) And this is what our rankings look like when we took all of the outpatient facilities and compared them. So this is 2020 data. And we have an expected kind of centrally. But you can see that some of the facilities were much better than expected. And some of the facilities were less, they were worse than expected. So these on the right hand side, these facilities that had more complications than you would expect would be the place that you would start to concentrate on. (DESCRIPTION) New slide. Text, Driving network performance on complications: site of service. A bar graph shows number of providers along the y-axis and percentage better than expected and worse than expected along the x-axis, with 0% in the center. The tallest bar is 12 providers at 0%. The shortest bars are 1 at negative 70%, negative 50%, and positive 60% and positive 70%. Text, The best performing provider was 70% better than the worst performer. 1 year, greater than 2,500 complications, costing $13.3 million, for providers performing below expected. (SPEECH) And this is just another example with a smaller data set. This is a statewide. And particularly, this had an orthopedic surgical center that was 70% better than expected. And it also had an orthopedic surgical center that was 70% worse than expected. And if you think about all of these complications that are happening greater than expected, these impact your margin. These impact your patient safety and outcomes. So it would be good to know where are the issues coming from. (DESCRIPTION) New slide. Text, Data by service line. A flow chart shows Procedure subgroup PSGs along with their total procedures, complication seen in E-D, complication inpatient admit, and risk IP and E-D. (SPEECH) And when we did this, we talked about doing it in anatomically. But we also did it by service line. So you could go and look in-- and I use spine because I'm orthopedics. And orthopedics and neurosurgery both do spines. But we took that all apart, that kind of division by utilizing cervical spine procedures or fusion procedure, which is a higher risk group. And same thing with lumbar. If it was a lumbar spine procedure versus the ones with fusion and higher complexity. And we could see by how many procedures which came to the ER and what was that risk of having a complication. Sandeep, I'm going to pass this on back to you. (DESCRIPTION) New slide. Text, AM-PPCs, documentation and coding quality, Top 10 procedure by volume: angiography and catheterizations. Coding Guideline: PSG 53, Coronary Angiography Procedures: at-risk cases, 564,725. Rate, 1.84%. PSG 52, Left and combined heart catheterization procedures: at-risk cases, waiting on data. Rate, 2.83%. Left Heart catheterization procedures have an expected AM-PPC rate that is 57% relatively greater than coronary angiography procedures. (SPEECH) Thank you, Mickey. Really appreciate you kind of reviewing all the insights that one can get from this approach. And listeners on the call, we welcome your thoughts and your thinking about how kind of a 30-day insight for the procedures that are going on with your groups could be useful. And I think the piece that we're really excited about is being able to show some of those benchmarks, that how are you doing against folks that look like you? And we're getting more and more data into our database. But the other part that we've been looking at is how important coding may be on capturing and making sure that the case is in the right risk group. And so in this case, if the procedure was a coronary angiography, we kind of expect a 1.8% revisit rate. But if they did also go into the left heart and do an EF, an ejection fraction, and you go through the aorta, that becomes a more complex procedure. And you're increasing the complication rate by a third, almost to 3%. And so we're talking to some of our coding experts. And the professional [? may ?] not look that different. But often times, this may be an example where if this isn't coded with the left heart cath procedure and just got coded as coronary angiography, we'd be putting that patient into a different risk group, just as Mickey was describing. And so we think as folks look at their outcomes that there's both a chance to do a check on the coding accuracy, as well as starting to look at the quality procedure. So Mickey, we can take a look at the next slide too. (DESCRIPTION) New slide. Text, AM-PPCs, documentation and coding quality. Top 10 procedure: cataracts. Coding guideline: complex cataract removal: CPT 66982 is defined as "extracapsular cataract extraction removal with insertion of intraocular lens prosthesis (one stage procedure), manual or mechanical technique (e.g., irrigation and aspiration or phacoemulsification), complex, requiring devices or techniques not generally used in routine cataract surgery (i.e., iris expansion device, suture support for intraocular lens, or primary posterior capsulorrhexis) or performed on patients in the amblyogenic developmental stage. Routine: PSG 103, Routine Cataract Procedures: at-risk procedures, 487,922. Rate, 0.19%. Complex: PSG 104, Complex Cataract Procedures: At-risk procedures, waiting on data. Rate, 0.31%. Complex cataract procedures have an expected AM-PPC rate that is 63% relatively greater than Routine Cataract procedures. (SPEECH) I also just wanted to pull an example of a low risk procedure, cataracts. That's something in geriatrics we're very keen to address just as a leading cause of blindness that is reversible and treatable. But you can see exactly what we expect, which is a pretty low complication rate, 2 in 1,000. But if it's a complex procedure, that increases, again, by a third. And this can be very tricky on the coding side in terms of getting the-- it's complex coding rules. But I think that this would be another area kind of in addition to the quality improvement areas that we would expect folks, depending on your procedure suite, to now have more of a quality lens for why coding is really important to kind of complement the traditional being made sure you're being paid in full for the service being rendered for the complexity of the procedure. We think having this quality safety lens will further contribute to kind of accurate, complete coding. (DESCRIPTION) New slide. Text, AM-PPC mechanical failures. Vignette: AM-PPCs reveal higher than expected mechanical complications from the new joint center. Patients were being admitted for joint revisions less than 30 days following a joint placement for one facility. Outcomes: This rate was noticeably high and points to the device or application. What can we learn: High rates of procedure complications relative to a benchmark can serve as a trigger for review (or basis for performance improvement incentive). (SPEECH) And so this is another kind of vignette of where we think there will be insights. And I invite you to think about in your own settings for the common procedures that are being done by either by the doctors on the call or by the practice sites that, in this case, this example of being able to look and see if the rates of mechanical complications are higher than expected, and then drill down and say, OK, let's apply a lot of the root cause analysis and all of the total quality management tools that we have at hand to improving that post-acute experience, and identifying root causes, and building on the practices that I think are in place, but now being able to extend them to perhaps more procedures, and have benchmarks, so that you're able to get a richer sense of relative performance. Next slide. (DESCRIPTION) New slide. Text, Establish baseline for elective procedure complications. 1, Integrate into your BI environment. 2, Input your billing, EDW or claims into grouper which feeds your analytics engine. 3, Conduct system-wide analysis by site, service line and provider for assistance. 4, Benchmark to Medicare or national norms. 5, Assess, aggregate, site, service line actual to expected. 6, Create actionable data for clinicians, HOPD, service lines. (SPEECH) And we're kind of wrapping up here. We are interested if there's folks on the call who would like to learn more. We see this system as informing your business intelligence environment. And we are coming to the table with some Medicare and national benchmarks, but can complement that with information from your own system and start to build out more recent data. Always the drag with these payer databases is that there's a lot of latency. And and we're able to bring them more current with your information, and then invite that analysis by procedure, by physician, by site, how are we doing on post-acute care events, and where are we doing really well, and let's learn from those sites, and expand that out, and then ultimately kind of driving towards actionable results that are kind of aligned around patient safety. Next slide. (DESCRIPTION) Next slide. Text, Summary: Promote member safety in outpatient settings. Growing need for ambulatory standards of care and safety is great. Identify actual versus expected variations. Source comprehensive, trustable, actionable outpatient methodology and data. Foundation to VBC in rapidly expanding ambulatory care. (SPEECH) And so in closing, we're looking forward to spending a little bit of time discussing your questions and kind of advancing the safety lens both from the procedure to that 30-day window. And as Mickey had mentioned, US News is evaluating this system on the outpatient facility side. But we built the system so that it also can look at events at ACs and even office events. And so we're able to look at procedures in a wide range of where they occur. And part of what we think is really important is recognizing that there are risks with every procedure. And that that's why that slide that showed the different rates, we don't have a strong opinion yet on what's the right rate. For us, this reminds us of what re-admissions looked like 20 years ago, where that slide that Mickey showed you was not a normal distribution. It had more of this tent approach. And there's a lot of variation. And on that second slide, you saw a lot of what we call fat tails, where you have outliers. And so we expect with more of this focus on 30-day results to see that curve kind of take on more of an upside down U and the extremes get pulled in. And we see more concentration around the center. And we start to see that move to the left. And so we would love to invite a follow up if partnership or collaboration on that would be of interest. And our team works very closely with many payers on value-based care contracts. And so as your thinking about commercial or Medicare or even seeing more Medicaid procedures going into value-based care contracts, we'd welcome partnership. And one thing I forgot to mention earlier was we did build the system to also include pediatric cases. And so we have been showing you a lot of Medicare data. The Medicare data tends to be a little bit more available. As a former Medicaid guy, I think a lot about kids and moms. And so we have built this out to also look at patient safety event for those adolescents and kids that are getting inguinal hernias, or getting tubes placed in their ears, or the whole range of procedures that kids get to are also included. Final slide. This is the slide that always makes me nervous, Mickey. Let's go there. (DESCRIPTION) A new slide shows phots of Miki and Sandeep. Text, Thank you. Miki Patterson, PhD, NP. miki dot patterson @ m m m dot com. Miki is spelled M-I-K-I. Sandeep Wadhwa, MD, MBA, swadhwa @ m m m dot com. (SPEECH) I feel like there's truth in advertising. Are we matched up here? I feel like the hair is getting whiter. And the Vision's getting a little worse. But we wanted to leave you with our contact information and invite follow up directly. And Daniel, if there's questions, we can spend a few minutes going through. That sounds great. Thank you, Sandeep and Mickey. Great presentation. We do have some questions that have come through. And just as a reminder to everyone, just go to that Q&A button down there. You'll see that icon at the bottom of the screen. Click on that and drop your question in there. So first question up, how much of these growing inpatient numbers to have procedures at ASCs are contributed to COVID worries and not wanting to be at major hospitals, do you think? This attendee says, I hear this a lot in my surgical practice, not as much recently, but absolutely heard it a lot in '21 and 2022. Lauren, thank you for that question. I think it's a very perceptive question. And your hypothesis, it follows our sense of the trend too, that we see institutional settings as being increasingly reserved for higher risk and complex procedures. Particularly during COVID, I think we were getting our hands on more ASC data and saw that trend with more procedures going on with ASCs, as you did too. I'll make one side comment, which was Lauren, we saw lower infection rates at the height of COVID. And I hesitate to say this. And I think we'll probably prepare a paper on this. But we're seeing those infection rates begin to climb up. But it looked like in 2020 and 2021, this nationwide focus on sterility everywhere sort of played out in our data. So Mickey, I think we're super eager to run the most recent data set we have to see that trend. So Laura, I think your point is well made that COVID accelerates this trend. And we just think there's this broader trend around convenience. And that's kind of favoring these less institutional sites of care. Mickey, anything to add on that? No. Well, I just think that that's right. It did sort of force the hand of people who wanted or needed procedures done to go outpatient. But I think it was starting to happen. That sort of gave it a giant boost. And I don't think it's going to go back. I think that's just the way it is. But the data didn't show that we had many less infections and complications from the 20s, 2020, during the COVID time. OK, thank you all for that answer. Next question up, how do you anticipate this could or will in the future be used from the consumer side? Knowing that the ASC I'm headed to is 70% better or worse than the benchmark would influence my decision. Anonymous attendee, excellent question. This was somewhat of the exact motivation on why we were investing on building on this system, was to give the public, to give the whole system visibility on performance and to do so in a way that was thoughtful, that's risk adjusted, that's clinically related. And this is, I think, where US News kind of jumped on this, which was to give consumers more visibility on patient safety in outpatient settings. But their work tends to focus on hospital outpatient facilities. So I think the ASC visibility is still something to come. And as we talk to all payer claims databases, or public reporting efforts, or people who have complete-- I think particularly, Mickey, within different states is my guess where we'll see the ASC sites being included. Now, on the other hand, anonymous, I think that there's an opportunity for the groups themselves to tell a safety story within their market. I think so much of the-- you guys tell me. I'll see you at some event in person at some date. And we'll have a drink of your choice. But I think that people are making choices oftentimes on reputation or name, Daniel, and that there's an opportunity to start for folks to say, hey, our patient safety rate is half the regional or norm in their area. So I think folks will start to distinguish themselves independently to the consumers as part of their efforts and how they describe themselves. And this can help tell a broader story. And I think we're seeing that, just to answer your question, I think this is kind of where I think US News is going on the facility side. And I think we'll start to see, Mickey and Daniel, more folks looking at bigger databases and wanting to have more informed decision making for consumers. And I think the other piece of that is we compare apples to apples. So no matter where you did this procedure, on the C spine, or the knee joint, or whatever, everybody's in the same pile. They're all grouped together because of the procedure, not necessarily where they go. But when we did do a little glimpse into the 2021 data from CMS, the ASC had much less complex procedures being done. You would agree that that's where they go. However, there really isn't a lot out there to say how well is this particular ASC. And I think this is a tool that will help. OK, just a reminder to everyone, we do have some time left. So please drop those questions into the Q&A. Got a couple more here for you all. What are some examples then of action plans after these insights are raised? And how quickly might you expect to be able to close these care gaps? I'll take that. So one of the things that-- we looked at total joints. So I love to do this because it makes sense to me. But when total knees, when we looked at the complications for them, the number one complication was anemia and acute anemia. So as a clinician, as somebody that's going to go look at this, did they have an adequate crit before they started? Did we do a lot of cauterization? Did we use trans anemic acid? Was the tourniquet up? There is a lot of things that you can start looking at in that realm. So I think once you start to understand what your complications are, you can start to do something about it. People just don't know. I just think a procedure was done. And then they went home. And unless they came back in a couple of weeks to get their stitches out, you don't really know what happened. You think they did great or somebody called them. But if they did go to an urgent access clinic or they went to the ED, you may get notified. But it's a while from now. But not everybody knows those outcomes. And now getting this data group that we can start looking at national benchmarks, I think that's going to be really helpful for the future. That's great. Let's go to the next question. We've got time for that. Are there any ways to leverage these insights proactively when considering incremental risk contracts? Mickey, I can start on that one. I think that if groups are equipped with their performance, it really can help inform your negotiations with the payers. That you're able to start to have a story that expands beyond access and price. And often, I wanted to say this earlier too, sometimes people will focus on mortality as a measure, Daniel. And it's great. It's just those are very rare events. And so when you are able to look at ER visits and inpatient, you start to be able to distinguish performance a lot more readily. And so I would say that the piece I'm excited about is the groups being able to showcase their strengths in a way to the payers that I don't know if the payers appreciate. And I think it just starts to influence the negotiations. And then the other piece is kind of this opportunity within a contract to say, OK, are there opportunities here, where a simple change or a focused effort with a particular provider or site can not only give better patient experience and safety, but also reduce some costs that kind of fit into this movement towards bundled payment? I mean, there's a lot of talk-- the action-- I mean, you guys are closer to it. We hear that a lot with orthopedics and oncology. But as we start to get interested in that, I think getting the experience on performance, Daniel, starts to build up the expertise and experience with insights. OK. Got one more question here. And again, everybody, we do have time for probably one or two more. So if you do have something, drop it in that Q&A section. So we've talked a lot about the data. So where are the blind spots? Where are the blind spots in that data? Mickey, you want me to start? Sure. I can do-- I think the-- I think some of the blind spots on the data are the payer data for our benchmark tends to be complete but not recent. And so one blind spot is when we're pulling data from our partner sites that they may not have access to the emergency room or inpatient experiences. And so that can be a blind spot if there's not a relationship with kind of regional ERs or inpatient. We can catch that, but dated from the payer. But I think that's one blind spot that we're trying to get a sense of systemness on when we're pulling data out of the group's databases. Mickey, your thoughts? Yeah, I think one of the things that we thought of when a large hospital system, if they're running their data through it, there will be a little bit of leakage because they might go to a local ED near their house if they came to a big center to have their procedure done. So there is that in the mind. But when we looked at this particular 100% Medicare to start with, we did have all comers. But it is years old now. And as we do every year, we'll catch it up. But I think that is the blind spot that we have. OK. I think we've got time for one more. So why are inpatient PPCs so much easier to identify than the outpatient ones? They're a captive audience. So if you're inpatient and you have a complication, you're already still inpatient. It's in your chart. It can be captured. The problem is when you have your procedure done at the ASC and you go home and it closes at 5:00, you're not actually going to go back there with your problem. So they can't necessarily catch it. And I think that is the ease or not ease. But the fact that we now can do this through 30-day either billing or claims is worlds ahead of what we-- we were not even able to find out what all these complications were. All right, well, Sandeep, anything you wanted to add to that? You feel good about Mickey's answer there? I do. Daniel, there's no amount of time I can't fill with my own voice. So I'm trying not to do that, Daniel. So I thought Mickey's answer was great. OK, perfect. Well, I want to thank you both for a great presentation. And I want to thank everyone in attendance for attending, first of all, and then also helping drive that conversation during the Q&A. That was great. Thank you all for doing that. (DESCRIPTION) New slide. Text, Thank you! Logos: 3M, M, asterisk, Modal. Text, 3 M dot com slash H-I-S. (SPEECH) The education for this session is now ended. We would like to thank 3M Health Information Systems for sponsoring today's webinar. You can visit their website at 3m.com/his to learn more about improving patient safety and all things capture to code. (DESCRIPTION) New slide. Text, Thank you for attending! Additional content questions? ask 3 M H I S @ m m m dot com. [REQUIRED], Fill out the evaluation to be able to claim CE credits. A screenshot from the MGMA page shows a green arrow pointing to a link that reads, Evaluation: Zoom Webinar. Text, Webinar questions? Contact E D at mgma dot com or www.mgma.com/events. (SPEECH) And if you do have content questions, email ask3mhis@mmm.com. And if you have a webinar question, contact ed@mgma.com. And if you haven't already, please be sure and click on the evaluation link. It's located in the learning management system. After completing this evaluation, which is required to claim credit, please allow 24 hours for system communications to process credit for this session. For more information on all of our conferences and webinars, please see the MGMA events page. Thanks, everyone. Look forward to seeing you at another online event soon. Sandeep, Mickey, thank you again. That was awesome. Thank you, Daniel.

(DESCRIPTION) People sit in video call tiles to the left of a slide presentation. The slide reads, On24 for a better webinar experience! Below, a sample layout maps out presentation features. The sample slide is headed, 3M. Science. Applied to Life. A great company is showing what interesting applications a fantastic product° can bring for motivated users. Arrows point to boxes down the lefthand side of the example presentation. They say, Media player, Resources, and, Have a question? Let us know here! The center of the sample presentation has a box for slides and an Ask an Agent button below. Arrows pointing to features down the righthand side say, Meet our speaker! and, We want to hear from you survey! The next slide begins the actual presentation. It reads, 3M. Science. Applied to Life. MaineGeneral uncovers opportunity, improves CDI efficiency using natural language understanding. August 2023. In a photo to the right, a man wearing a stethoscope speaks to a woman leafing through papers on a clipboard. (SPEECH) Good afternoon and welcome to today's webinar where we are hosting MaineGeneral as we talk about how they uncover opportunities and how they improve their CDI programs. Before we get started, I'm going to just do a couple housekeeping items to talk about the portal and the platform that we are using. We are using the On24 platform. So we do recommend that you utilize Google Chrome, close out a VPN or any multiple tabs that'll help with the bandwidth. If you are having any audio issues, check your speaker settings. And if you do a quick refresh of the browser, that usually clears out any issues with that. (DESCRIPTION) The next slide reads, Housekeeping! On24 Webinar Platform for a better user experience! Use Google Chrome and close out of VPN/multiple tabs. Check speaker settings and refresh if you are having audio issues. Ability to move engagement sections. Ask questions! Chat with attendees. Certificate of Attendance available to download for live webinar sessions. Engagement tools and CC available. Check the resources section. Complete the survey. The information, guidance, and other statements provided by 3M are based upon experience and information 3M believes to be reliable, but the accuracy, completeness, and representative nature of such information is not guaranteed. Such information is intended for people with knowledge and skills sufficient to assess and apply their own informed judgment to the information and is not a substitute for the user's own analysis. The participant and/or participant's organization are solely responsible for any compliance and reimbursement decisions, including those that may arise in whole or in part from participant's use of or reliance upon information contained in this presentation. 3M disclaims all responsibility for any use made of such information. No license under any 3M or third-party intellectual property rights is granted or implied with this information. 3M and its authorized third parties will use your personal information according to 3M's privacy policy. This meeting may be recorded. If you do not consent to being recorded, please exit the meeting when the recording begins. (SPEECH) There are several engagement tools today. In the media player, if you do need closed captioning, that is available. If you do have any questions, and we do encourage those throughout, we'll get to as many as we can at the end. Go ahead and put those into the Q&A. That will be easier for our speakers to see the questions as they come in. But we also have the attendee chat turned on. So if you do want to chat with attendees and make any comments about the content, please feel free to chat with each other in that attendee chat section. There is a resources section where there is a certificate of attendance as well as a few other resources. Some of our previous webinars, if you're interested in listening to those on demand, those are there. And then at the end, we always appreciate hearing how we did. So we'd appreciate if you complete the survey that will launch at the end of the webinar. So again, we really are looking forward to hosting MaineGeneral. Again, today, we're going to be talking about looking for ways to advance your CDI program. (DESCRIPTION) The next slide reads, Meet our speakers. Susan Brennan, RN, CCDS, BC, RN Clinical Documentation Specialist. Cathy Rice, RN, CCRN-K, CCDS, Clinical documentation specialist, MaineGeneral Health. Gracie Gillis, RN, BS, RHIA, CDS, CCDS-O, CDI adoption specialist, 3M Health Information Systems. Moderator: Kathy Harkness, RN, BSN, CCDS, CDI client engagement executive, 3M Health Information Systems. Copyright 3M 2023. All rights reserved. (SPEECH) MaineGeneral has done leaps and bounds with the implementations that they've done. And so we're just going to be kind of having a candid conversation with some of their team, with some of our team as well. And it's really going to be diving into the financial opportunities and how they're operating more efficiently using some of those advanced reporting capabilities. So with that, I'm going to go ahead and turn it over to Kathy Harkness who is our CDI client engagement executive where she will be, again, having that candid conversation a fireside chat, if you will, with MaineGeneral. Kathy? (DESCRIPTION) The slide transforms into a photo of a beachside campfire at dusk. Kathy Harkness speaks from a video call tile below two other tiles. (SPEECH) Thank you, Lisa. Well, I'm very excited to talk today about MaineGeneral. They've got some great story to tell. This is our second fireside chat that we've done and the first one was well received. And so I'm very excited to be able to chat again with you all and bring these speakers and their stories to you today. So just to start off, I'd like to ask Cathy and Susan to just introduce themselves, giving themselves a little time to tell you about who they are and what they do at MaineGeneral. And Kathy, if you can start us off, and that would be great. (DESCRIPTION) In a tile above, Cathy Rice sits next to Susan in a conference room. (SPEECH) Sure. I'm Cathy Rice. I'm an RN. I've been a CDI for 12 years. Prior to that, most of my experience was in the critical care unit. And Sue? Yeah, and I'm Susan Brennan. I'm a CDI nurse here at MaineGeneral since the start of the program 2001 and was sort of in the float pool and all kinds of nursing prior to that. Well, I mean, what a transition from the float pool to CDI. And so Sue, since you've been there from the beginning of the program, would you be able to give us some history on how you got started and talk about the CDI program pretty much prior to when you started with CDI collaborate? Sure. As I said, our program began in May of 2001. And I was working as a utilization review nurse. I just started about that time. I'd only done it a few months. And the concept was to implement CDI into the UR role. We were two campuses at that time about 20 miles apart, and we trained with about five other small community hospitals. Credits due to our CFO at that time who had a vision. And he reached out and contacted the other hospitals. And we all shared in the cost of the training and it was done at our facility. And as far as I know, we are the longest continually running program in Maine. We had a week long class with a book provided by the consultant. We did it on paper in the book. And after a week's training, we were off and running doing both UR and clinical documentation. Our focus was on capturing the CC's and increasing the case mix index. We learned the CC's, MCC's didn't exist. Initially, we had a steering committee that directed the coding task force. We started out, we focused on a topic and provided education on that topic for both inpatient and outpatient documentation. We did the CMP tidbit posters that we placed in documentation areas, and we wrote an article in the physician newsletter. Eventually, MCC's were implemented and our first goal was to capturing the type and acuity of CHF. In 2005, we got our first software, and we used it until it was retired 15 years later. We remained two separate campuses doing UR and CMP under case management. Around 2008, new UR nurses were hired. And when they were trained, UR and CDI were separated. We had our own manager and our own physician advisors. We attended our [AUDIO OUT] All the CDI nurses were motivated to obtain it. And for many years, we all had our certification. In 2012, our manager at that time, Kathy Salucci, was selected as the CDI specialist of the year at the actor's conference in San Diego. And I was able to attend that conference. It was very special as we were the only two remaining from our initial training. Kathy had served on the committee for the first physician query practice brief and also on the revision. She taught us well how to write queries. And when we have been audited, we've always been complimented on our physician queries. In 2013, the two campuses merged into our current facility of 198 beds. We were now under HIM and not case management. We obtained 15 minutes on the schedule on new physician orientation. We developed a tip card to give to them and go over that has our names, the hospital guidelines for acute respiratory failure, for malnutrition, explains clinical validation, and other documentation tips. We give these to the residents also. Our manager began doing the coding and the clinical denials. We went over them at a weekly staff meeting. The denials merged into a weekly meeting with appointed members, including our physician advisor and coding leadership. CDI can join on a Microsoft Team's invite. ICD-10 came along and more training, and we started to work two days a week remote, and then we were back on site for COVID. Currently, we are 100% remote, and we come on site for physician education, staff meetings, and other needs. We had a large number of CDI nurses leave for various reasons-- retirement, relocation, and our software program was being retired. We needed to train new CDI nurses, and we wanted a program that would survive and keep us at the forefront as we always had been. We identified our needs, and Kathy will be sharing further the process and reasons we chose 3M collaborate. Kathy was the point person in our selection. Kathy Harkness, think you are on mute. Hang on one second. I guess we should [INAUDIBLE]. Try that. (DESCRIPTION) Kathy Harkness mouths, unheard. (SPEECH) Nope, you're still on mute. Did they hear me? All right, let's try again. Oh. Sarah, are you there? Yeah, you're there now. OK. Yeah, I was wondering if you muted all because [INAUDIBLE]. I don't know why. Go ahead, Kathy. OK. There you are. Sorry. Some technical issues. OK, so Cathy or Sue-- I was going to shoot to Cathy. But Sue, what an amazing story that was and how proud you are of that program. And so thankful to us that you're still with us and that you're still able to tell us that story. I think I was at that conference when Kathy Salucci got that award. So when you mentioned her name, I remember that name. So we're going to pivot just a little bit because it sounded like you guys have had quite the journey and are very experienced and obviously know what you're looking for when you want to grow your program and take it to the next level. Every year, every challenge that has come down, CDI and the changes that have made-- I mean, I've been in the industry for since I believe 2007. So I've seen a lot of changes myself. So Cathy, when you came to MaineGeneral and started looking for some technology and the current program where you were, what was it that you were looking for in that vendor selection process? It was really difficult when we were looking at different vendors because there's so many possibilities. And then when I found out that 3M collaborate had the MModal option, that was what really piqued my interest. We looked at immortal and it was a standalone product, and we loved it. But due to IT concerns and financial concerns and timing, we just couldn't go ahead and move forward with MModal. So this was a win to get 3M and MModal together. So that was our biggest draw for the 3M. It had the encoder which we didn't have an encoder in our previous software. So there was so much coder logic that was missing. We had mismatches that were really-- we knew they were going to be a mismatch because we didn't have the technology in our software to fix it. And we wanted more opportunities and triggers to help with prioritization, and we wanted allowance for customization in a product and ongoing customization. And 3M collaborate had all of that. So we went with that, and we've been very happy and very pleased with our choice. So basically, you had some technology, but you had a lot of manual processes at that time. Oh, it was like 80% manual. Yeah. It was laborious. It really was. Yeah, OK. So when you purchase collaborate and you set up with our implementation team and talk to getting started, what were the goals when you were implementing? We wanted something that was easier to use. So we wanted to learn the product and use it fully. We needed to streamline reconciliation process. Prior to 3M collaborate, the only way that we knew that a case was coded would be to go through our list of cases that hadn't been reconciled and look through another software to see if they'd been coded. And it was like an individual process. It was really, really time consuming. So we wanted to very much fix that. And we wanted to avoid missing things in the charts because we didn't have ways to search for them. And how many FPEs did you have at that time when you started? [INAUDIBLE] there are four? Four. Four FPEs. OK, I just want to make sure everyone knows, kind get a sense of the size of your organization and where you were in your program at that time. So what was it about CDI collaborate that piqued your interest? I mean, was there a certain functionality, was it efficiency, what was it that, when you saw the product, that really stood out? The NLU. OK, let's talk more-- yeah, I mean, I remembered you telling me that in our conversations before that. So let's talk more about the NLU for a moment. Well, first of all, for the audience, let me explain what NLU is. Natural Language Understanding is a set of rules that are built on top of what traditionally is called Natural Language Processing or NLP. Our product has the ability to use our base code set as SNOMED because it gives us a lot more richer information and the way that we're mapping the documents and being able to surface information and look for information for our customers. So with that, when you found out about that technology and all of the rules that we have, how did you select that NLU content or let's talk about that NLU journey [INAUDIBLE]? To select the rules and what was going to be queried for-- we looked at what our main queries were currently. So it was the sepsis and the acute respiratory failure, acute kidney injury, encephalopathy. All the things that we queried on, we wanted something to be able to identify and trigger those so we didn't have to go looking for them in the chart that it would be brought to the forefront. And that's how we started. And the thing that I really loved about it is that there are those rules that are created, but there's the ability to add other rules. When the CKD 3 specificity came out, we wanted to capture that. But we didn't want to have to go pouring through all the lab values to find it. And working with our adoption specialists, we created a rule for CKD 3. And so every time it was mentioned and it wasn't specified, we got an opportunity trigger that alerted us to that so we could send queries for that. So it grows with us and it has been very successful. So you told me in-- oh, go ahead. When we started, we didn't start with the NLU. The first month or so, we just used the basic product to get used to the product first. And then we added the NLU. And the thing that I liked about that as well is that over time, it got smarter. It learned how our physicians document, the language that they use. So we were getting less false triggers and more positive triggers. So do you feel like this was now an artificial intelligence that was helping me be more efficient is what I'm hearing? When you talked about the fact that you have four FPEs, you also share with me that you had two new staff. Was this something that they found was helpful to them? Or talk to me about how you were able to onboard new CDIs using this technology. You want to say? I think they found it very helpful. They are very into the software. They look at the jelly beans and all the CDI information for guidance. The code book references, the edits-- Cathy and I had learned on-- I learned on paper. And then our software program was so antique, didn't have all the bells and whistles. And they're teaching us on it. They are. They look at those jelly beans all the time. And we're like, oh, gee, we didn't know that. So it has been we're like, OK, we've got to look at the jelly beans more because embedded in the encoder are all this wonderful knowledge that we just never had the opportunity to look for. So it's not in our habit. But oh, my gosh, they're like, no, that's not right. [LAUGHTER] We're like, OK. Well, how great for a new CDI to be able to bring value right away because having them have that ease of use. And for those on the call, I just wanted to clarify jelly beans. Those are the little tooltips that are embedded in the CRS encoder. And that is connected to CDI collaborate. So when you are using the tool, you're having CDI collaborate as a workflow, but you're also doing concurrent coding using the 3M encoder that has those tooltips, the references, and all of those things. So I also use the tabular reference and I still think that there's value in understanding codes and things like that. But oh, my goodness, technology is so exciting that it has all of that at their fingertips. So how was the onboarding process? What's the difference? I mean, obviously, having all of those efficiencies and things, how would you how would you describe that process? It was much faster and there was less learning time. It used to take forever to learn all the rules. If you have sepsis and you have a UTI, sepsis is the PDX. And the jellybeans really helped with all of that. They didn't have to learn all the little rules. It was a reference right at their fingertips for that particular diagnosis. So it really went quicker. I think it was less frustrating for the newer folks because learning CDI is overwhelming to begin with. And to have that extra help and nudges to learn is really important, I think. I'm very grateful that we have it for the new folks. So was the NLU something that was also-- I mean, how are they reacting to that piece? I mean, the encoder, the tooltips were just coding guidelines and coding rules. You had mentioned that the NLU itself, when we talked about queries, how that is affecting the query process. It helps with the clinical validation when they're writing queries because when you have the NLU opportunity that's triggered, it shows you all of the supporting evidence and where it was found. So you can easily see what is significant to support that diagnosis. So you're learning through that as well. Absolutely. Having that ability to through documents and having that information tee up I'm sure would be a huge help so that you not only-- sometimes, you don't even know what you're looking for. And I used to hire CDI specialists. I told them, it's like looking for Waldo, where's Waldo every day. And sometimes, Waldo isn't even there. So it is a real challenge when you're training new people to think that way and look for things. But let's go back to when CDI collaborate was implemented. We know in our business that implementation of technology can be quite painful and it's not something that most people look forward to. But let's talk about your implementation process for CDI collaborate. What did you do to help prepare and how would you describe your implementation process because that's one of those things where people ask about? How did it go? We had our challenges with that because it was at the start of COVID. So we didn't have any on site folks to help with the implementation because everybody was in their segregated areas. A lot of stuff was done in teams, and we hadn't learned all about the connecting via computer that was new to us. Yeah. We were learning everything. So there were a lot of challenges with that as well. But there were some really wonderful things that happened with the implementation because selecting the rules-- that was really good, and tweaking them. We were able to bring a lot of our old tools to the new collaborate product. It wasn't that we bought CDI collaborate and then we had to do everything the 3M MModal collaborate way. We brought our worksheet template. So that was the same. We brought our query templates. And the way that the queries were sent to the doctors remain the same. So there was a familiarity about it which made it much easier to transition to a new software. Absolutely. And with this success and longevity of your program, we definitely wanted to build on those successes and be able to-- especially familiarity. That's a very key word, especially when you're talking with physicians. If they're used to a certain template, you don't want to change everything on them because you might not get the outcomes that you used to get just because they don't understand the new form. So I can imagine that would be very helpful. With CDI collaborate product, it also includes an adoption specialist for the life of the contract, which is pretty unique from our perspective. And it's something that we value as an organization. Describe how this person works with you to support your CDI program and goals and how critical she is in your success. She is on the line today. I'm not sure she's on the phone. She might be having some audio difficulties, I think. But I'm going to ask Sue and Cathy to talk about Gracie for a minute. Gracie is very knowledgeable and she's been a huge asset to us. We had sort of lost our education and budget cuts. And we were a strong program, but this thing's changing all the time. And Gracie keeps us up to speed. Whenever we have a question, she answers us. And she has reports at our fingertips, and she taught Cathy how to run the reports. And Cathy shares all the reports with us. Do you want to explain the reports? Yeah. And there's many different nuances with the CDI collaborate. And she helped us use it most efficiently. The reporting structure has helped us for three months. Every three months, she goes through them. It helps us improve and elevate our program because we can find out what we could be doing better with the queries. If there's something that 3M collaborate does, we can get a report about it. It's really kind of amazing. The queries, how many we're sending, how long it takes for the doctors to respond, how they're responding-- that was another thing too that we can tell now. We can tell if they are agreeing, not agreeing, if they're doing the dreaded unable to determine. We have all that information at our fingertips. So we can figure out who do we need to talk to, what education do we need to do. Right. And we didn't have any reports really. So this was huge. Yeah. And so going back to some of your newer CDIs-- were you able to use the dashboards and the-- I guess they're productivity, but they're really down to the CDI level. Is that something that you use with them to kind of give feedback? Absolutely. We run reports monthly. And so everyone can see how long they're spending in a case, how many cases they open, how many queries they send, who they're sending them to, how they're answered. So it really is effective. Especially for the newer folks, you're like, oh, I'm opening new cases now, or I'm opening a higher percentage than I did two months ago. And my queries are getting answered. And Oh gee, it's not taking me an hour to do a case, I can do a case in 30 minutes. So those little things that show progress really are helpful to encourage and foster the growth of the CDIs. And there's also an audit tool that Gracie helped me set up. So I can go through and audit charts and see what's going good, what's not going good, needs improvement. So all those things are very helpful to the newer folks and those older folks. Yeah, I have found that feedback to staff and whether it's things that they need to improve on or their successes is really important to encourage especially sometimes when you feel like you're just sitting and got your head down and you're in cases and you don't see the bigger picture. I think that really helps. I think it feeds that part of people that have the desire to do this kind of work. It's really helpful. I wish I had had that because I know when I was managing, a lot of us say, well, how am I doing, what's going on. And I had to do it manually. I had a team leader that that's all she did was audits and chasing down things to make sure that we could give that feedback not only to our staff, but to the providers as well. Providers really want to-- they're very competitive and they want to have that information as well. So having the ability to run those reports yourself, Cathy, and then have Gracie to be able to look at trends. And our adoption specialists like Gracie-- she is a CDI specialist. You can see her credentials at the beginning of our talk today. All of our adoption specialists or solution advisors, I think we have a name change coming in. And most of them have CDI credentials, they really are. And for size hospital that you are and smaller, it's like adding another person to your team. So glad you're finding it helpful. We're going to talk a little bit about the rules. I mean, sometimes understanding our technology, that can be an ongoing thing. And you mentioned being able to add new rules. And I assume that Gracie was a part of that. I think you mentioned that she was for that CKD stage III specificity. Was that something that Gracie worked with you with? Yes, she did. She helped us create that. And she was a huge help. We had a different adoption specialist when we first started. And then when Gracie came on board, she helped us go through because over time, you realize that what you thought you wanted for rules may not be what is working for you. So we met as a team, and she went through every rule that we had and made sure that it was appropriate with our guidance. And we tweaked them. And they've been working much more effectively since then. She has a very keen understanding of a rule and what it asks for because those are very concrete, and we didn't get them right in the very beginning. We didn't realize. But until you're using them and the NLU is picking up, you can't really tell. So it was good to do a look back and say, OK, what's working and what's not, and let's change them up if stuff was triggering too often or not enough. And it's been working much, much better. CDI is very dynamic. I mean, when the government rules change every year and sometimes even more frequently than that, having something that you can reach out to that person and not a 1800 number but actually a live person to be able to talk through these changes-- I'm sure I would have loved to have had that as well. So I'm glad that that's a great relationship for you. Looking back, what did you learn that you think you'd want to share with others? Oh, actually, let me pause for a second. I don't know if Gracie-- While we're still talking about Gracie, Gracie, are you able to join us and give some comments about MaineGeneral and what you've learned from them and how that's going? Sure. Can you hear me? Oh, yes, loud and clear. Are you able to hear me? Great, yes. OK, yes. Well, MaineGeneral has been awesome to work with all the way around. They are very in tune to rules and pay attention if they're over triggering. Like Cathy was saying, when we rework them, we've done that a couple of times and just even different rules like their skin lesions. I know we had a little bit of trouble getting it right. And then we did get it tuned because of their use of that a lot. Yeah, they are very knowledgeable. And I will say, one of the best of all of my customers using the NLU for their queries. So we can measure the percentage of the opportunity to queries and 66%, which is a very good number, of their queries are coming from using the NLU. So kudos to them. They're a great team to work with. Love them. Well, we do measure a lot of things because we want to make sure that the tool works for you and that you're not working for the tool. And I think that was pretty evident when we talked about implementation that you were able to bring your successes from your program and have it be complementary to the product. Those are the types of things that when we're working with customers, we're looking for. I always feel like Dr. Phil is like, how's that going for you, how's that working for you, because it's important that we don't come in and upset the apple cart. But like I said, implementations and change management, sometimes we need to change, and we want to make sure that we're a part of your successful journey in doing that. But Cathy and Sue, looking back, I was to ask this question again-- what things did you learn that you'd want others to know that you want to share with others today? One of the things that I wanted to just mention that is so, so great about this product is that it has the search function. So if you are curious about a diagnosis, you could type in any word. And anywhere throughout the whole record, it will show up where you can find it. It's really important when you're looking for ongoing documentation of a diagnosis like sepsis, you could put it in. They may have set it for the first two days, but then they haven't set it for two weeks. And that's evident just by a quick search. Or if someone comes in with CHF and you want to make sure that hypertension is documented, you can do a quick search for that. And Sue was saying this morning that she does it. If she's going to send a query, she sometimes types in the diagnosis to make sure that it wasn't documented somewhere that she missed. So she's not asking for one that-- That is there and I didn't see it. And the-- That's a great use case. Yeah, that's a great use case. And then there's a place to write comments in many, many places which are so effective in our day to day work. Opportunities aren't always just the thing that's identified. If a diagnosis is considered completely documented, you get those alerts as well. So this morning, I was doing a case and sepsis was documented. But I was able to make a comment that said the sepsis was based on CERS plus an infection. So then I know to look out for organ dysfunction that may happen and try to get it linked to the sepsis even though it's documented, more cement that. And if we are on vacation, we easily can assign our cases to our coworkers and leave lots of comments to say, hey, look for this or this one's pretty much all set, just keep an eye to see if anything goes awry. And it's very, very helpful. And we leave notes to ourselves all the time under the different opportunities and then just the day to day comments, what we're looking for or what our thought process was. So in case a denial comes down the line, we're going, OK, this is what I was thinking. So those are all things that are helpful. And to be patient with it. It takes time. The younger nurses pick up on some of this technology quicker than us older ones. But you can learn it, you can do it. And I love the fact that we were able to-- the familiarity that we could use our old templates design. The company worked with us to design the query template the way we wanted it. And one of the biggest things and lessons I learned was that you can't break anything. No matter what you do, you're not going to break it. It can be undone. I know from experience that sometimes it takes a while to do the undoing once you've really messed it up. But you've never ever broken anything. So you just have to keep plugging along and doing your best. I love that because it can be like, oh, my gosh, I deleted something, or I've clicked something and I can't get it back. That's a really good thing to share because it's not that everything can be recovered, but at least that there's this sense of it's there to assist you and not be a barrier to your work. And that's really-- and CDI collaborate was built for CDI specialists. I mean, it's not a coding tool. You talked a lot about searching and codes and diagnosis. Just to be clear, this is not a computer assisted coding tool, this doesn't have any kind of tech or those kind of things. It's basically there for the CDI to be able to use it to surface information and then select their own codes in the encoder. But I love the search key. That's an addition to the NLU rules that you turn on. So we typically only turn on-- gosh, Gracie. How many rules-- we have a library of over 500 and growing every day because in general, other people create their own rules as well. But we typically don't turn on more than 15 to 20 in the beginning. And just for example, the sepsis rule might have 13 different versions of the sepsis rule with or without organ dysfunction. So the search function gives you that other ability to search beyond what the NLU is doing for you. What do you consider your biggest wins with-- I mean, you've talked about a lot. But I'm just trying to summarize for the folks what you think is your biggest wins and maybe things that you wish were different. The reporting has been, I think, our biggest win. We're able to track our productivity which allowed us to go fully remote. So that was a big win for us. We're able to track our financial impact and it really drills down to like did you affect the severity of illness, risk of mortality, did you add HCCs, MCCs, CCs, so you really can see what you're doing and in terms of dollars too because, of course, the financial impact is important. And then it also tracks protective billing which we never had any way to reflect all the work we did with our clinical validation queries. So now when we do our clinical validation queries and we eliminate those diagnoses that aren't valid and will be denied, we have a way to measure that as well. And all of our indices have improved. Our productivity, our financial, our protective billing have all improved since we started with this. It's wonderful having that feedback. And Gracie meets with us routinely and provides us that feedback. And it's great having the adoption specialist to do that because she looks at things from a bigger picture. Sometimes we get mired in what we're doing day to day. But she has that objective view to say, hey, look at this, and it's so, so helpful. Great. So as you continue your partnership with us, what's next for your program? Our biggest challenge now is getting our outpatient CDI going again. We had started it for maybe a year and it went by the wayside. And that's what we're going to be working on next is our outpatient CDI program. Great. Well, I think that's all the questions that we had selected to go over to try to give you a nice overview of this great program for everyone to understand where they started, how they grew, where they are now. But we'd like to open it up now to any questions from the audience. So Lisa, do we have any questions from the audience? Yeah, we do have a few. So, yeah. And thank you to you all for such a great presentation. So let's get started with our first question. How long did it take your eyes to get comfortable once you went live with the technology? Comfortable. Yeah, I know. Not that long. No, I was trying to think of-- I mean, we were doing it like immediately. We had the little bit of training because as I said, it was during COVID and teams. So it's a technology that you can get up and running. So I guess we've-- It didn't take me long, and I'm the most challenged. Yeah. [LAUGHTER] We appreciate your honesty. From the perspective of the new CDIs, they were up and running and flying with it. Yeah. So if you look at it that way, no time at all. No time at all. I had no problems and I learned on paper. So I was up and going. (DESCRIPTION) The slide now reads, Questions, over a blue gradient background. (SPEECH) It is intuitive. It's definitely user friendly. Awesome. All right. Next question is how many-- and you might have answered this, I apologize. But I think you had said, how many kids do you currently have on your team. And then secondly, are you still with a 198 bed facility. We are a 198 bed facility which is amazing to us because we're like a standalone hospital. And when I look at all of these big hospital corporations, we're just 198 bed hospital and we're doing really well. We have really good quality outcomes, and we've been recognized by CMS. We're a five star for quality and patient experience. And US News recognized us as well for many different areas as well. So we are. And there was a second part to that. Was it how-- How many CDIs do you have? Four. And we have a fifth too, Kathy Salucci. But she works on denials. But we have four doing inpatient [INAUDIBLE]. Awesome. And then kind of as a follow up then, a question that just came in-- then do you have any coders as part of your CDI staff? Not as our staff, but we work so closely with them. We have a really good working relationship. But we're all ends. And we do the [INAUDIBLE], so we're working with them very closely. Awesome. Next question, so the CDI gets the trigger for specificity, not the MD? So I'm going to answer that one. So CDI collaborate is really facing only the staff. Coders can log in and use the tool to see the same information. We have another product that can nudge the provider through computer assisted provider documentation nudges that the same rules that we would send a CDI, we can send those to providers through our Fluency Direct toolbar. But that is a different product and not currently something that MaineGeneral uses. So that would be something that they could add on and send the nudges and queries through that toolbar. But that's not a product that MaineGeneral has at this time. And I say at this time because you never know, right, [INAUDIBLE]? Next question, Lisa. Sure. Staying with a product and comparison question, can you talk about the differences between CDI collaborate and how it's different from 360 and compass? Yeah, I'll take that one as well. So the CDI collaborate again was built for CDI. So it's really clinically centric. It's got clinical rules, it's teeing up clinical information. It does have a link to the 3M encoders from CRS that allows them to do some concurrent coding. But it's the CDI that would be selecting and using the encoder and entering the words or codes and it would be entered into sync up with collaborate. 360 is encompass is our premier CDI encoding platform. So that's going to be a platform that CDI and coding use together. And it has computer assisted coding and it's a different technology. And I think I mentioned during the talk that CDI collaborate is not a computer assisted coding tool. All right, next question. Going back to the NLU rules, any advice as to which ones were more helpful than others with the criteria for selection if there are multiple rules for a specific diagnosis. I don't know how to answer that off the top of my head because there's so many little nuances with each little rule. Gracie maybe could speak to it more. But I think each institution would have their own rules that would work best for them under the categories. That's how we figured when we chose what we thought we wanted. And then, oh, no, like for sepsis, let's say, no, instead of going with these four rules, we're going to drop two and add this other one underneath that. So-- I'm sorry, I don't have more specific information about that. Yeah, and we can provide guidance to have those rules are selected. We can use data if you want to look at what your query volumes. There's a lot of different ways that we can help you make that selection. Going back to the differences between CDI collaborate and 360. Are you able to use both? And also, can you use CDI collaborate on top of 360 and compass? Great question. So when 3M acquired MModal, CDI collaborate was an MModal solution and it had the NLU, the clinical evidence sheets, and the natural language understanding built into collaborate. When we joined 3M, we took that technology and we embedded that into 360 and we call it CDI engage one which is adding those evidence sheets into 360. Typically, an organization would not have both. They would either do CDI collaborate which is what we consider a product that is more suitable for customers that have maybe a critical access or a community hospital that maybe not has the needs of the CDI coder platform. So it's usually a decision that the organization makes to decide whether or not they want to purchase CDI collaborate or they want to go with, again, a larger platform product, which is our 360 encompass. And I call it our premier CDI coding product because it has a lot more features and functionality. And in larger institutions, that would be something that they probably are looking for to take advantage of. But you wouldn't need both. But what's exciting is we were able to supply the NLU technology to enhance our 360 product. All right, a question-- yeah, those were really great questions and really understanding the integration and how it all works together. Another question for MaineGeneral. Do you do 100% of all inpatients? We do not. That's our gold standard that we're aspire to. But right now, we currently do DRG payers. And we-- [INTERPOSING VOICES] --for opportunities to prioritize because we've had two new CDIs training and their productivity is increasing by leaps and bounds. But we use the opportunities to prioritize which cases we open. Well, this is a good follow up question. So do you review all payers pediatrics, OB, NICU, behavioral health? No. No. Since we are running close to time, I'm just going to ask one more question. Does your CDI do patient safety indicators, and how do they collaborate with your quality department? We do work with quality and look at those PSIs and have a communication with them. So if stuff's identified, we alert quality and we use our queries to help determine if it's a PSI or not. And there is in the template for queries, you can select what type of query. So if it is specifically for PSIs, present on admission, you can select that and track and trend that. That's available. Are you all using that as part of your program? [INAUDIBLE]. Great. There was one more question that I happen to see in the chat. Are the CDIs able to access their own metrics? And with access, do they have access to all the teams or only their own? They do not have access to their own metrics. They get the reports. So they do have access that way. They can't look it up themselves, but they can look up their own. And then no one else shares data about each other except we'll look at overall how we're doing. But they can see in the reporting, the percentage of cases that they open compared to their peers. So one might do 25%, one might do 40% of the cases. So they can measure themselves that way, but not individually. They won't know oh, so-and-so is doing more than me or-- that's kept confidential. Great. No, I think that's a great last question. Again, we do really appreciate your time today. The information was really great. So to all of our attendees, we really appreciate you joining today. This session was recorded. We will add it to our website in the next couple of weeks. Before you do close out of the presentation, make sure you do download the certificate of attendance. That's not a CEU, but you can utilize that certificate to request CEUs at an accredited association. And again, there are a couple other resources there as well as some of our previous on demand webinars. And be on the lookout. We do have a webinar that is going to be scheduled in early October where we'll host another one of our customers as well. And so again, really appreciate the information today. We appreciate everybody joining. We look forward to seeing you all again in October. Thank you. Have a great afternoon, everyone. (DESCRIPTION) The slide now reads, Thank you!

(DESCRIPTION) Presentation, on 24 for a better webinar experience. Image, a screenshot of an interface for the presentation. In the center are the slides. On the right side is the speaker bio and survey. On the left side is the media player resources and a box for Q&A. In the upper left hand corner of the screen is the video feed for the speaker of the presentation. Logo, 3M, science, applied to life. Slide, 3M's approach to providing payer solutions across the continuum of healthcare. Miki Patterson, PhD, NP, 3M, AM – PPC product owner. Jessie Murphy, MHA, 3M methodology consultant. (SPEECH) Good afternoon and welcome to our June webinar where we're going to be talking about the approach to providing payer solutions across the continuum of care. Before we get started, I'm just going to go over a couple of housekeeping items. We (DESCRIPTION) Slide, housekeeping. A bullet point list. (SPEECH) are using the ON24 webinar platform that has a lot of engagement tools. Because this is a web based platform, we do encourage you to use Google Chrome and close out a VPN and multiple tabs that will help with your bandwidth. If you are having any audio issues, do a quick refresh because that is through your speaker settings. So check those settings, do a quick refresh. And if you are having any of those issues, that typically clears that up. Again, with this platform, we do have multiple engagement sections. We do have the Q&A section that we encourage you to ask questions. We have a lot of content on this webinar. So if we have any time at the end, we will answer a few questions. But if not, we will reach out to you after to follow up with those questions. But please feel free to ask, and we'll follow up with you. Within the media player, if you do need closed captioning, that is available for you there. We also have a resources section. So if you'd like the handout for today, you can download it there. And then there's a few other guides and things like that might interest you as well. And then we also have a survey. We always like to hear how we did, so please fill that out at the end of the webinar. Like I said, we like to know how we did, so we always appreciate that feedback. If you have any questions throughout the webinar, just go ahead and pop those in the Q&A. (DESCRIPTION) Slide, 3M team introductions and agenda. A numbered list. (SPEECH) And again, because we have a lot of content today, we will get to as many as we can at the end. So without any further delay, I am going to go ahead and pass it off to Jessie who's going to introduce herself and Mickey as our speakers today and go over the agenda. Hi. My name is Jessie Murphy, and thanks for joining the presentation today. Can you hear me OK, Mickey? Wonderful. OK. Well, again, so my name is Jessie Murphy. I'm a methodology consultant with 3M. I've been with 3M a little over a year, but I've been using 3M's products for almost a decade on the payer side of the business. And in my work, I support implementation and rolling out of these products for payers and provider clients. And I'll let Mickey introduce herself. Hi, everybody. My name is Mickey Patterson, an orthopedic nurse practitioner by trade. I do work for 3M and have been part of building the ambulatory potentially preventable complications software. So I will talk to you a little bit about that in a little minute. OK. So today, we're going to do a quick overview of the 3M client experience. There's a lot of information that goes into these slides, so bear with us as we go through these pretty quickly. We're going to talk about facility payment and quality group or application, and then how you can add on population health groupers and value based care to that. We'll then review bundled program analytics, and then Mickey, you will end with going over our new ambulatory complications software. (DESCRIPTION) Slide, the 3M client experience. A square diagram with four quadrants that are labeled as states/international, payers, business partners, providers. (SPEECH) All right. So the 3M client experience. So 3M has a decades of experience in health information systems as far as working with payers, whether that be coopers or payment transformation or that sort of thing. We do a lot of things with providers. Providers are very familiar with our products not only from an HIS standpoint, but we are ingrained in every form of the hospital. In addition to that, we have partnerships internationally and in state agencies. We do work with NHS. We have Mickey was just talking about working with some in Puerto Rico this morning. So a lot of things internationally, and then we have business partners. So companies like EHRs like Cerner is a partner of ours where we will embed our products into theirs. (DESCRIPTION) Slide, methodology consulting, areas of expertise. Four squares labeled value-based programs, facility payment and quality, population health, strategy. These lie over a two directional arrow that reads integrating 3M's grouper methodologies. (SPEECH) And so that's a high level overview of where we're at. As far as what we do, we have a broad spectrum of things that we provide from a methodology consulting perspective, everything from more developed value based care programs all the way over to helping with strategy. So we'll talk about a lot of these things today. But from a value based perspective, we might be helping build an ACO program or implement budgets for a very developed program. We might be working on payment transformation, so facility payment quality. We might be doing something along the lines of doing an EAPG or DRG update. We support all of those type of things. And then we do broader things such as population health where we might be looking at something like social determinants of care or we might be doing something along the lines of creating a population health program for a specific group of people, such as a maternity program. We do a lot of things with strategy and all of those three columns that I just discussed, but we do have things that we do with strategy that are on its own, such as helping develop products along with our clients. This is a list of the groupers that we provide. We (DESCRIPTION) Slide, 3M clinical classification methodologies. A table. (SPEECH) do a lot of things, but these are the groupers that we're going to specifically talk about today. We're going to focus a little bit on some of our payment groupers, so you'll hear me talk a little bit about APR DRGs EAPGs, CRGs, which is our clinical risk grouper. We also do offer assistance with HCC, even though it's not on here. We are going to talk a little bit about all of our quality groupers, so preventable complications both in the inpatient realm, and then our new ambulatory that Mickey will speak about. We will talk about our potentially preventable. So that's a group of softwares that would include readmissions, admissions, Ed visits, and then potentially preventable services, which would encompass a large bucket of things such as labs, imaging, things of those nature. (DESCRIPTION) Slide, facility payment and quality grouper application. (SPEECH) OK. So really, let's start with the backbone of what we do. So (DESCRIPTION) Slide, 3M APR D.R.G. and E.A.P.G. groupers. (SPEECH) all of our groupers are in some way developed from these payment groupers. So APR DRGs and EAPGs are widely adopted payment groupers. Many of the people on the call today probably use these in their day to day payments setups. I think that these are adopted by over 30 states, whether it be APR DRGs or EAPGs. They are more granular payment setups. So they provide a support system for stabilization to decrease payment variation. So you can see when you look at a chart where a client had a different payment model before and they might have implemented EAPGs where it might have been up and down, and then we implement the EAPGs and settles over a period of time. And it allows a payer to increase payment by a net neutral payment set up. In addition to that, it can help refine the payment process. There's a lot of editing that can be implied with EAPGs. And then also with APR DRGs, and EAPGs, they can really help push programs towards that value based care, offering more of a payment based on quality and efficiency than just fee for service. (DESCRIPTION) Slide, Medicare pricing modeling. (SPEECH) In addition to our payment methodologies, we actually do offer Medicare pricing modeling. So that be inpatient, outpatient, and critical access hospital as far as the modeling that we provide. So a lot of our payers that are getting EAPG pricing or APR DRG pricing from us will also get Medicare pricing, which will help them make contracting support decisions. What percentage over Medicare are we paying? Do we need to rein that back in or are we not paying enough? So it's just another point that we help when it comes to payment. (DESCRIPTION) Slide, impacts of 3M APR D.R.G. and E.A.P.G. groupers. (SPEECH) So we'll take a lot of the information that is provided from running the APR DRGs, EAPGs and we will help our players set benchmarking. So what payment are we looking at for a peer group? So all of the academic medical centers and a particular network, what are we looking at for payment on them? Who's above? Who's below that? And same thing with rate setting. We help them set a base rate for the APRs or EAPGs. And then based on the acuity of the services, there would be a weight that would be more or less than a standard rate that is provided. That's then used for contracting, and also as I mentioned previously, there's claim editing pieces to these softwares. (DESCRIPTION) Slide, 3M approach to facility programs. A circle broken into three parts labeled meaningful measurement, clinically based and risk adjusted. There is a clockwise arrow around this. (SPEECH) And then on that-- so on that payment side of things, we add on to this quality groupers. And so we have a standard structured payment model that provides some stability. And then on that, now we can say, OK, let's look at the quality piece of things. Let's look at readmissions. So our potentially preventable readmissions are different from all cause readmissions of Medicare. And the fact that they're clinically based-- it's a clinically based software looking at readmissions that are clinically related to a previous admission, it is not going to tag all readmissions. It's going to say, hey, this person was in here for COPD. A week later, they fell and they broke their leg. That's not clinically related. We are not going to tag that as a readmission whereas an all cause readmission software would. But if they came back a week later and they were having some kind of breathing exasperation or pneumonia, or something of that sort, that might be clinically related because they're both pulmonary admissions. And so that's the value of the PPR readmission software. It's one of our most popular facility softwares. On that PPR, you can add the PPR-ED. Does the same thing as the PPR, but it's looking for people that didn't necessarily come back and get readmitted, but they had or came back and had an emergency department visit. In addition to that, we have our inpatient potentially preventable complications. This is looking at complications that might occur in a hospital. We obviously know the complications that are going to occur no matter what you do in a hospital because it's just the nature of a facility. But the value of our software just PPR, PPR-ED, PPCs, and as Mickey is going to talk about with AM-PPCs is that we use those bedrocks of APR DRGs and EAPGs to set a standard for a network or for a state or for national network. So you can see and compare yourself to your peers. So yes, we are seeing that we have complications, but are our complications higher based on our risk adjusted population to peers that look similar to us? And so an example of a potentially preventable complication might be someone that got an infection or had a sponge left in them after surgery, so on and so forth. But again, the value is their risk adjustment. They're clinically based. We have providers that look at these things and make decisions about if what we're doing is clinically accurate. I won't go into AM-PPCs because will talk about that later. (DESCRIPTION) Slide, impacts of mental health on payment. Three bar graphs of different years, 2019, 2020, 2021. Text, impacts of presence mental health and substance abuse PPR allowed PMPM. (SPEECH) Here we have an example of how we might use a combination of our groupers to provide insight for risk adjustment insight or building a standard for how you might look at clinical transformation on a network. So this is-- using a couple of different groupers, this is looking at potentially preventable readmissions, which obviously use APR DRGs. It's looking at a chronic population. So it's looking at people that have a chronic clinical risk group so it's risk adjusted, and then it is comparing people that have mental health and substance abuse or readmissions to those that don't, but they're all a chronic population. So this is a common analysis that I've seen done with ACOs where they're trying to say, OK, well, what does mental health look like on our network and how can we set a baseline for improvement? So you might look at something like this, create a program, and then look at it in the future to see what kind of changes you have made. (DESCRIPTION) Slide, PPC network hospital profile. Three tables labeled percent difference from expected across three network hospitals, percent difference from expected hospital B by PPC group, percent difference from expected hospital B by complication. (SPEECH) This is an example of PPCs and how we would do an analysis on PPCs from an inpatient perspective. So here, we are looking at a hospital network that has three different hospitals. We've got a hospital A, B, and C. They have around 3,500 to 5,000 candidate cases each year, so that's looking at their total admissions each year. And we're looking at the number of cases, so observed cases where it's an actual admission that had one or more PPCs on it. From that, we have an expected case count that is created based on the risk of that hospital and the patients that are being seen at that particular facility. We compare the expected to observed as we do with all of our software, and look to see where these hospitals are performing against expected. You can see in this example, the hospital A and C are performing below expected. So they had less complications than what was expected of them. So below expected is actually a good indicator for a hospital that they're performing well. Hospital B is at over 17% difference from expected, which indicates that they had a significantly higher amount of complications than what was expected for their case mix. With the PPCs, you can dive in a little further. There are groups that they are grouped into, such as perioperative, you can see here, or infectious complications. And so when we dive into hospital B here, it looks like they are having a significant amount of complications that are related to infections. We can then dive in that into that even further from an inpatient standpoint and see what complications specifically was driving that from an infectious complications group, and you can see that C diff popped for this. So obviously, we're starting at a high level looking at performance across hospitals diving into groups of things, and then looking at specific items that are causing issues for that particular facility. That's not to say that facility has done something bad. There could be several things that cause this particular issue, but it's a great way to highlight that area for clinical transformation possibilities. (DESCRIPTION) Slide, population health, VBC grouper application. (SPEECH) So when we take this a step further from facility, we move to more of a population health or a value based care model. And (DESCRIPTION) Slide, the foundation for value-based care. Three circles labeled quality cost and risk. (SPEECH) the bedrock for anything when it comes to transformation, if that's in health care or any type of service, is variation. And so I'm a fan of Deming. I love all the things that he says, so I always use this quote to say, uncontrolled variation is the enemy of quality. And really, what that's getting at is we always need to look at where people look different, whether that be where they look different from a good perspective or a bad perspective, we can learn from both. But really, that's important when it comes to the foundation of value based care. The foundation of value based care is really built on three big pillars, and that is quality. Quality is always number one and most important because we don't want things to be just be based on costs. We want to make sure that we're actually improving the care for the people that we serve. So from a quality cost and risk standpoint, 3M has a lot of options and things that we provide from a consulting perspective to support payers and providers in developing strong value based care programs. So obviously, I started to mention our preventable groupers, how we compare those with expected rates. We can then look at things such as preventative care, chronic care management, consistency of care. And when I say consistency of care, I'm talking about, who are we attributing this care to? How often are they seeing that patient? Are they really driving the care, whether that be primary care or specialty care? We have a lot of things that we do to help build value based care around those things. In addition to that, we combine that with quality. So we do a lot of setting for budgets, we look at trends, and especially looking at preventable from a cost perspective. And then you always have to answer that question that every provider is going to say, well, my patients are sicker. And so we have to account for that. And so risk is always really important on any value based care because we need to make sure that we are accounting for a higher acuity a case mix population. So we have our clinical risk groups that have started to touch on that we can provide from a risk standpoint there. We also do provide consulting around HCCs. We will create budget risk scores and we'll do population segmentation to help with the risk adjustment portion of value based care. (DESCRIPTION) Slide, program types. A flowchart that reads from left to right quality P4P programs, shared savings/risk, bundled payment programs, capitation models. (SPEECH) So our clients are all across the board as far as where they are from a payment model. When it comes to value based care population health, we might have people that are just starting it from a quality P4P type model, so they're just dabbling in those upside risk contracts trying to get into that ACO game. All the way into people that are in capitation models where we are really full force bringing that provider into the fold and trying to really reduce that cost of care. And (DESCRIPTION) Slide, foundation of population health and VBC program design. A flowchart that reads from left to right client background, client program goal, program design, model and reporting development, program implementation and support. A two directional arrow beneath all of this that reads 3M population health and VBC consulting. (SPEECH) so we do have a broad spectrum of experience across the value based care models. So what does that look like? So we spend a lot of time getting to know our clients, understanding their program goals. A lot of time is spent and the program design, which I'll go over that in just a minute here. We look at what do you want from a reporting and model perspective, that's really going to help your program goals, and then we offer a lot of ongoing support to make sure that we are running things through the group or appropriately, that we are pulling together insight for consulting. So (DESCRIPTION) Slide, program design. (SPEECH) when we look at program design with a payer, going to be looking at things such as eligibility of services, exclusion things. Obviously, the typical membership questions that you're going to ask in a payer world. On a business, what type of member months are we going to recommend for these type of things? Which physicians? I think that's something that people don't always think about. So we spend a lot of time on who is included and who's not included from a physician standpoint, and then obviously, the program design itself. So how are we going to set budgets, periods, time for things like that. The (DESCRIPTION) Slide, risk stratification and segmentation. A graph. (SPEECH) typical things that you would set in a value based care model. This is a typical graph that we provide to all of our clients. I've seen it done numerous ways. But really, it's just trying to show how we add that quality aspect in the care model. In this particular example, we're looking at risk stratification. So we're using those clinical risk groups. We're looking at the population and trying to dive into a segment of the population where we might be able to make an impact, whether it be from a care management perspective or some population strategy. So as you can see here, we are looking at people with the highest number of potentially preventable dollars compared to their total medical spend. So you can see obviously, the healthy population is that really bright blue bucket down there as the bottom, the biggest bucket, lowest cost. But then this is where we can dive in to really identifying those complex chronic-- multiple chronic type populations, people with a lot of comorbidities. How do we find them? How do we put our arms around them and making sure that we are helping reduce their cost of spend and improving their quality? This is the people that they say, that 20% of people spend 80% of dollars, this is how we find them. (DESCRIPTION) Slide, harnessing potentially preventable events. (SPEECH) So this dives into that nitty gritty level here. So I'm talking really high level. I've gone into some of those groupings and how we do these things. But really, what does this mean from an individual level? How are we really improving lives with some of the things that we do at 3M? And so I just gave a few examples here of how we might identify different people or populations of people with these groupers. So we have an example of a paraplegic that was inpatient complication due to a lack of appropriate wound care from a hospital stay. That person might be identified by potentially preventable complications on an inpatient-- from an inpatient setting. A new mother that was readmitted for wound complications following a c-section 10 days prior. This the group of people that I think are often missed as far as potential people that we can put programs around. So this would be identified as a potentially preventable readmission. The last example is one I've seen time and time again from a rural population perspective. So communities struggling with high pediatric Ed utilization where they would be more appropriately treated in the office. A great example of this is, I was working with a client once. They had a ton of Ed utilization in the community, but they really hadn't done any analysis to what was driving that until we did some roll ups using that potentially preventable Ed visits, did some slices by age. Realize that the most common diagnosis that was being treated in their ER was otitis ear infections. And so what do we do from that? Why are kids that are having ear infections being seen in the Ed so frequently? Well, then you have to think about access and why are we driving them to the Ed? Is it something that we're actually saying from a provider perspective? So a lot of things that can be done there from an analysis perspective. (DESCRIPTION) Slide, supporting payers and providers in population health. A table. (SPEECH) And then this is just a short story that I like to tell because it's so powerful as to how clinical data behind the scenes can improve lives without people even knowing that it's being improved. And so I want you to focus on the left hand side of the screen for just a moment here. And when we think about how we do presentations from an ACO, I don't know how many people have been-- in the room have been in ACO conversations, but a lot of the time, it's around cost. And so we spend a lot of time talking about quality, and cost, and if you do this, you can save this type of money. So you might see something like the columns on the left where we say, oh, well, if you reduce papers by 5%, that would equate to about 13 cases, and that's a $267,000 saving. I've seen that presentation so many times in my life. But who are those people and what does that mean. So on the right, this is actually an example from a client. This is made up data, but it was something similar to the lines of this. Where a patient was found-- she was a primary care ACO, and the patient was being seen by the local hospital. They looked similar to hospital 2 in the scenario, and they were also seeing this patient for primary care. From their data alone, their hospital HR data, they were in their primary care because they were a primary care was owned by the hospital, they were able to see that they saw the patient three times for primary care. They had three hospital visits and two Ed visits for this. They didn't see everything else that was going on with this patient. So in working with the payer in an ACO arrangement, they were able to see that this patient actually had 30 hospitalizations. Almost all of them were considered a potentially preventable admission or readmission. They had a couple visits with a specialist that was connected to another hospital and several home health care claims. But really, no one was taking accountability for this patient. This patient was bumbling around between hospitals, really lack of care management that was going on, until the software identified this patient as someone that really needed to be captured on to sooner rather than later. The pair and the provider worked together with care management, and that patient had less than 10 admissions the following year. That is 20 less admissions, and it's something that improves someone's life in a way that they don't even know that a software did that for them. So I just-- I think it's really important to understand that although we think of things in terms of the left hand side that these are people, and we are impacting lives. (DESCRIPTION) Slide, analyzing drivers of utilization. Four rectangles labeled disease cohort, geography, population trends, quality of care. (SPEECH) So in addition to the actual outputs from the software, we'll think about a lot of different things when it comes to consulting. We will think about different disease cohorts. So I've had providers that have wanted to look specifically at a disease state. So hey, we want to look at our COPD patients. OK. Well, we can do that. We will identify that from a primary EDC, which is an output of clinical risk groups, and we can look at that across all of our different groupers to see what is happening with that particular population? We will look at specialist ordering patterns. We will look at medication adherence, which I'll go through in just a minute. Things such as geography, population trends, and quality of care. I have already touched on a lot of these things, but we don't just look at the output of our groupers in a vacuum, we really think about what that means on a larger scale. (DESCRIPTION) Slide, PPS analysis using 3M E.A.P.G.s and rehabilitation service line. A bar graph. (SPEECH) So for this example here, this was looking at some analysis where an ACO was having some trouble with really high rehab spend from an EAPG service line. On this particular example, they were struggling to say, well, why is it that our PMPM is so high on physical therapy? And so through a combination of analysis using our potentially preventable services software, we actually realized that it really wasn't that ACO that was driving that cost. People were using physical therapy outside of the ACO even though that ACO offered physical therapy, and that was driving up that PMPM spend. And so in doing analysis, you can find that it really was the referral pattern of the primary care, and so the ACO spent a lot of time actually trying to drive those referral patterns back into the practice to reduce care-- the total cost of care because these patients were using facility physical therapy, which was double the cost of the office physical therapy that was provided by the ACO. (DESCRIPTION) Slide, providing insight to drive clinical decision-making. A table labeled asthma medication adhere example. (SPEECH) This is another example of clinical transformation that I've seen done through 3M. So for some of our more advanced clients, we can offer some medication adherence reporting in combination with our groupers. And so this is just an example of how you can look at rates of medication adherence and how that might impact Ed utilization. This is just fictitious data to show how this could be done, but I have seen this done in ACOs where they'll look at adherence rate. So you'll see in the middle column there, this provider a 75% adherence rate, but they have a PKPY of 96 for potentially preventable Ed visits on that asthmatic population, whereas provider C has an 84% adherence rate and a much lower potentially preventable Ed visit PKPY. So it's just an example of how you can look at multiple combinations of data and provide correlations to support clinical transformation. (DESCRIPTION) Slide, bundled program analytics. (SPEECH) So I'll end with just our bundled program analytics. So (DESCRIPTION) Slide, why are episodes important. (SPEECH) we do have episodes that we provide. So those are really helpful. And so PFE is the grouper that we'd be talking about when it comes to episodes. That's our patient focused episodes. And so that software is looking at the continuum of care. So we're looking at a particular cohort of people or a particular event, and then we're rolling up the cost of that care over time so that a payer or provider can look at bundled payments. (DESCRIPTION) Slide, case study, maternal care. A timeline with delivery in the center, antepartum care on the left, PFE cohort, and postpartum care on the right, PFE event. (SPEECH) So this is just an example of that from a maternity perspective. So you can just see how we could have a cohort where we would look at the ante partum care. We can look at the delivery, which you can pretty much do with the DRGs, then we can also add that in and look at up to 90 days following that delivery in an event window. And so when you do something like that, you can create a profile of an episode. So (DESCRIPTION) Slide, event profile of allowed amount by episode window. Two bar graphs that both have money on the vertical axis and compare the difference in price between vaginal delivery and cesarean delivery. (SPEECH) you can look at, OK, well, for this particular network, we're seeing that the average vaginal delivery is like 12,650 whereas the cesarean is coming in around 21,420, which anyone could really do that from a payer perspective. But then to break it out between ante partum, the actual delivery, and the trailing, and then to set a rate for a network and compare it across providers, that's where you really get into the potential change for clinical transformation. (DESCRIPTION) Slide, cesarean rates compared to patient acuity. A bar graph. (SPEECH) And so this is an example of how that can be done. This is a complex example, but we're looking at a group of fictitious ACOs here that are OB/GYN groups. We are looking at, in the light blue, the rates of intermediate and high risk pregnancy. So we're looking at these higher risk pregnancy populations, and then we're comparing that to cesarean rates. And so what you can see here is that community care really stands out in this example because they have the lowest number of intermediate and high risk pregnancy and delivery, but they have one of the highest rates of cesareans. So it's not saying that they're doing anything wrong, but it is saying that they look different from their peers. So it might be something that we want to dive into. So it gives you an example of how we can break into those bundled care arrangements while doing some analytics on that to show where providers are standing out from their peers. So there's a lot of information at one time, but I will turn it over to Mickey now to talk about our new AM-PPCs. (DESCRIPTION) Slide, 3M AM – PPCs. (SPEECH) Thank you very much, Jessie. That was great. So I'm going to talk a little bit about ambulatory potentially preventable complications. (DESCRIPTION) Slide, current opportunities and challenges. (SPEECH) And the reason that 3M started looking into this is that as you probably know, more and more procedures are being done in the ambulatory care setting. There's more people buying ambulatory surgical centers, ASCs. And if we have such a high level of procedures being performed but we still don't have outcome measures, we don't have safety and quality measurements. So as 3M, because we love to do these kind of challenges we can look at data and start to slice and dice it to find out what's actually happening. And we know that complication is what is going to impact costs. So I'll talk a little bit about that in a second. But (DESCRIPTION) Slide, 3M AM – PPCs. (SPEECH) first, just to define what an AM-PPC is, it's a harmful event like an accidental laceration or a negative outcome-- infection, bleeding sepsis. It has to happen after an elective procedure. So we're not talking about any procedures done in an emergency room. It has to be elective scheduled, and it has to be performed in ambulatory care setting. And when we started to look at this, we wanted to make sure that there was a clinical plausible relationship between that procedure and the complications. So we weren't looking for progression of disease like diabetes or A1Cs getting worse, we're not really looking for that. We're looking for something related to the procedure. (DESCRIPTION) Slide, logic to developing 3M AM – PPCs. (SPEECH) So in order to do this, we did a several pieces of logic, and the first one was to divide into procedure groups. So there are some 2,900 different procedures that we believe gives a patient a risk, meaning somebody put a knife, a needle, a tube, anesthesia or medication into a person for a procedure. And we started in atomically. So we started with what is being done on fingers hands, wrists, elbows, and we went to do more in terms of complexity or risky. And I'll give you an example of this. If you had a trigger finger done, it's a very low risk procedure. However, if you had a total shoulder replacement, that's a much more complex and high risk procedure. We can dice this by service lines, and I'll show you a little bit about that. And what we did we started looking at these procedures, and we chained them out for 30 days because we wanted to see what kind of complications were happening. And just an example of procedure groups-- colonoscopies, upper GI endoscopies, cataracts, things like that are the procedures we're talking about. The next level of logic that we put in was in complication groups. And there are some 4,500 complications that we could plausibly relate, but we wanted to keep them in groups. So they're are lung type complications. They're are bleeding type complications. They're infection type complications. And we wanted to utilize time limits as to when that complication would or could show up or when we don't want to count it anymore as in related to that procedure. (DESCRIPTION) Text, four complication types. (SPEECH) And then finally, we wanted to find out, where were these complications showing up? And this is probably the most illustrative piece that this component can deliver. We saw that there were complications showing up in the emergency department, complications that caused an inpatient admission, and there were complications that were seen in outpatient. We have a nice little special little logic that actually can look into a type 4. And if you know anything about Medicare, if you do a procedure in that outpatient procedure and they are admitted within 72 hours, you cannot bill for that outpatient procedure. It gets rolled into that inpatient claim. So we have a way to actually look for that. And when we did, when we started to look at the data, we wanted to make benchmarks. But we wanted to make sure they were significant complications, so we only used emergency department, visits, and in-patient admissions as part of our Medicare and national benchmarking. (DESCRIPTION) Slide, 3M AM-PPC, risk adjustment approach, First at procedure level. Age, disability status, oncology, correlated with chronic conditions. (SPEECH) And what we found out when we looked at risk adjustment, we first wanted to do risk adjustment by the complexity of the procedure. So it would be apples to apples, anybody doing that procedure. But our data showed us that age made a difference. So if you are over 75, you had higher risk of complications. If you were over 85, you had higher risk of complications. We also noted that if you were under 65 in our Medicare data, which mean that you were in disability status, there was higher complications associated with that. Also in our data, as we expected, if you had an oncology diagnosis, there were five of a procedure groups that had higher risk of complications, but not all of them. And I think this is important, and we'll get a little bit to that in a minute. But we also saw that there was a correlation of these A procedures and B complications associated with chronic conditions. (DESCRIPTION) Slide, samples of complications. (SPEECH) So some simpler complications-- and I'm showing you how we group them, but there is a pneumonia in lungs. There's aspiration, there's pulmonary embolism. There are also bleeding complications like acute anemia or hemorrhage hematoma or venous thrombosis. An infection and mechanical. And by mechanical, we actually made it specifically to the procedure being done, and I'll show you that in a sec. (DESCRIPTION) Slide, sample procedure directed complications. (SPEECH) And this is when we wanted to actually make some of the directed complications to that procedure, and I'll give you this example 101 and 102 are post procedural infections of the eye and adnexa. That would only be an AM-PPC for procedures of the eye and adnexa. Same thing with GU or vascular or cardiac. We're not going to look at eye infections after getting knee arthroscopy done. That makes sense. (DESCRIPTION) Slide, top 20 PSGs 3M AM-PPC rates 19/20 Medicare FFS HOPD. A table with the columns PSG, description, at risk procedures, ED complications, IP complications, AM-PPC rate. (SPEECH) So what does this look like? And I'm going to just take a minute to go through this eye chart of data to show you what the top 20 procedure groups that we found in that 2019, 2020 Medicare fee for service hospital outpatient, because I like to tell you you, can use this for any data you want, . But I want to just report to you what this particular grouping of data. It's 100% claims. It's over these two years. And when we looked at it, look at the first year, we're talking upper GI endoscopy procedures was a number one procedure done for these Medicare patients in '19 and '20. There was a million and a half of these procedures done. Out of that, some 7,500 or so were seen in the emergency department with a complication directly related to that upper GI endoscopy. Of that group, there were 23,000 unplanned admissions for complications related to that upper GI endoscopy procedure. So we have an AM-PPC rate for that procedure group of 1.9%. And just to show you how this methodology worked, we knew upper GI wasn't really a very complex or risky procedure. But think about a cataracts, very low risk. There's almost a half a million done, and only 114 went to the Ed with the complication, and less than 1,000 were admitted with complications from that cataract procedure. So that left a rate of 0.2%, really makes sense to us. And to further show you this, something very complex, a dialysis shunt procedure. This procedure is only done on patients that have kidney failure. So that chronic disease is grouped because of this procedure. And we know they're very sick individuals. So now, there's a quarter of a million of these cases done in those two years. And 7,000 of these go to the emergency department with a complication from that shunt procedure. And some 11,000 are admitted for complications, and that raises that complication rate to 6.7. And this is important that we now have national benchmarks to look at these seemingly simple outpatient procedures. You think, it's really no big deal. I'm going to have it. I'm not being admitted. I'm going home. But there actually is some complications associated with that, and we're now able to illustrate that. (DESCRIPTION) Slide, data by procedure subgroup PSG 70 upper G.I. endoscopy. A bar graph. (SPEECH) So we looked at that upper GI endoscopy. Remember we saw there was a million and a half of these cases, and we have that 1.9% risk. Well, our component is actually able to tell you what they came into the hospital for. So we know the vast majority came in for a GI peritoneal complication or significant bleeding. Now, these people had a procedure, they went home, and they came back and either came to the emergency room or were admitted for this particular complication. So you can now not only know how many complications you had, you can tell what that complication was so you can start to do something about it. (DESCRIPTION) Slide, data by facility/provider gastroenterology complications. A bar graph. (SPEECH) And we can also slice and dice this data another way. So if we looked at those upper GI and we went to a higher level, the service line of gastroenterology procedures-- so there's about five or six procedure groups in here, we can look at how many complications happen per facility. So if hospital A or outpatient group A might have had 300 complications from these procedures in that service line. So it gives you a way to start to compare yourself to peers. (DESCRIPTION) Slide, data by service line and PSG. A flowchart. (SPEECH) And we're also able to do it by service line and by procedure group, and here's an example. I love this because I'm orthopedics. And there are people who do spine that are orthopedic, and there are people who do spine procedures that are neurosurgery. They do a same kind of spine procedure, either a cervical spine or a very complex cervical spine fusion procedure. Now we can compare apples to apples. How many procedures were done? How many were seeing the Ed? How many were inpatient? And now we can actually see in that procedure group, what the complication of having these outpatient procedures are? So if you look at the bottom lumbar and sacral spine fusion, it's almost 3%. 3 out of 100 are going to come back to the Ed or be admitted for a complication from that outpatient procedure. (DESCRIPTION) Slide, top 10 service line ortho PSGs, Medicare FFS 19/20. A table with the columns PSG, PSG description, cases, ED counts, IP counts, OP counts, ED and IP counts, completion rate. (SPEECH) We can look at it by service line. So in that Medicare data, we put all of the orthopedic procedures. And the top 10 orthopedic procedures that were done, the number one was total knee arthroplasty. So this is a total knee joint. There are 231,000 of these being done. 2,600 are coming to the emergency room with a complication from that total knee. 3,000, almost 4,000, are being admitted to the hospital with a complication from that total knee. We also can report how many were seen in outpatient, which is helpful if you're the providers or the payers to find out, where are they going to be seen? But we can also see that complication rate, 2.8%. So you say to me, so what? (DESCRIPTION) Slide, complications of PSG 13 total knee arthroplasty. A bar graph. (SPEECH) We can find out, what did those total knee patients come into-- what complications do they have? The number one complication was post hemorrhagic acute anemia. They bled and they had anemia. This is really important because when you see what the complications they're coming back for is now for you to look at your processes and procedures to see if you can prevent that. And I'll give you a little example. Total knees were done in the hospital. And in my early days, they were admitted for two weeks. They had physical therapy. They had labs. They had all kinds of nursing visits. And we would watch them the next day and check their blood to see what their hematocrit was. If their blood was low, we would transfuse them. If it was a little low or if they had low blood pressure, we'd watch them another day, check the labs again. But now this is an outpatient procedure, and we're sending them home. If that person had anemia, they might have chest pain. They might have fainting. They might have palpitations. They might just be really fatigued. And if they went to the emergency room, that surgeon that did that total knee is not going to know about it. If the emergency room is attached to the group that does the total knees, maybe they'll hear about it. A note might go to the primary care doc. A note may eventually get to that surgeon, but they don't actually really know the volume. And now you say, so what? So what do we do about it? Well (DESCRIPTION) Slide, prevention, post hemorrhagic anemia/acute anemia. Four rectangles titled preop, intraop, post op. Each one has a ballpoint list. (SPEECH) if they told me that we had patients that were having anemia coming to the Ed or being admitted, I would go back and look at my practice. Are we doing the best practice for preventing that anemia or hemorrhaging? Do we look for preexisting? Do we do things pre-operatively? Do we do things intra-operatively? Do we do things post operatively? And we may change practice. We may now have somebody do a hematocrit for a home test after their discharge knowing that we have this complication. So it all kinds of feeds themselves. (DESCRIPTION) Slide, driving network performance by site of choice state level. A bar graph. (SPEECH) Another way that we can look at this, and I think this is really, really telling, is, we can look at this as state level data. And if you look at this chart, we're looking at various facilities, different providers, and what their expected outcomes are. And if you look at the center, this is 0%. So this is where the expected complications lie. It's not zero. But when we look to the far left of this chart, we actually see-- this is an orthopedic ASC that did 70% better than expected for outcomes. And if we look over to the right, this orthopedic ASC in this seam, I'll say state, had 70% worse than expected for outcomes. So when we see this, we'd love to know this data. Everybody on the right hand side of this, if we improve them at all, if they know that they have these complications and they can start to get towards expected, we're going to have better care, and we're going to have lower costs. (DESCRIPTION) Slide, performance across health systems. A table with the columns at risk cases, expected rate, actual rate, actual/expected. (SPEECH) So another way to look at this, if you are a health system, you can actually look at your hospital's actual to expect it if you did 33,000 cases, and you would come up to close to expected. One is the expected rate. You could also drill down to all the hospitals in your health system. Your main hospital is doing good. Your ASC is doing really good, and perhaps the opportunity is in the regional hospital number two. (DESCRIPTION) Slide, network scorecard example. A table. (SPEECH) You can look at it by provider or facility. You can look at how many cases they are doing, how many are going to the Ed, how many are being admitted, and even how many are being treated in an ambulatory setting? And it may be less complications. It also may be less costly. So as a network, you might have a different actual to expect. Your network might have a 1.2. So anybody doing better than 1.24 in your network is actually doing better than expected. But it's good that we can see facilities that are doing better and maybe learn from them. (DESCRIPTION) Slide, state level data by service line conservative costs of complication. A table. (SPEECH) And this is just another way to slice and dice the data. This is actually a state level-- it's using that Medicare database that I have. And starting to look at a really, really conservative cost of a complication. When we do Medicare, we use very, very small numbers such as an ID visit cost you $679, average just for that visit. An inpatient visit is $12,000, average just-- so utilizing those very, very conservative costs. If you took some of those 1140 for complications from urology and improve, there's a potential of 3 million or so dollars. Probably way more than that because these are really conservative costs. (DESCRIPTION) Slide, financial analysis of potential associated savings. A bar chart. (SPEECH) And then finally, look at the analysis by, where does that service line and provider send their complications? Are they inpatient or are they ambulatory? So if you look at this teal color at the end, this represents the patients that had complications that were treated ambulatory setting. And it may be that a provider or service line just says, oh, go to the Ed or go to the hospital with this problem versus having a program set up to see a patient urgently if they have a complication. (DESCRIPTION) Slide, summary. A bullet point list. Image, a blue gloved hand rests on a patient's hand. (SPEECH) So basically, an AM-PPC can really, really start to illuminate the issues that are out there that we never knew before. It's going to help us promote patient safety in that outpatient. We might even use it for standards of care or, as Jesse was talking about, as a foundation for value based care. What is happening with this facility or group compared to what is expected? And really, if a payer were able to give this data back to an ASC or a provider or a group, they would finally see what those complications, and they would take action to improve. You wouldn't even have to try to do something, just provide that information, and I think that would be really helpful. (DESCRIPTION) Slide, check out the resources and complete the survey. (SPEECH) So I think that's all I have to say. Lisa, do you want to come on back. Yeah, awesome. Thank you both so much. Anytime I hear all of this content, it's just so incredibly impressive with just the knowledge. So it's been a great presentation. Really appreciate it. I think we have time for just a couple of questions before we wrap for today. And like I said, if there were any other questions, go ahead and plug those in now, and we'll get to-- we'll follow up after. So one of our questions is, how do you think AM-PPCs can impact changes in network performance? So that's a really great question. And I think one of the ways that it's going to impact is to actually give people that information. Because we really never knew, if you had a procedure done in an ASC, it closes at 5 o'clock. People with complications are never going to go back to that surgery center to show the complications. They're going to go someplace else. So they had no insight or window into what complication-- how many were coming? What they were-- what complications were? What service lines or complications? And it's also maybe a way to showcase different groups or facilities that are doing really well in a network. Great. Another question we have, do you have any analysis where you can show Medicaid benchmarks? Oh, absolutely. I didn't put Medicaid benchmarks on here just because it's easier with large volumes to see that these huge amount of cases, these are the complications. When we do Medicaid benchmarking, we have that in our national data for our benchmark. So it's not all Medicare. But we are able to see that different procedures bubble to the top. So in Medicaid, one of my states that we work with, they actually have ENT and facial as the highest cases, which makes sense for tubes and tonsils and things like that. So we're able to see not only that or circumcisions or other Medicaid related procedures that are being done and what the complications are. Great. And I think we have time for one more. In what ways have states in their payers implemented 3M methodologies to improve quality performance? So a loaded question because it is happening. it's happening across the US. So we have-- I think it's over 30 states that have implemented APR DRGs or EPG. So obviously, the implementation of those methodologies helps with level setting on payment. In addition to that, there are several states that actually use things such as PPRs or PPCs currently to impact payment adjustment. A lot of payers are looking at things such as-- the same things that states are, PPRs, PPCs. Some of those population health metrics. So we've seen a lot of payers use potentially preventable Ed visit submissions, that whole PFP wing of preventables to build ACO models on or MCO models on or any type of large population health program. That was really long winded. No, that was fantastic. I think it was a great answer. So let's go ahead and wrap for today since we only have a couple more minutes. We appreciate everybody who did join us. We'll be posting this on our website in the next couple of weeks. So if are interested in looking or listening to the recording, it will be available. Don't forget to check out the resources where the presentation is there, as well as some other e-guides and other information. And again, we certainly appreciate you completing that survey to let us know how we did. So Jesse and Mickey, we really appreciate your time today. The information has been great. Both of you just really well-- very well spoken and really great presenters that we appreciate. So (DESCRIPTION) Slide, thank you. (SPEECH) again, thank you for our attendees and our speakers, and we will go ahead and close for the day. Thank you so much. Thanks.

April Independent Physician Practice Webinar � Described Video (DESCRIBED VIDEO) Text appears on screen that reads: Fireside chat: How payviders can level the playing field as risk-bearing entities Munish Khaneja, MD MPH, CareAbout Sandeep Wadhwa, MD MBA, 3M HIS April 2023 Lisa, the host, begins speaking and explains the format and tools for the webinar. (LISA) Welcome to our webinar today. We have a great fireside chat with our own Sandeep Wadhwa and we are welcoming Dr. Munish Khaneja from CareAbout as they talk about how payviders can level the playing field as risk-bearing entities. But before we get started, I am going to go through just a couple housekeeping items so you can see the engagement tools and make sure you are familiar with what we have. So we are utilizing the ON24 platform. So it's a great experience because there are those engagement tools, but because this is a web-based platform, we recommend that you use Google Chrome, close out a VPN or multiple tabs because that will help with your bandwidth. Because this is a web-based platform, there is no dial-in number. So if you are experiencing any audio issues, do a quick refresh of your browser and that typically clears up any issues that you might have. Within the engagement tools, you have the media player, you can make that larger as well as the slide area, you can minimize those as well if you'd like to do that. Within the media player, if you do need closed captioning, we do have that available. So using the media player, just look for that button to turn that on if you need that as well. We encourage you to ask questions throughout utilizing the Q&A box. So we'll get to as many questions as we can, but we encourage you to ask for our speakers to get to as many as we can. We do have a resources section. We don't have many slides today, so we're not offering the presentation, but we do have just some more information for you if you'd like to learn more about our solutions, as well as a certificate of attendance for today's webinar. And then lastly, we do have a survey that we always like to hear how we did. So if you can complete the survey at the end, we'd really appreciate it. So let's go ahead and get started. There is a speaker bio section, so if you want to do read a little bit more about our speakers for today, please take a look at that as well. So I am going to go ahead and turn things over to Sandeep (DESCRIBED VIDEO) The image on screen transitions to a photo of a fire burning in a fireplace with wood stacked in a wrought iron bin to the left and fireplace tools to the right. The image remains on screen throughout the entire webinar. (SANDEEP WADHWA) Great. Thank you, Lisa. Thank you attendees. I'm really excited about the conversation we'll be having today. Dr. Khaneja, do you want to maybe kick us off and tell us a little bit about CareAbout and then I'll come back and share a little bit about 3M and we can dive in? (MUNISH KHANEJA) No problem. If we can advance the slides, that'd be great. One more. All right. So CareAbout is a physician-led organization. Basically in its structure, it's an MSO model where we are physician led. Much of the executive team has been an operator as we talk about it in terms of being able to manage practices, manage value-based care relationships. We've all come from the payer or provider part of the world. Our vision is to continually and relentlessly improve healthcare for all, and we believe that the vehicle for that is what we describe in the mission, which is the provider. If we can create structural support for the provider, we can move from fee-for-service to value-based care. The key here is we also look at this in all lines of business, Medicaid, Medicare, commercial. There is an ability to support what the payer and the supporting payment system and payment model requests of the payer and we believe we can help them get there no matter what line of business. Go to the next slide. Three core facets of what we do is the empowerment, enablement and energizing. Provider satisfaction is one of the most important ones and I won't spend a lot of time reading through all of these, but I want to highlight some of the key ones. Making sure that the provider understands what's needed as part of value-based care and at the same time incentivizing the model in such a way incentivizing them the practices, even the support staff that's in their offices is vital to success. If that's not part of everyday work, if value-based care is not part of everyday work, it gets forgotten. And when we look at measures such as HEDIS Measures for pediatrics or for some of the other adult measures, there's date and time requirements where items must be done, so it has to be built-in. Between that you add on the support function, so improving the clinical documentation, coding accuracy, quality scores, all of that are vital to the enablement of that value-based care that we just described, and the hope is to excite them, so energizing the model by adding on more providers. Many of the providers we're meeting our middle of the pack in terms of their timing to retirement. So they're looking for something that says, "How can I work in a universal platform with other doctors and get additional support? I can add more doctors to an existing practice I already have." So we help them all the way through that process. And at the end of the day, I would be remiss if I don't talk about the importance of key performance measures. We have to continually make sure that what we're building to has a clear outcome that we've delineated at the beginning, so that that same provider feels that, "Okay, this is what I've been asked to do and I'll compete it at the end." Next slide. The pillars of growth we look at are the provider actually is an equity partner. So one of the differences we look at in terms of the way we work with practices in creating a series of different singletons in different regions is they have to be aligned to not only an incentive model that's based on a year-end bonus, they should also be the ones who gain at the other end. Again, being private equity backed, there still is a investment thesis and a process. We are working with vulnerable populations, so I believe the mission is it's more widespread and that's been supportive, but at the end of the day, we are seeing a investment thesis, the provider should benefit from that. Overlaying a singleton is an IPA. So part of this model is also to support docs who may not be ready to make that jump into a singleton practice but may want to continue to work with other doctors, I'll call it, in an independent but supported fashion. And so this way all of the technology, the tools and the support we're working on, we are also bringing to bear with the IPA networks that we're building out in each of the regions we work with. Having spent so many years on the payer side, I always talk about payer partnerships. No negotiation has happy people at the other end of it, but negotiation shouldn't be screaming people. It's about making sure that the payers and the providers are aligned at the needs and we can come to a functional middle ground with the endpoint being an improvement in the value-based care aspect of the performance. Otherwise, it's just a conversation about rate. And I think plans have had too much time just trying to build up rates. The market can't bear conversations that just talk about fee-for-service rates on. So these are our four pillars that we work with whenever we talk to anybody. (SANDEEP WADHWA) Great. Munish, let me just do a quick recap for the audience on the 3M story, then let's dive in. But for folks who are maybe not as familiar with 3M Health Information System, we're very much focused on these three pillars around rev cycle inefficiency and rev cycle accuracy, creating time to care, very, very interested in maximizing the time physicians have with patients and looking at ways that we can support the technology approaches that decrease the administrative tasks associated with delivering outstanding patient care. And then our third pillar really centers on the tools and methodologies around driving value-based care. We think there's a role in the market for a private organization like ourselves that can maintain and enhance measurement systems around value-based care. And it's further reflected in some of the particular solutions we offer working directly with about a little over a quarter million clinicians on their efforts between simplifying document capture and all the way through towards population analytics and performance in risk contracts. Thank you, Lisa, for putting up the fireside chat part of the conversation. Munish, there's so many places to start with. It's so clear in your introduction that you're really centered around creating happy physicians and then driving that forward to create a great patient experience. And I guess I just would love to get some early feedback from you on how you're measuring that success with physician satisfaction and engaging physicians. And it sounded like it's both workflow and parts of that structure that have an ownership piece, but could you just elaborate a little bit more on if you feel like how that journey is coming along with delighting the physicians (MUNISH KHANEJA) Delight is a strong word, it's a very fierce. As I said, satisfaction is key. So let's unpackage what we were asking and I think it's important. The physicians have spent a very, very long time living a 70% overhead model. They live, eat and breathe a credentialing problem. A plan changes their network, not at whim, there's usually a thoughtful approach, but because they're so busy, they forgot to open up that last document because it came in paper about what the network changes are going on. They're used to creating a benefit model where they're basically requesting this, that or the other thing for a patient and suddenly that's not a covered benefit anymore. There are just so many interactions that a single doc practice can't work to for anymore and it's not sustainable at the 70% overhead world. COVID, especially I think made folks think about, "What are we spending all of our time in?" We can add on to that that right now between 60% and 70% of their time is spent on an administrative tax and not actually spending any face-to-face time with the patient. So the journey that the docs had come to us with already has been a tough one. And so their ask is, are you going to make my life easier? Yes, because we take on all of those rules, that is the MSO function. Are you going to make sure that I can create a long process where I'm not worrying about every dollar every day? Absolutely. But we have to align the way you pay yourself as a single docker, as a group, the way you pay your employees. We have to align that to the payer's value-based care model. I am not at all saying that the 46 HEDIS Measures are not important and most of our programs actually take into account an entire quality package in terms of a quality incentive program where all of the HEDIS Measures are in. But the general model we look at is the value-based care portion takes a subset of those and really hones in on shared savings, shared risks or full risk programs where you actually connect those to a smaller subset and that's where the plan can say, "These are the ones I'm having trouble with. And you have 500 or 1,000 or 5,000 attributed lives, it would be great if you could help us move the needle on this." So that's the model we bring it to and once we explain it that way, they're generally supportive. And obviously as I said to you, the equity is that super long play, long conversation, not the two to three year conversation, it's the five and 10 year portion of the conversation (SANDEEP WADHWA) Once the physician's part of the organization, are they seeing that 70% come down immediately, or how is that journey from 70% overhead? Is it tangible yet or is that also a long-term event. (MUNISH KHANEJA) Based on fair market value in each of our regions, you can absolutely make more call that the 30%. The 30% can absolutely increase, however, it is based upon incentives connected to improving value-based care. (SANDEEP WADHWA) Okay. (MUNISH KHANEJA) As long as you're doing the right thing by the requirements of our contract support models, which we spend a lot of time fighting for the right thing. Payers are payers, I've been in tho that world a long time and there are ways that the payer thinks, "So we spend a lot of time advocating on the patient and the provider's behalf, but when we come to an agreement and that's signed, we then need to connect that to the incentive structure for the doc." So sure, the 30% will increase as long as they're supported through the incentives. (SANDEEP WADHWA) Munish, it feels like a lot of practices are looking to health systems or payers as potential owners or I shouldn't say owners, but that payers and health systems are also looking to increase their investment or partnership or even ownership of practices, and it seems like you're offering a third path. Could you just share with the audience how you're discussing with their practices, their options and what's resonating, what's not resonating versus what you're seeing with a national payer investing or a health system expanding and doing more of that horizontal integration? (MUNISH KHANEJA) Well, I'll give it to you in the opposite order. I think the national payer investments are usually not in small practices there when they're much more built up and involved. So I've not found many national payers that have taken small regional practices, very small and I'm talking like county, multi-county, they usually go multi-state. So they're not the ones you're going to generally speaking, there are some exceptions here and there in terms of regional plans, like a priority spectrum, kind of a structure. But again, you're still talking about a health system owned health plan. So now I'll talk about the health system owned model. In the health system owned model and I see a lot of health system folks on the list, so pardon the structure, until a health system moves away from the concept of being a cost center, untethered value-based care is hard to, because the tether is still referrals into the system, beds being filled, totally not unimportant. My first training, if you get to go through my bio, I started as a hospitalist, my entire day was in and out of supporting improvements in the hospital. So I value and I completely understand the need and the importance, but value-based care requires a cost reduction. Just to complete that, a commercial admission that lives in the tens of thousands of dollars where it can be avoided is a cost center. So that second option that you offer, the problem with it is it's aligned to another part of the value-based care chain, that's a cost center often. (SANDEEP WADHWA) How do you talk to physicians about that? Because in a way what you're describing maybe that may feed the traditional relationship for a physician is referrals and centralizing. What is the story to the physician in terms of that conversation where there's value in that untethered relationship or less tethered? (MUNISH KHANEJA) So two or three different pieces, we're valuing what you are doing based upon the attribution alone, that's the biggest driver. In a PCP world, it's the attribution. Then it's, do you already understand value-based care? And that value-based care, have you already gotten into equality? Are you already creating a 4, 4.5 stars? And I mean the actual Medicare stars approach. Are you having report cards come in from plans that are showing you're doing that? And on the Medicaid side in a similar fashion, are you in the 75th or 90th percentile of HEDIS Measures? That's just the starting point. Are you involved in shared savings approach? So our value is built upon that versus with health systems it's a little bit more of what types of patients and what referrals can come in. So the question is the equity value is built differently, and then the last thing is equity. Many hospitals are for-profit systems, but many are not for-profit. So there's no equity to be creating. Now, I am not against hospital systems involved in this at all, I think there are many regions where hospital systems and them bringing in PCP practices have been amazingly well-organized and we can cite tens of examples, I don't think hundreds, but tens of examples across the country. It's just that one still requires you to get tethered into that model where you will need to make sure that you're connected into it unless the hospital system lets them live a little bit separately. (SANDEEP WADHWA) I'm super eager to get to the value-based care side of this, but just one other quick question before we jump to that which is, how are you related to hospitals in your model? How important is that relationship? How have you partnered with such a strong physician payer? Could you talk a little bit about your relationships with the non-participating health systems or even broader specialists or other provider groups? How do you view that relationship structurally? (MUNISH KHANEJA) Sure. So I'll start back. We have specialists in CareAbout, in our singleton practices and as well in our IPA, so the connected specialties, some folks call them primary care adjacent specialties, cardiology, GI, neurology, gastroenterology. These are pretty commonly connected practice where the volume of referrals are so important and we want those practices to be involved. Endocrinology is another one in there, but we would love to have those involved in what we do because it builds up the ability, get a complete picture of the patient. Hand surgeons, orthopedics because they're very specialized and it's also can you generate enough volume to support these specialties internally. That's a very important thing. It gives a little bit of a support arm to that specialist. So some specialties we are involved in. The hospital is not the problem that I didn't want to be very clear, we work with them, we have hospitalists that we bring in regions. Our hospital team works with the hospitalists over there because our goal is to get that discharge. As soon as the admission happens, let us help you. Soon as the discharge happens, let us help you. We are very supportive of whatever that is going on in that timeline of the admission, but the before and after, our success with the patient honestly doing better depends upon being very well-connected at the before and after. So that's what we live on. Transition of care work is what we do a lot of. (SANDEEP WADHWA) So those relationships are loosely coupled with the hospitals or formal relationships? (MUNISH KHANEJA) They can be formal. In some cases, we'll sign some formal documentation. In some cases, it's going to be, it hasn't started yet, but when we build up the process, we'll put in our own hospitalists if the hospital supports that, and some of the regional hospitals you can do that. Other places have their own hospitalists and don't let you bring in somebody, but it can be as formal or as loose as the hospital will support it from our point of view. (SANDEEP WADHWA) Well, thank you. I think your business model is so interesting. It's one that I think the audience... I'm glad you were able to share to me there's not one right solution and having these, it's a testimony to there're being choice in the market and different value propositions. And so I guess I want to just switch a little bit to the value-based care portion. The one theme that around physician participation ownership. But value-based care, I guess, Munish, it struck me that you're approaching VBC and you stated it that it's not just a Medicare advantage story, that you're looking at these relationships with the commercial and Medicaid, and I guess I'm really intrigued by your Medicaid VBC because I think all of us on the webinar, every day we get messaged about Medicare Advantage and provider relationships and it sounds like that's important but not the whole story. Can you maybe share with the audience the story with having such a diverse approach to your value-based care and let's unpack that a bit? (MUNISH KHANEJA) Sure. Again, my experience in Medicaid taught me that some of the people who work in the plans feel like, "Well, we're not able to activate any conversations with the patient, we can't do much. Why 35% of all of the mail that we used to send out would be returned, unopened, and basically said you have the wrong address?" Now it's amazing because the address we have is the same one that they gave to Medicaid when they enrolled. So this is the world that we were dealing with. We had 7% to 10% of our population as high risk, and of that 7%, we were only able to activate one seventh of that. We're talking about it was so hard to activate true change. And I think the part that clicked as we started working in the Medicaid population is it's all about the engagement that the doc can create. The plans are not living on major margins, Medicaid managed care plans generally live on razor within margins. But when you take volume conversations, when you're talking about 250,000 lives, 500,000 lives, 1 million lives, which many of these plans have now merged and consolidated into, it's become way more possible to create a functional value-based care model where you are working on 200, 500 Medicare Advantage patients. That same conversation can be had, it's just that you have to have it with 5,000, 10,000 Medicaid patients. Now, same rules. With those Medicare patients, it's actually the risk is distributed more openly. In Medicare and Medicare Advantage, the top 20% are actually the ones where you have to do a lot of work on and it's the other 80% where you may not be getting as involved in care plans and other things. Medicaid, while on 10,000 lives, you're going to be working on maybe 200, 300. I think I used the 7%. So it's 700 people out of 10,000 is what you're working on. And that's more manageable. But the key of the you, it's the payer that's been saying, "Can you do these things for me?" And we're happy to do them as long as we can align value-based care revenue to that. (SANDEEP WADHWA) Well, could you talk about that a little bit in terms of how do you evaluate your ability to take value-based care or risk arrangement with the non-Medicare? My sense is folks are so familiar with the Medicare quality measure set and with HCCs on the risk management side, what are the keys to success for a non-Medicare? And all of this conversation, Munish, my sense is, I just say for the audience's benefit, I think the patient's health is what we're achieving here, but in order for this to be sustainable, I think the risk adjustment and the quality measure set or the success metrics need to be achievable. How are you approaching that from a contracting point of view with the payers? (MUNISH KHANEJA) So you can't take risk on lives you don't understand, so we spend a lot of time asking the plans for data, two, three years of data minimum going backwards so we understand how we did in quality, especially how we didn't claim. We need to understand which hospitals. And yes, when you're asked that question about do we have a relationship with hospitals? Absolutely. And then will we refer patients to hospitals that are lower cost yet the same quality? Absolutely. That's part of our job. So we need to understand all aspects of the care that's being rendered by the claims' data that we're getting. That's part one. And if we can't get to a clean data process, we don't going to have VBC conversations at all, then it sticks with quality. Part two then is, "All right.... Okay, go ahead. (SANDEEP WADHWA) What's your read on the data quality? Are you seeing the payers able to share the data in a way you can digest it or have you had to say no? Just before we move to 0.2, just what's the market readiness between payers and providers around data access for that retrospective look? And then I presume the go forward you'll want data coming in. Where are we on the life cycle there? (MUNISH KHANEJA) It is way better than it was 10 years ago when I was in the plan world fully, like 2012, 2013, it's way better. I believe we were doing a good job giving doctors a report card. We were telling them what's going on, we were telling them which patients to work on, but I don't think we were handing them data that they could then ingest into their systems. So that's the piece. We have created tools that the docs can use. As they're sitting in the EMR, the docs can use these tools to basically say real-time, "Should I work on this? Should I work on this? Should I work on that?" So it is way improved. I think the Cures Act is going to help us in the next two years to really improve it with smart on fire and those API structures, I think we're going to get even better, but it is way improved. So coming back to the second point is if there is shared savings to be had, then the question becomes, at what points do we go shared savings, shared risk, immune full risk? Depending upon what we can, depending upon what we can ensure is what we decide to work with the plan on. We're not finding plans rush to tell us about full risk, so the good thing is plans also don't want to put the providers in a situation where they can't handle the financial pressures of the process. I think most plans have been good that way, but at the same time, they are under their own pressure. So there's a few different constructs you could look at. People talk about an MLR and we've got to get to that MLR. Well, sometimes these medical loss ratios or medical expense ratios are higher than 100. What do you do then? Well, if it is a higher than 100, then there's a world that where if we can work with the plan to drop it from 100 to 90, we get shared savings on the opposite side. So there has to be creativity that plans weren't always applying in the past, so from a contract point of view, we look at every contract with really open eyes, new lens because depending upon where we're starting, we have to create an output that works with them. But Medicaid, again, has the ability to do that. The only last thing I'll say with Medicaid that has a little bit of a planning process is you have to understand you don't have five years to do what you thought you had because Medicare Advantage patients stay on longer. A Medicaid patient in most regions can be in and out of the same plan in one year. So if you're going to create savings, if you're going to create process, it's rapid, pretty immediate on that first reason the patient came to see a doc for that year because two years from now they may not even be on Medicaid. (SANDEEP WADHWA) Munish, can you talk a little bit about your national footprint? Is New York where most of your patients resign, and then how are guys approaching different geographies? (MUNISH KHANEJA) So New York, New Jersey and Florida are where we have started. We are looking at a few other states including California and Texas. It's really coming from folks that understand and are aligned to what we're trying to build. Like I said, delight is what we'd like to get to, but they have to be satisfied with the model. So right now we're heavily involved in New York, New Jersey and Florida. (SANDEEP WADHWA) Munish, New York's Medicaid program is a little different than a lot of other states. They've had their own risk adjustment approach, and Munish, from your prior lives, I think you're kind of a national expert folks on the bio in terms of your familiarity with these different risk adjustment approaches, and it may be helpful to share your perspective on New York's approach to risk adjustment and how you've had to adapt to succeed in that environment. (MUNISH KHANEJA) New York is definitely an interesting place, we always like to do it differently. The rest of the world, the rest of the country calls something MLTSS, we have to name it MLTC. So as a native New Yorker, I can tell you we like to be different. The model that the state uses with 3M, I think, is a fascinating model. First of all, having used the Grouper since my training days and even in Medicaid to us help understand how we waited admissions and understanding that. It's a fascinating model in that it takes into account a whole host of different risk levels. Many risk stratification tools are numbered, are simple measures, this, that and the other thing, and you add these three measures and suddenly you get a number. The beauty of the 3M tool, and that's why I do enjoy working with it overall, is that it has a real ability to take somebody all the way down to healthy all the way up to catastrophic and has a lot of different opportunities. What makes it hard in New York is that what we get from a 3M Grouper as a downstream pay provider versus what the state government does is a bit different. So for those who don't know in New York, the 3M Grouper does organized job in figuring out what the entire Medicaid experience is and then takes a two-year lag to it. So that makes it a little bit tougher for us, but takes a two-year lag to it and then says, "All right, we have X number of dollars, it's a finite number so we have $1 to spend." We have to normalize that across the lowest risk person and the highest risk person. If the plans have the highest risk person, we're going to edge a little bit of the dollar over here. For the plans that have the lowest risk person, they're the ones giving it up. So when we were working in the plan world, the common thought was if we had a risk score below one at the end of the year, we would look for the plans that were a risk for above one and we're like, "Okay, we just paid them," because that's what it felt like all the work we were doing. So it's great that they normalize it because they have a budget structure. The problem of normalization is we don't have a clear ability to increase the pool versus Medicare, not that the pool is a never ending pool, especially recently it's changed, but the pool had a little bit of an ability to grow above where it is. Again, the second problem as I relate is that two year lag, that basically means that today in '23 we're actually getting paid on the experience into one. So it does create a little bit of a financial adjective for most people along that process, but that's pretty much how they use the 3M Grouper for risk stratification and payment. (SANDEEP WADHWA) And are you serving a pretty diverse set of patients in New York? Anything you want to share about the range of complexity or the range of patient populations that CareAbout is engaged in? You mentioned the LTC population. (MUNISH KHANEJA) So we're taking care of all. So the next piece of I think CareAbout's unique model, we don't walk in and say, "Okay, I'll talk to you about your Medicaid contracts. I just want your Medicare contracts, don't care about what you do with commercial." That's not the practice. That physician satisfaction and that support, each one of those situations needs help. And I'll give you a real interesting, so many, many years ago while I was at Emblem, we sat with all of the large employers. So all the health plans sat in a forum with all the large employers, J.P. Morgan, I mean, just name 10 of the biggest employers in that area and they were there, and couple of the consulting groups had people there too because it was national contracts that Aetna, Anthem were managing, we were a regional player. And we talked to them about wonderful things we're doing in value-based cares. Now, this is like six or seven years ago, and so we're talking to the commercial payment model and they said, and two of the very important words, "You're telling us now you can save these additional dollars by doing value-based care, we thought we were paying you for that all the time, we expected that out of you." So the commercial world is a little different. They actually are under the same price pressures except the price pressure is the employer. And it's not that different, we have seen a huge increase in high deductible programs, 5,000, I've seen 10,000 out there, that's becoming way more commonplace. Second, I remember a time where I could go to a employer who had a commercial payment system and I could pick from three commercial plans. It's not common anymore. You got one and you got to pick it. You might get to choose a tighter network or a looser network, but you have one plan. So the world where the employer had more money to spend on healthcare is not there. And that's why, to answer your question, even in commercial plans, there's an ability to create shared savings. I don't think full risk in commercial is easy because of ASO models, because, respectfully, the health plans don't have enough skin in the game other than in their own internal fully insured population. But in the self-insured employee, employer population doesn't work. So we look at commercial, we look at dual eligible, nothing is off the table as long as we can create value-based care around them. (SANDEEP WADHWA) Well, I'm going to press on that, which is you came so close to saying direct contracting with employers, and I guess I want to just press if you think that is viable or you do think this is a three-way conversation. What's the path forward with ASOs? Because I think the ASO model is very much focused on the plan side on administrative efficiency versus in a way, just you said it, I mean the risk bearing is on the fully insured side. So are you engaging in direct contract conversations, or is there a way to engage the payer? Tell the audience about if there's some steps. You're also sending the audience a signal that you're a realist, Munish, that this notion of everything being perfect is not kind of a Disneyland. Are there positive steps on that self-insured path that came out of those conversations or that CareAbout is engaging in? (MUNISH KHANEJA) So it's a great direct to employer conversation. So here's the problem. A payvider taking on all aspects of payer claims payment, IPA management, delegated credentialing, utilization management, member appeals, provider appeals. I mean, for those who've worked in plans, it's a plan, that's a plan infrastructure. And think about it, on $1 billion revenue plan, the government looks at the world and says, "You should be able to do this in $150 million," meaning the 15% margin. How many plans do we know that are skating by on way lower than that? The MLRs are higher, the spend is higher. So the direct to employer, I don't think that's the end game for what we are doing. There are models that are there, it is a three-way conversation. We want to start by becoming a preferred network for the patient. We would love it if they look at our IPAs as we're developing them going forward and say, "Wow, these guys have got it." They've got their own singleton in that universe, they're doing a great job there and they're surrounding it with an IPA. And many of the docs, it's never all, but many of the docs and practices in the IPA are also letting those tools and support functions improve what they're doing and we create incentive. After you become the preferred network for a payer, then you go to stage two. Are we moving to global capitation and are we going to go on and become a full payer functioning piece? Yes, it could come in overtime. It will vary by region, but it is not the end goal we're building to at all currently. (SANDEEP WADHWA) Well, it's helpful to get caught up, I think, for the audience because it seems like a lot of organizations talk about that and it seems to your point that be careful about taking on overselling a position where those other functions are devalued, but if they don't occur, well, Munish, it's a disaster if you get the claims process thing and your network. So I appreciate. Let me turn it over. Let's see if there's some questions, but could maybe just back to Medicare, it sounds like you're taking all lines, but is that Medicare line pay the bills or it's in the same mix as your other lines of work or how do you think about the Medicare direction for CareAbout? (MUNISH KHANEJA) I would've said to you many of the folks who've been done this from 2010 onward when risk adjustment was open and there was very little stringency on the process. At that time, Medicare was paying the bills for anybody taking risk. That's not true anymore. The federal government is just changed the regs on that. One of our providers had said and this is a great way to describe it, "So the federal government is telling us that we actually have to do what they're paying us to, which is manage the patient." And I was like, "Yeah, that's basically it." And coming from a plan side where as a CMO, the utilization management and the care management team was under me and the whole goal of what we're doing is to manage the patient or manage the patient's care. I think that it will become a way more even keeled model where we will create care coordination support, we'll spend still more of our time on Medicare because of the patient population, a little less time than that on Medicaid and a little less than time than on the commercial population. But if we can create the standardized care model that we are looking to create for all these approaches, we can find success in all three. Again, the revenue piece, one might be full risk, another one might be shared savings, but both are achievable. (SANDEEP WADHWA) Munish, a couple questions, just clarification on what is meant by singleton versus IPA? (MUNISH KHANEJA) Yeah, so singleton basically means a practice that's one unified practice and in our world that means they're also on the same EMR, so that when we affect a quality recommendation, when we make a recommendation, it hits all of the docs every time a patient with that disease state or that gap comes in. So that's the singleton, which basically means larger and larger single unified standardization. The IPA would be a series of docs or practices sitting around that singleton, so the singleton is the hub who's really working in one EMR, but other docs might say, "Hey, love what you're doing, just not ready to join a singleton, happy the way I'm doing it right now, but I'd like to see if there's an upside that I can get." Also, I think part of what we're doing with the IPA probably reduces some of their 70% overhead will help delegate credential, then we'll do care management support. So we'll add in some support to them in the process. (SANDEEP WADHWA) One other question about the balance between care management and physician engagement. There's some concern that care management is more hype than value. How do you react to that criticism around population care management? How are you distinguishing between its promise versus the way it's sometimes discussed? (MUNISH KHANEJA) As a health plan guy, it was the bane of my existence and the highest frustration that we would spend money on care management could not get effective change. And I came to my personal realization whether others agree to it or not, I'll let everybody decide. We are too far away from the patient. You get a phone call from your payer, you don't answer, you get a phone call from your PCP, you'll trip over furniture to try to answer it at the time, even today because it might be something important or you're engaged regularly in the portal with your patients. So even telehealth and a commercial and medicare population, you're sending information, "Hey I have this rash, look at it." It's a whole different interaction that the docs are doing. So I believe care management in the payvider space where we are working with the providers who are talking to the patient every day can have tremendous success. It is what the health plans do from a distance that has not been able to get the outcomes. (SANDEEP WADHWA) So, Munish, it sounds like it's not your greatest frustration now then that it- (MUNISH KHANEJA) Not anymore. Now I enjoy the conversation more than ever. (SANDEEP WADHWA) Okay. I mean, I get it, it's not perfect but it sounds like it's gone. And I think that's really valuable for folks to hear, I mean, there's been this talk of if pick the skyscraper in town and floors, 12 to 16 were care managers and now think about that infrastructure being tied to a practice. But I don't know if there's a sense that that's paying off. I mean, it sounds like you're saying yes, that it's worth the investment, it's worth the dollars that you're spending to it. Are those people enjoying their jobs? I mean, this provider satisfaction is that care management team, how are they feeling about their work? (MUNISH KHANEJA) They're feeling good most of the time because most of the time, whether we're a national company, this is not being done from the phone. These are regional people who understand their regions, who grew up in those regions. So that's yet another piece. Many, many of the patients complain that how do you know sitting so far away from me on the phone what I'm going through. I'm not knocking remote care management, but it does add another piece of complexity that the patients don't often understand. They don't quite understand how is this going to help me. So that's why the satisfaction thus far has been great because they're here, they're local and they're doing what they want to do. The second piece is there's a care coordination arm around. So when we talk about care management, that's usually a bit more of a clinical conversation. In the Medicaid and duals population, we spend a lot of time on care coordination to make sure in transition of care those things are done, were you able to get your DME supplies correctly and fast, were you able to get your home care done quickly and fast? So it's the care coordination piece that adds onto it. The home visit programs that I talked to you about getting into the doc patients houses, so much better. It's much better here. (SANDEEP WADHWA) Munish, one question about drugs. What's the role of the Part D equivalent in commercial and Medicaid, and are they part of your risk deals, and if so, how are you managing specialty costs biologics or what's thinking on drug costs? (MUNISH KHANEJA) We're going to tackle that more in the next couple of years because right now we're still laying a lot of the groundwork of what I described thus far with the three E's. But to answer your question, it depends on the plan and depends on the payer. The plan structure might have Part D in it, they might say, "Well, we don't even take Part D. Part D is sitting somewhere else, a union plan. Part D commonly sits with the union, so we don't even have any conversation." So it depends on the plans pieces. Then we go to that second piece that I mentioned, which is the payer's overall model of value-based care. Sometimes Part D is in, sometimes Part D is out. I will caveat folks, if you are going to consider Part D, you have to have a very good understanding of the rebates, the very good understanding on the specialty side how is it being billed in. If it's being billed in purely on the medical pharmacy side, then we come back again to the hospital administered medication versus the office administered. So there's just such nuances that we do understand that we're not getting deeply into. It's variable. I have some risk deals where we have Part D in, some that are out. (SANDEEP WADHWA) Okay. But it sounds like that's earlier on the crawl, walk, run or on the learning curve, it's still to the left side of that adoption. Munish, I think we've covered the waterfront here, so I want to thank you. Great conversation. We covered a lot of ground. Any parting thoughts for our audience before we break? (MUNISH KHANEJA) No, happy to support the model. I'll say the one thing, I wish the state was more transparent, less opaque about the way they use the 3M modeler for the work we're doing. That would help us really create a explanation as to why the patients we're taking care of are sicker. So it'd be great if they did that and the two-year delay. So while we've enjoyed working with 3M, the state could do a lot better. So I'll make the case for that if and ever we can help on that. (SANDEEP WADHWA) And I'll choose not to comment on that, Munish. But I appreciate you sharing your thoughts today with this audience, and really appreciate your leadership and passion and look forward to folks have follow up questions, look forward to folks staying in touch with us and we'll look to make connections here. Lisa, do you want to close us out? (LISA) Sure. Yeah, this was a really interesting conversation and again, thank you both for taking the time today. So again, in that resources section, if you do want to learn more about our services and solutions, you can find more information there. (DESCRIBED VIDEO) A slide appears on screen that says: Interested in learning more? The slide then lists two offerings from 3M available to webinar guests. Real results: A profile of seven organizations and their success with 3M Clinical Risk Groups (CRGs). In this eGuide, read about actual results that payer and provider organizations have achieved when using 3M CRGs. 3M ambulatory solutions. Check out our artificial intelligence (AI)-powered technology that can help you address multiple sources of revenue leakage while making physicians� lives easier. (LISA) And I did want to also call out here in May, we will be in Atlanta for our Client Experience Summit. So if you are a current client and you're interested in learning more, please feel free to go out onto our website to see the great sessions and speakers that we have. Typically, we are in Salt Lake City, but we moved to Atlanta this year. So we have a really great agenda packed full of our clients really talking about our solutions and how we've impacted their organizations. (DESCRIBED VIDEO) The slide switches to an image of the Atlanta skyline with details about the 2023 3M Client Experience Summit. The future is now. Let�s go. May 22-25, 2023. Atlanta, Georgia. (LISA) And as Sandeep said, if you do want to keep in contact with us, you can feel free to email us here. But also within the portal, there is a learn more button that you can also click on to let us know if you'd like to learn more information. And as always, please fill out that survey just to let us know how we did. So again, thank you both for joining today and we really look forward to seeing this group again in the future for future webinar. So thank you all again. Have a great rest of the day. (DESCRIBED VIDEO) A final slide appears that reads: Further questions? Comments? Contact us: ask3Mhis@mmm.com Screen fades to black.

(DESCRIPTION) Presentation. Slide. Text, on 24 for a better webinar experience. Image, a screenshot of the webinar interface which includes a media player panel, a resource panel, a question submission panel, speaker profile panel, and others. Three video feeds on the left side of the screen. Logo, 3M. Copyright 3M 2023, all rights reserved. (SPEECH) Good afternoon, and welcome to our webinar today. Before we get started, I'm going to do a couple of housekeeping items, and then we'll get right over to our panel of speakers. Just (DESCRIPTION) Slide, housekeeping! A bullet point list. (SPEECH) a couple of housekeeping things, we are using a platform called ON24. So if this is your first time joining us, we do recommend using Google Chrome, closing out of VPN, and multiple tabs that will help with your bandwidth. Check your speaker settings as well. If you are having any audio issues, do a quick refresh of your browser, and that should fix anything that you have. There are a lot of engagement tools with this. So you can see the media player. You can make that bigger or smaller. If you want the presentation area to be larger, you can make that bigger as well. Within the media player, if you do need closed captioning, you are able to turn that on as well. We do have a couple of sections for you to take note of. We encourage you to ask questions in the Q&A box. So please feel free to ask as many questions throughout, and we'll get to those at the end. In the bottom left hand corner, you should see resources. So if you do want to download the presentation, that is there. And we also provide a certificate of attendance. That's not for CEUs. But if you do need that certificate for submitting any education, you can download that as well. And we do have some other resources to talk about. We'll get to those at the end. So please feel free to download that information. And then we also have a survey. So please feel free to let us know what you think of today's presentation. We always appreciate feedback. (DESCRIPTION) Slide, intermountain reaps benefits from standardizing CDI technology systemwide, March 9, 2023. Logo, intermountain health. (SPEECH) All right. So let's go ahead and turn it over to our speakers today. We have a panel of awesome speakers from Intermountain Healthcare, who are going to really go over their CDI program, and how they have extended that into the community space. If (DESCRIPTION) Slide, meet our speakers. Three profile pictures. Text, Korey Anderson, MD, CHCQM, FACP. Medical Director for Intermountain physician, advisory services, CDI and quality, intermountain health. Kearstin Jorgensen, M.S., CPC, C.O.C., operations Director for Intermountain physician advisory services, intermountain health. Takiko May, MD, CHCQM, FACP. Hospitalist, intermountain local regional hospital, position advisor educator lead, CDI, physician advisor, appeals and denials, Intermountain healthcare. (SPEECH) you'd like to learn more about our speakers, there is a bio section, so you can learn a little bit more about them there. So I am going to go ahead, and kick things off, and pass it over to Kirsten. (DESCRIPTION) Slide, meet Intermountain healthcare. Image, a tall building with a great quantity of windows. Bullet point lists. (SPEECH) Thank you so much, Lisa. So I'm going to give you just a little background about Intermountain Healthcare. We are a 32-hospital system. That includes one virtual hospital, and most of these are in the Denver or Utah area. We're an integrated delivery network and system. And so we actually have our own payer, and we have greater than 1 million members within our payer group. We provide generally over $200 million in charity care here. We have over 2 million patient engagements a year, close to 60,000 caregivers. And my very, very favorite thing to share about Intermountain Healthcare is that, to my knowledge, we are still the only civilian hoist operation via our air ambulance. (DESCRIPTION) Slide, intermountain office of patient experience. (SPEECH) So I'm going to tell you just a little bit of a background about our CDI program. So a couple of years ago, Intermountain underwent a transition to centralize several of our system functions. And so as part of this that Intermountain physician advisory services, which houses our CDI program, and is all the way over to the right on the screen. We were aligned under the Office of patient experience. And this division includes several functions, including quality, and external reporting. And it's honestly been a really nice alignment to be able to have our physician advisors really work with CDI, and more specifically, quality to figure out how we can impact our publicly reported data, and some of our patient safety via documentation initiatives. I'll tell you just a little bit also about our physician advisor slash CDI program. So we were established in 2010, and we had been previously outsourced at that time, and had some challenges with our vendor. And so we were actually brought in-house. And the initial scope, included utilization review, and over time. We actually included appeals and denials. And then we brought physician advisors, and CDI together. And about that time, we brought 3M on, and that was probably around 2017. (DESCRIPTION) Slide, CDI challenges and improvement goals. Two bullet point lists titled challenges and goals. (SPEECH) And so with our program, give you just a little bit of a quick history about our program. We face some challenges. So we started our program in 2007. And we really were staffed by our finance crew to adjust just a Medicare specific population. In recent years, we've actually expanded to all payers, which was really good, but challenging, to make sure that our team was able to find pieces that would bring the highest value. And so we also faced actually another significant challenge during COVID, which I know we all did. But we were cut several FTEs, and had some of our team members redeployed. So we really saw dramatic impact on our overall footprint in the acute hospital space. Luckily, a year before COVID, we had started some conversations at 3M really about figuring out how we could use technology to be able to expand our CDI footprint. And this really is an important objective to us as we generally have CDS RNs friends that are larger facilities focusing on that work at the expense of some of our community and rural sites. And so really, overall, we've been able to partner with 3M, and use their technology suite to be able to elevate our CDI performance, and have an impact that's really much greater than if we were just doing manual reviews. (DESCRIPTION) Slide, computer assisted physician documentation. A bullet point list. A diagram titled closed loop CAPD. (SPEECH) So I'm going to give you just a little bit of background on the technology that we're going to be talking about today. So we ended up deploying encompass, 3M encompass around 2017. And in 2019, we started engaging with 3M about some of their more advanced technology capabilities in the product suites that they offered. And one of the things that we were really interested in is their CAPD technology. I want to give a little bit of background as to what that is because we're going to be talking about that throughout our slides. And so essentially, what this is there's the ability to have suspect conditions, or specificity logic that's based on biomarkers, or discrete data fields, or lab values that would be indicative of a certain condition. And the logic essentially scans through the record to see if this condition is documented. If the condition, or the specificity is not there when the physician is actually completing their note, whether they're using some dictation system, or typing in the EMR, then an AI nudge will pop-up. And I like to think of it very similar to like an instant message type of functionality. And it pops up, and lets the provider know that they have a potential condition to evaluate for documentation. And one of the things that this technology has done to really gain trust of providers is that it actually looks across the record to see if the condition is documented somewhere else like the H&P, or maybe a different progress note. And if the condition is documented somewhere else, that not just suppressed. So it doesn't go in front of the providers. Another thing that I think is important to call out is that this technology really has the intent of decreasing rework with the provider as they can adjust documentation gaps while they're actually in completing a note. And so it reduces CDI queries, at least, for those specific conditions. This technology, also, I think another important callout is that the technology reduces some of the low hanging fruit of the CDI queries that have been sent in the past, and it allows our nurses to evaluate for more nuanced, and advanced topics as they're doing their reviews. With this CAPD technology, we really tried to ensure that what we put in front of providers is highly accurate. And actually, one of our intensivists calls, some of the nudges head slappers, are really essentially conditions that are pretty basic that the provider maybe neglected to document in the middle of their very busy day. So that's just a quick overview of when we talk about CAPD, or some of our physician nudges. That's what we're referring to in the presentation. (DESCRIPTION) Slide, project strategy. A diagram with an arrow that points to a bull's-eye. The steps along the arrow are labeled as such, CAPD, evidence sheets, priority score. (SPEECH) So as I mentioned before, our overall strategy really has been to leverage this technology as an extender to increase our CDI footprint. And the CAPD technology, nudge technology was really the first tactic in our tool belt for this. We also leveraged what 3M calls evidence sheets. And this is almost like CAPD for CDS RNs. We have the ability to leverage technology that looks for suspect conditions, or diagnosis specificity to highlight for our nurses some specific documentation opportunities. And the nice thing about this tool that I really favor is it's really user-friendly. You end up seeing in the left hand side of your navigation pain the different conditions that are triggered. And then in the middle of your 3M pain, all of the different evidence for that particular condition pulls up. And so it makes it so it's really simple, and easy to review if that particular condition is relevant. Our evidence sheets are basically our second tactic in expanding our footprint. And then our third is the priority score. And so we've really been able to marry the manual work that we do with our CDS RNs with the 3M AI. And we've been able to do this, I feel like in a pretty seamless way. So the unanswered CAPD nudges, and then the evidence sheets are basically prioritized for us in our daily workflow. (DESCRIPTION) Slide, implementation steps. A line with five points that are labeled as follows, leader engagement, Service line socialization, provider profiles, provider education, provider adoption. (SPEECH) And so when we started basically our CDI, or rather CAPD, and evidence journey, we actually did it while we were in the middle of the pandemic in Utah. And so we really needed to make sure that we deployed the workflow in a simple way that didn't add any more time to providers. And we needed to really make sure that we were intentional on each of the different steps, so that we really used everyone's busy clinical time in an efficient, and effective way. And so you can see several steps that we're actually going to go through for you in this presentation, including leader engagement, service line socialization, making sure that we have provider profiles set up, provider education, and adoption. And so I'm going to start out with leader engagement. (DESCRIPTION) Slide, leader engagement. 1, Bullet point list. (SPEECH) And within physician advisors, our mantra is that the medical record is the patients. And we owe it to them that their accurate is accurate. And so we really conveyed this to our leaders as being the y for the strategy around deploying nudges, so that they would be able to help us champion this work. And we did a couple of key things as we engaged our leaders. So the first thing is that we shared improvement data with them. We talked to them about making sure that we have the ability to align strategic directions with some of the other work that we have going on with our board goals, and other facility goals, or care set goals. And then we really tried to get their leadership support as physician leaders, but even different service line division physician leaders as well. (DESCRIPTION) Slide, leader engagement. A graph titled domain percentile ranking. Mortality at 26.3% is ranked 37th top decile and highlighted. Ratio titled improving risk adjusted ratio, two potential avenues. The first number is Avenue number one observed. The second number is Avenue number two expected. (SPEECH) And so what you can see on the screen, we've got some data on the left-hand side, we were actually in the 37 percentile in mortality. And this was actually measured by visit data, so that we could understand where we were at as a national percentile. And if you scroll your eyes over to the right, that just is a depiction of how we get the outcomes data. More specifically, the observed or the numerator is the care delivery to patients, and then the denominator is the expected, or basically the acuity that we are representing how sick those patients are. And so we were able to essentially go over, where we were at in some of our publicly reported data as measured by Vizient. But we also were able to explain that algorithm to our physician leaders, and talk to them about some of the levers that we needed to do to improve. And our physician leaders, and I think physicians, in general, are quite competitive. They want to make sure that they're the best. But even more than that, we really want to be able to represent the care that we give in an appropriate way out to the communities that we serve. And so we really were able to engage physicians once they saw the data and understood a little bit more about how that data is computed. And so from that, I'm actually going to turn it over to Dr. Takeme to go over the socialization of how we were able to basically get this project out, and expand our CDI footprint. Thanks, Kirsten. (DESCRIPTION) Slide, socialization. 2, Bullet point list. (SPEECH) So one of the things that we knew we needed was that good physician engagement, and physician champions to help support this work. In our system, our hospitalists, for the most part, in our larger hospitals are all employed physicians, and they are generally highly engaged with improvement efforts. So for example, one forum where education is in our hospitalist project counsel, which is the leads from all of our hospitals that have embedded hospitalist groups, we were able to share the technology with them, what the goals for that technology were, and then also work on developing, which nudges we would turn on that would be most relevant to those teams. And we were able to do this with surgical services as well. And then because this was a new tool, we also wanted to pilot this, and basically take it for a road test before we rolled it out widely. We did encounter some early bumps in the road, and we were able to provide feedback to 3M, and get the tool improved. We had some issues, where it was hard to get a nudge to resolve unless you had the things spelled out exactly right. But we knew that getting this tool into the best usability format was the best way to get engagement from our frontline clinicians. (DESCRIPTION) Slide. A table with data. (SPEECH) So we also, being Intermountain, and loving data, went back to our data to try to figure out, which nudges we're going to be most useful in our different subspecialties. So here's an example of some data showing how our service lines captured certain conditions compared to top performers in their cohorts. And so this would point us in the direction of which nudges were going to be most useful, to make sure that we were leveraging this tool to the best of its ability. We knew that if we turned on numerous nudges, I think there's over 250 of them, that we would end up with nudge fatigue. And we didn't want to disengage our frontline clinicians before they even got used to the tool. So we were very intentional about how we went after the implementation of this. (DESCRIPTION) Slide, provider profiles. Three, bullet point list. (SPEECH) We also recognized that the provider profiles needed to be a little bit different for different types of providers. So our APPs have slightly different structures in some settings. For example, in our surgical specialties, the APPs may work with more than one service line. But we also know that the nudges that we send to a hospitalist, or a cardiologist look very different than those we might send to a general surgeon. So we used our specialists from our different divisions to help us make sure that we were turning on the nudges that were most relevant to them. (DESCRIPTION) Slide, provider profiles, common CAPD conditions enabled. And assortment of images for each of the diseases. (SPEECH) So this is some examples of some of the nudges that we turned on. So kidney disease looking at getting stage on chronic kidney disease, or identifying acute kidney injury, or cause of acute kidney injury. We recognize that sometimes, people are putting altered mental status, instead of capturing a more risk-adjusting diagnoses. So we turned on encephalopathy trying to get our acuity, and specificity on heart failure diagnoses, et cetera. One of the things that we figured out going forward is that there were things that we didn't recognize. So shock, for example, was one of the nudges that we turned on. We happen to have an ICU in our largest trauma facility that is called the Shock Trauma ICU. And it was triggering the shock nudge every time they wrote a note. And so obviously, that was not going to work out. So we suppressed that nudge for them. We also found, for example, that although we thought capturing the difference between MRSA and MSSA was a big one, we were so naturally compliant that we really didn't need that nudge. So we decided to suppress that one. And then with COPD, we recently figured out that it's the way hospitalists, who are the people who see this nudge the most document. It's assumed that it's stable COPD, unless they say that it's exacerbated. And so it was hard to get the nudge to resolve in the way that clinicians normally document. So we've suppressed that until we can figure out a better way to make sure we're capturing what we want to capture. So it's not as, though, you can just activate this tool, and expect it to be perfect from the get go. You need to have forethought into what you're doing to turn the right nudges on, make sure that your clinicians know what you're trying to get at when you're giving them the nudges, and then listen to their feedback, and respond to that feedback to help with engagement. (DESCRIPTION) Slide, provider profiles. A table with a column titled nudge concept. (SPEECH) So this slide just demonstrates which nudges we initially turned on for which types of providers. So the medicine is our hospitalist groups. And you can see that they have almost all of the nudges that we've chosen to turn on, turned on for them. Whereas the neurosurgeons, we recognize that there are only a few of these that are going to be most high-yield for them. If there are two types of clinicians or more, working with the same patient, the hospitalist consulting on the neurosurgery patient will get different nudges than the neurosurgeon would get on that same patient. (DESCRIPTION) Slide, provider education. Four, bullet point list. (SPEECH) So, again, as we roll this out, we felt that provider education was the key to successful implementation. And when we went live in each facility, we offered a brief education three times a day on site focusing on those providers that we thought were going to see them the most, including our hospitalists and our general surgeons, for example. We had a fact sheet that we shared with them, so that they would understand what the tool was, what the components of it were. And then we had boots on the ground doing site rounding to help troubleshoot, including IT presence to help if there were concerns, or if the tool wasn't launching as we expected. They were also there to help answer questions that came up as the clinicians were getting the initial nudges. (DESCRIPTION) Slide, provider education. Three screen captures of an app. (SPEECH) We continue to use this same type of education going forward because we're adding new providers all the time. Some folks might not have gotten initial education. Some folks may forget the education that they've had. You can't expect us to educate once, and have everybody buy in. So we know that ongoing education is necessary. We're also good about sharing data, about the use of the tool with teams, so that they know what's actually happening in real-time, and then trying to reengage those clinicians who may not have previously engaged with the tool. One of the nice things early on when we were starting our education, before we went live with 3M, we would get requests from frontline clinicians saying, why can't you give me my query in the flow of my natural work? And this tool is able to do that, give them their nudges, and their queries in the flow of their workday, which we think is a huge benefit as Kirsten alluded to previously. (DESCRIPTION) Slide, provider adoption. Five, bullet point list. (SPEECH) So in order to continually engage in provider adoption, we've tried to make sure that this project remained visible by sharing it in group meetings, both with the hospitalists, and our surgical providers, as well as our intensivists. We, again, share that data transparently. We share query response data with leaders in each division, so that they can share that with their teammates, and make sure that everybody knows that this is something that we are expecting them to participate in. We hope for a 90% or better query response rate. And for most of our specialty specialties, we are there. But it is important to track the progress. And if there's a problem, then you need to go back and try to figure out why your clinicians aren't engaging, is it because they aren't seeing the tool that they've somehow managed to minimize it, or are they intentionally minimizing it, and then go back, and reeducate. (DESCRIPTION) Slide, graph. A line graph of engage one active providers by month from January to December. The line increases. (SPEECH) This is the overview of our adoption. And it shows us going live across the system in multiple facilities. So you can see that we started back in 2021 in January with one of our larger community hospitals. When we first went live, we were using document based reasoning. So it was only looking at the current node, and then we were able to implement the encounter-based reasoning that looks at the entire chart. We then rolled it out successively to different facilities, again, with that boots on the ground approach to help support the rollout, all the while, getting feedback from frontline clinicians to improve the tool, and sharing that feedback with 3M, so that we could get the optimized version of the tool. And you can just see that it's taken us quite a while. And even at the end of this slide, this was not all of our hospitals, as Kirsten mentioned. We have 32 hospitals. And some of our rural hospitals were the last to go live, and some of them we have just brought on recently. There are a few challenges, I think, with the rural facilities, in particular. There's a documentation culture that comes when you have a number of physicians that are doing the same work, and following up on another provider's patients. That isn't the same when you have maybe a family medicine doc in a rural hospital, who doesn't do this very often. And so I think that's been one of the challenges in the rural facilities. But we are getting them to use this tool, and doing some more education where appropriate. (DESCRIPTION) Slide, provider adoption, facilities without CDS RNs. Image, the entrance to the Intermountain Sevier Valley Hospital. (SPEECH) So in terms of provider adoption, this severe Valley is an example of a rural facility and we don't have any CDI RN presence there, or specifically, supporting that hospital. And this is an example where that you'd see in several of our smaller hospitals. But we share the data about where our documentation opportunities are with the administrative team, administrative leadership, as well as the chief medical officer, and trying to get them engaged to help engage their community-based physicians. And we really think that this sharing of data is key to the success of the rollout. We also recognize that sometimes the things need to be rolled out a slightly different way in a smaller facility. But as we get more and more focused on clinical data, and how our documentation impacts our outcomes data, we are getting greater and greater involvement from our rural facilities. And they are now asking us to come back, and do education with their providers to help implement the tools. (DESCRIPTION) Slide, evidence sheets, nudges for CDS RNs. A ring segmented in three parts labeled AI use for clinical indicators, evidence of conditions in the record, suspect condition prioritized for CDI review. (SPEECH) And now, I am going to turn this over to Corey. Perfect. Thanks, Taki. A couple of things I would add on the provider side as well is it's an ongoing labor of love I think with the nudges. And I think one thing we did learn early on is it's very nice to have a test group of physicians that you can leverage to turn on new nudges, and trial them as Taki talked about a little bit to see where the kinks are, where the issues may be, is it doing what you want. It's really challenging to roll out something new to a huge swath of providers, and have it not be successful. Because while it takes a long time to get by in and engagement, it does not take a very long time to have something be unsuccessful, and completely lose engagement. And so we've used our physician advisor group as that test environment, if you will, to turn on these new nudges, and see how they're working, take their feedback knowing that they're engaged in the work, they really understand the value, and the why. They're not going to be too dissuaded when things aren't successful, if it's not functioning well. And we can bring that back work with 3M to say, here's what's working, here's what's not can. We change the criteria? Can we change the wording? What can we do? Somewhat in line with Taki already mentioned of how we approach this. The other thing I would say is, the other piece we probably still have not optimized as well as we should, and that we've learned is, it's not turn it on, and it works, and you forget about it. I think it's continuing to review your data as a group, and seeing, is it getting what you need, and what is your data showing in terms of our new opportunities arising? Are there new nudges we should be bringing into the fold? Are there ones that are either not doing what we need them to do, or they've done all they can do? And it's really not adding much now that we may be concerned sunset. Having some process, or come together on a recurring basis, where you're looking at the data, and comparing that with the nudges, and what they are, or are not doing, and where do we need to go from here in terms of bringing in new nudges to add to the value of the work, and what the technology can do for us. So with that, I'll talk a little bit about the CDS or the CDI nurse side of things. And this is really what we were the evidence sheets come in. And evidence sheets is really just a fancier way of saying nudges for our CDS RNs. It's effectively doing the same thing that it does on the physician side. But it's for our CDI reviewers to, again, help them be able to prioritize where the money is right in terms of looking at the cases. And I say money, but it's not always cases that have dollars, but it's cases that have quality aspects or safety aspects to them as well. And so as Kirsten mentioned, our program is staffed in a way where we only probably have about 55% to 60% coverage with our CDS nursing doing reviews. And so as we're looking at whether we're talking about CCs, and MCCs at Intermountain we're about 60% fee for service and 40% value-based. HCCs are very important in our organization. How do we bring those reviews into the mix? How do we still look at our other quality and safety data around mortality, patient safety indicators, et cetera? There's a lot of things, and a lot of plates we have spinning in the air for our CDI nurses. How do we help them juggle and manage all of those things and go and do reviews where there's actually opportunity identified, and they're not losing or wasting time, if you will, looking at cases that honestly don't have any opportunities available? And this is where evidence sheets are really played a huge role for us. And so using the CAPD, or the AI technology, and we've customized a lot of these to reflect criteria that we're using at Intermountain, right? So for example, sepsis. It's not that we're making up our own criteria, but sepsis is one of those conditions, where some adhere to the sepsis three criteria. Some are using the step one step two. We just are able to customize it for the criteria that we're using as a standard within our organization. And so we use those clinical indicators, whether it's lab values, certain vital signs, imaging, findings, et cetera. And they can be built to have that CAPD technology looking at those reports, and that data to then nudge our CDS nurses. If something is identified, and it it escalates up to-- and we've turned on evidence sheets that we know are CCs, or MCCs, or they're HCCs, or they're variables in our quality and safety data like mortality. And so where there's evidence that these things in the record, it will then nudge our CDS nurses, and it nudges based on where we've scored this in the priority score for our CDI review. And we're able to give whatever sort of value we want to these evidence sheets higher or lower, so that as certain evidence sheets are identified based on the criteria, it moves up in the priority score for our CDS nurse to take a look at sooner than maybe something else that scores lower on the priority score. So this is really how we've tried to incorporate this evidence sheet, knowing that we don't have humans to look at every case. And we need to really be efficient in the cases that we are reviewing to better prioritize the work. Next slide. (DESCRIPTION) Slide, evidence sheets. Nudges for CDS RNs. A screenshot of an interface. (SPEECH) And so along with that, again, this is just an example of what you can see, or what one of the CDS RNs would see is that there's a potential condition, or a couple of conditions that are identified. And in what category they fall into based on the criteria that we've built. When it was identified, so is this something that was found a couple of days ago that nobody's paid attention to, or is this something that was just found this morning? Also, the provider, it can tell you that whether the provider was nudged. And so we can see sometimes some of these evidence sheets are also similar conditions that the physicians have as nudges. Obviously, sometimes the physicians get to those nudges, sometimes they do not. And so knowing whether we nudge the provider, and the CDS nurse can go and take a look, and see maybe they just disagreed with it, maybe it didn't meet the criteria. What was the reason maybe the provider was not responding to that nudge? And then after further review, does it still warrant pursuing? Does it warrant the query perhaps to the provider, or is it something that we should dismiss, and move on? And you can also see, again, it's using, or tethering the evidence to the conditions, so that the CDS nurse knows what are the potential lab indicators, or other findings, what are the potential treatments that are being provided that would be supportive of the condition to, again, better help them be able to know, is there something really here to further look into, and potentially a query opportunity, or is there something that's not really that substantiated that I can quickly move on, and get to the next potentially prioritized case. (DESCRIPTION) Slide, evidence sheets. A table labeled active evidence sheets filled with different diseases. (SPEECH) As far as the evidence sheets, and what we're choosing to use, Kirsten mentioned, we've tried to make this more nuanced. And it goes back to our approach with the providers, where it's those head slapper conditions, right? Heart failure, I think is, or the stage of chronic kidney disease is the classic head slapper. Really most physicians working in the hospital should be able to document the acuity and specificity of heart failure. It's such a common prevalent condition. It really should be something that we're not missing. The same thing with the stage of chronic kidney disease. It's so common and prevalent. Those should just be the things that the physicians get right all the time. However, in a busy day with complicated patients, and a lot of things going on, those conditions with specificity are still missed not infrequently. In fact, I think up until recently, and it may still be the same, heart failure continues to be the top. One of the top, if not the top condition that is still queried for by nurses, which makes no sense because it's one of the most common conditions that we see. And whenever we do education, it's one of the examples that we use. Yet the physicians still struggle at times. That's what the nudge can do for the physician. Help just capture that specificity, and acuity that you should be getting every time, and probably do 80%. But that 20% that, you don't. The nudge fills the gap so that we're not missing that. We can then use our evidence sheets for our CDS nurses to go for some of the conditions that are not as straightforward for the physicians, and are actually a little bit challenging to write a good nudge, that the physician can clearly understand, what is being looked at, or what is being asked potentially here. And so you can see some of those conditions include things like sepsis. And as I mentioned, going back to what is the criteria you're using, is it step 3, is it step 2, step 1? CERs, how many CERs do you want present, or antibiotics ordered before you would create a nudge? Those are a lot more nuanced. And we really want going to our CDS nurses to basically just say, we think something is here. Go look at this case with your clinical thinking cap on, and see if there's something here, and whether a query is appropriate, rather than having that face the physician, where it's interrupting them in a busy day. It's not clear. Maybe it's a pancreatitis patient that is a little febrile and tachycardic, and is clearly not septic. But it's going to ping the physician over and over again. We just didn't want to do that. And so we took some of these conditions, as you see, and made them more CDs facing than physician facing. And I think this has really helped our CDS team prioritize some of these cases where, again, sepsis, malnutrition, shock. These are big ones we don't want to miss, both from a clinical perspective, and the communication that the documentation provides provider to provider, but also from a reimbursement perspective, and a quality and a safety perspective. These are big conditions that we really want to make sure that if present, we are capturing the vast majority of the time, and we are not missing them. (DESCRIPTION) Slide, nudge next steps. A circle with the five layers labeled labs, diagnostics, progress notes, H & P, ED. The circle points to the text, end of life nudge. (SPEECH) So next steps, as we talked about a little bit, this AI technology, or the CAPD technology, it's encounter-based. When we first started, it was note-based. So it was only looking at the note that the physician was writing at that time. After we got our feet under us a little bit, and a little bit of experience, we then expanded it to encounter base. So it's looking at all the documentation, lab, imaging, physician, APP, even some of your ancillary services like dieticians, et cetera. And it's able to look for the different phrases, clinical indicators, et cetera, that would either reflect the condition has been appropriately documented, and that it will get coded appropriately with the relevant specificity, and acuity, or that it continues to be unspecified, and therefore, warrants either a nudge to the physician, or if it's one of those conditions would prompt the CDS RN. One of the interesting things that we've done with this is we've started to find other more creative ways to use the technology. Not just as far as it pertains to clinical documentation in the sense of capturing an ICD-10 code, but as we started to think about our work around mortality, one of the issues that we found that we struggled with was that we were having a lot of patients that we were waiting till almost the last few hours, if not minutes, in the end of life before we got palliative care, and/or hospice services involved. And really I think as most folks in the clinical sphere, while that is still probably useful, it's not optimal, right? The sooner we can get those services involved, there's clear data that shows those patients have a better quality of life for whatever time that is. Families are better supported. There are more resources available. It's just overall a better experience at the end of life for these patients when they're identified early, and we get the right resources provided to them. And so we created some nudges, if you will, around end of life that not go to our team, or not go to our physicians. But actually, goes to our care management team through a 3M workflow. And we created the phrases that we wanted the technology to look for in the documentation, things like end of life, poor prognosis, terminal, obviously hospice, or palliative care. Whatever phrases you want. And now, the technology can look at that, and it can ping our care management team. Each day in the morning, they have a work list, where a number of patients have been identified through the technology and the workflow. And now, as we have a daily huddle each day in each facility with care management, and the palliative care team, they have a list already created that they can run, and make sure, hey, are we aware of the right patients? Are we connecting with the right physician and team to make sure that the discussions that are being had we're involved if needed? We're there to support them, and we can get the resources to the patient, and family in a timely manner. It doesn't mean, we go in and say, hey, your patient was identified as an end of life. We're here to put them on hospice. It's simply to make all of our teams more aware of what's going on with our patients in the hospital. And so that we can get the right resources to the right patients, and the right time, and support the front line teams with those resources in the best manner possible. (DESCRIPTION) Slide, outcomes. Image, a doctor leans in close and smiles to a young girl. Slide, impact by nudge type, eight months of data. A table whose main column is labeled nudge concept. (SPEECH) So as you can imagine, there's some outcomes that you'll want to be aware of, and that you want to focus on around. What is the technology doing for us? And so you can see here, we have a few different buckets as far as whether it's financial impact, just overall final codes on the claim, and then quality impact from a few of the nudges. And this was a little bit early on in our implementation. And there's about eight months of data. But you can see, overall, as far as total resolve nudges, over 6,500 nudges with just I think about 15 or 18 conditions that are listed. There are over 5,000 specified codes included on the claims related to the nudges are tied directly to the nudge impact. You can see it had a significant impact in terms of MS-DRG, or APR-DRG, or both, as well as changing the primary diagnosis, or adding a CC or MCC, which, again, is most of us in a fee for service model that is highly impactful, and important to being reimbursed appropriately for the care of the patients we provide. You can see also there was an HCC impact, which as I mentioned in a value based world is very important. And then even some Alex Houser impact as well. (DESCRIPTION) Slide, nudge resolution, eight months of data. Another table with a column labeled nudge concept. (SPEECH) The other thing that you can see that we found that was interesting on the data is that with just that eight months of data, some key findings we found is that the impact is early on as well, right? And that's I think where a patient population from CDI perspective, we've struggled is to get some of those early reviews. Patients that are there for a couple of days, and through efficiencies, and good clinical care are now able to be discharged sooner. It's really hard to get our CDI team there to review those cases. Potentially, generate a query, get it responded to. And the nudges have been able to fill that gap where we're getting. As you can see, over 20% of all the encounters with a resolved nudge were for patients with a length of stay that were less than two days. And almost 75% of all the resolved nudges had a nudge date that was within two days of the admit date. So really this is helping us get some of these documentation deficiencies addressed in those short stays and early on, which also is helpful because we know as we carry consistency through the medical record in our documentation. The better it is for us from an audit perspective, or a denial perspective where if sepsis is only captured on the discharge summary, and a 10-day stay, whether it's true or not, just doesn't look as good as sepsis being captured 9 or all 10 of the days of a 10day stay. And what this actually translates into is almost a 4 and 1/2 FTE for a CDI, or CDS RN an impact. Meaning, we're using technology, where humans have had to fill the gap, which can be quite expensive. And this isn't to say that this replaces the CDS RN, but it's allowing us to extend our bandwidth without always having to go to the big expense of hiring additional humans to do the work. (DESCRIPTION) Slide, nudge response summary, small facilities. A bar graph with the nudge concepts on the horizontal axis and 0 to 100% on the vertical axis. Each bar has three colors which are labeled Active percent, naturally compliant percent, resolved percent. (SPEECH) So you can see here, when we break it down into our small facilities, you can see where the gray is nudges that were resolved, and the percentile, or the percentage that were resolved. You can see the naturally compliant bucket, which means after a physician has been nudged three or four times for the stage of chronic kidney disease, they generally start to learn. If I'm putting chronic kidney disease in the chart, I should put a stage with it, or I'm going to get nudged. And so that's the naturally compliant piece. It's almost a learning tool, if you will, that the providers of start to document a lot of these things naturally on their own. Whereas as Taki mentioned, that staff infection, really for the most part, we're capturing it without having to do any nudge. So that may not really be adding a whole lot of value to us, where with some of these others like shock, and COPD, we maybe have some work to do either with the nudge, or how it's worded, or configured. So this is how it breaks down in the small facilities. And then when you look at a system level, I mean, very similar. We see that the small facilities are oftentimes a microcosm, or reflection of what's going on at the system level as well. But again, a lot of them, good natural compliance. Some of them were not getting a lot of engagement on. And so that's some of the data that I mentioned in terms of going back after a few months, and looking to see what's working, what's not working, is this still an opportunity, is the nudge doing what we needed to do or not. (DESCRIPTION) Slide, APR risk of mortality. A line graph with years and quarters in the horizontal axis and units from 1.5 to 1.9 in the vertical axis. The line increases over time. (SPEECH) As you look at some of the data like APR risk of mortality, you can see over the last couple of years, as we've implemented this technology, which really it didn't start until about the middle, late spring of 2020. You can see where our APR risk of mortality has gone. It is consistently trended up. Now, again, this is partnered with the better access, and ease of answering queries, as Taki you mentioned through this similar electronic query delivery method. This has been coupled with a lot of provider education. And I think that's a part of this whole thing is not the standalone, fix all, be all, and all sort of thing. Technology, I think needs to be supported with robust education to the providers around the why of this work. Most physicians don't understand the finances of hospital. They don't understand what a query is going to do, how that helps pay for more staffing, right? How that helps pay for the new robot in the operating room. They don't understand how it impacts their quality data, and whether their mortality rate looks good, or it looks bad. So I think really having a foundation of education for your providers is key. And then using the technology to support that work, and help them be able to do the work effectively, and efficiently, and easily. So it sets them up to be successful in the way that they approach the work. (DESCRIPTION) Slide, next steps. A bullet point list. Image, figures in a circle with their arms around each other hunched forward in a huddle. (SPEECH) Next steps really as I talked about is, at least at Intermountain, how can we further expand the use of these nudges to different specialties? As you saw, we have about eight or nine. And we can probably expand out from that even further, and really get into the pediatric setting. We have a full children's Hospital within Intermountain, and we have a second one that will be opening early 2024. So this continues to be a growing area for us to be involved in from a CDI perspective. The other is, again, as I talked about a lot of work around mortality and identifying these patients who are potentially moribund, and would benefit from end of life services much earlier in their clinical course than what we've historically done, and using that technology to better support us in the work around that. (DESCRIPTION) Slide, lessons learned. Bullet point list. Image, a group of practitioners with facemasks pose for a picture outside the hospital. (SPEECH) Finally, just a few lessons learned from what we've talked about today, leadership support is critical. I think as a CDI team, we've learned we're influencers. And in order to best influence, we need the support of our clinical physician leaders, and system physician leaders to help one get fannies in the seats, if you will, to hear education, and learn about this, but then also, to realize, and recognize this isn't just important to us in CDI, this is important to all the leaders, it's important to the finance folks, it's important to the quality folks, and why. The other thing is, as we talked about, is you need technology coupled with the CDI review process, and you need to tether that to data, and use your data to help guide you where you're being successful, but also identify where you still have opportunities, and gaps. And that's how you can close that loop, and really create a successful program with using all of these different tools together. (DESCRIPTION) Slide, questions. Image, a woman raises her hand in a meeting. Text, Kory Anderson, MD, CHCQM, FACP. Medical Director for Intermountain physician, advisory services, CDI and quality, intermountain health. Kory. Anderson@imail.org. Kearstin Jorgenson, M.S., CPC, C.O.C., operations Director for Intermountain physician advisory services, intermountain health Kearstin. Jorgenson@imail.org. Takiko May, MD, CHCQM, FACP. Hospitalist, intermountain local regional hospital, position advisor educator lead, CDI, physician advisor, appeals and denials, Intermountain healthcare. Takiko. May@imail.org (SPEECH) And I think with that, I will end. And I think, Lisa, I will turn it back to you. And if there's any questions, or further comments. Yeah. That was awesome. Thank you. Thank you all. A lot of great information. And we do have some questions that came in. Going back to the nudges, do you line your clinical parameters for your nudges with any denials you're receiving? Or are you making any internal processes, or policy revisions based on those? So I would say, yes. But we're doing it even before some of the denials. I think as part of our CDI program, we've tried to have a core work group, including compliance in that forum to have some clinical validation documents within our Intermountain organization. Meaning sepsis. This is the definition we're using here are the clinical indicators, respiratory failure, same thing. We've had much like around the country, we get denials for drugs around sepsis and things like that. And we've been able to use the CDS review process with the prioritization of sepsis, to get appropriate queries, where some of these cases were questionable to our physicians, where they could clearly say yes, it's excluded, or confirm that the diagnosis is present, and reaffirm why it is present. We have been able to leverage some of those instances. With one specific payer, we were having a lot of sepsis issues with to get some of those overturned at the administrative law judge level. And really in the long run, change some of our contract language with that payer, so that we don't continue to get denied over and over for that same condition. So I maybe expounded more on that question than you would have liked. But to answer yes, we have. But we've coupled it, again, with some of the other in-house things we were already doing as well. Now, that's fantastic. And I think expanding on that, that definitely helps. Do you have any HIM professionals working in your CDI group, or is it predominantly RNs? It's predominantly RNs. We do have a small concurrent coding division within our group. In addition to that, whenever we're developing nudges in whether it's strategy, or involving our HIM professionals, we actually try to ensure that coding and operations from HIM are included, when we're setting up nudges, when we're setting strategy, or when we're reviewing data. And that's just really important because they have the ability to also provide insight into what they're seeing from a deficiency perspective, which really helps us decrease our discharge not final coded or not final builds. OK. Going back to the provider adoption, enabled provider adoption, what percentage of your providers use CAPD? I think we've actually set up almost all of them that have admission privileges. If not, most. So that's thousands of providers that we've actually set up on this technology. More recently, as we've been speaking to, we've really tried to focus on the smaller sites. And so I think we've got most of them set up. Our smaller sites are engagement in some of the responses to those nudges. It can vary, but we're seeing it increase, which is fantastic. And Lisa, I would say one thing, too, that we also had to learn early on is you still have a health care system, where you have a lot of docs who straddle both sides of the fence, right? Where they're in the office, at some point, in the day, or in the week, and then they're in the hospital at other times. I think early on, we had some instances, where the nudges were actually firing to the physicians in their clinics as well. And it was actually a little bit overwhelming. And we were able to work with 3M to reconfigure how that was set up, so that these are only firing for physicians when they're in the hospital setting, at this point. When they go to their clinic, if they do both sides of that fence, it's not going to be nudging them in the clinic. Now, some of the stuff it can do that. And clearly, it can. And at some point, maybe that's an area we move into. But we have not done that yet. We've configured it, so that only fires when they're in the hospital setting on those patients. And staying with the nudges, do you include case management issues in those? Well, I mean, with Dr. Anderson actually spoke to some of the work that we're doing in identifying patients who are actively dying. And what we have done specifically with those nudges is we've created a custom work list for care management to log into 3M, and they can identify moribund patients, and take them to a daily huddle. There's been a couple other topics that we've included. Some of the BP CIA, if anyone's familiar with that. The bundle program topics have also been included. And that really helps our care managers understand who's in the cohort, as well as some other details about the population. So in sum, we have log-ons for care managers. We have work lists for them within 3M, where they can go in, and actually work cases. And we have topic specific nudges that we have created. And before we wrap, we did have a couple questions around integrations with EMRs. And so just real quick, and then I'll pass it over to Kirsten to talk about the EMR that you're integrated with. So if you are interested in learning more about the integrations that we have, definitely let us know. There is a button in the middle of the dashboard that you can fill out some information, and we can definitely talk to you about your workflows, and understanding if it would be a good fit. Because we do integrate with, gosh, at this point, I think we say around 200 plus EMR. So it's definitely something that we would like to see if it's beneficial for your organization. Somebody did specifically ask about Altera. And CDI facing features are available today for those customers. And we're currently working on an integration for real-time nudges for those sites. So if you'd like to learn it more information, definitely reach out to us, and we'll have somebody from our team reach out to you to discuss. But specifically, for Intermountain, what EHR are you currently using with that integration? Yeah. We're actually using Cerner. I think generally speaking, it's gone fairly well. And I'll tell you just really quickly from an integration perspective, 3M was so helpful to us, brought project management, resources, and really managed those very technical aspects of making all the connections work within 3M. And specifically, the engagement product to Cerner. We do have some additional products that sometimes fight with 3M. And so we're internally trying to deal with that. I don't think it's Cerner. As we've talked to our IT professionals, they've identified different products that are fighting. And so we need to figure out how to deal with that we've got a lot of bull tons onto Cerner. That being said, overall, I think it's actually worked pretty well from an EMR perspective. Fantastic. Well, thank you all so much. I can't say enough just how great it is when this group is here. We've had a couple of times, I know that you speak at a lot of industry associations. So I encourage anybody that goes to tradeshows, to always look up this team that speaks often in the industry. And we're so honored to have you here today to talk about your successes. So (DESCRIPTION) Slide, want to learn more about how we engage with our positions to transform CDI efficiency? Click to learn more. Logo, intermountain health. (SPEECH) if you are interested in learning more, within that resources section, the presentation is there, and there's links within that resources section. So if you want to learn more about how Intermountain has been engaging with their physicians, to really transform their CDI, please take a look. There's a lot of resources there, some case studies. We also have a couple of videos as well. So take a look at that. And please let us know if you're interested in learning more. If (DESCRIPTION) Slide, 2023 3M client experience summit, the future is now. Let's go. May 22nd to 25th 2023, Atlanta Georgia. 3M Client Experience Summit for 2023, we are excited to be at a new location and venue at the iconic Westin Peachtree Plaza Hotel in the heart of downtown Atlanta, Georgia. Mark your calendars for May 22 - 25, 2023. Each year, 3M brings together our most valued clients and colleagues to discuss evolving trends in the health care industry, learn about new opportunities in care and technology, share best practices and celebrate successes in innovation. And each year, we keep expanding and adding new and exciting opportunities! (SPEECH) you are a client of ours, we definitely encourage you to join us in Atlanta this year. We are going to be hosting our 3M client experience summit, May 22 through the 25. And really, it's a great opportunity for you to come together with other 3M clients, or we talk about different trends. It's three days of sessions. But predominantly, we have our clients there discussing their success stories, and innovation as well. So we definitely encourage you to join us in Atlanta. And we do offer CEUs. So if you do need to get some of those, definitely consider joining us in May for our client experience summit. So with that, again, really appreciate this tremendous panel from Intermountain. Please complete that survey. We'd love to hear how we did. And (DESCRIPTION) Slide, thank you for joining! (SPEECH) I know we're working on our next community webinar. So please be on the lookout for that. And we hope you join us then. So thank you all again, and we'll see you the next time. Thank you. Thanks so much.

(DESCRIPTION) Text, A great company is showing what interesting applications a fantastic product can bring for motivated users. Arrows point to parts of a computer screen. At the left, an arrow, media player, points to a Live Stream window. Arrow, resources, points to a Resources window. Arrow, Have a question? Let us know here?, points to a Q and A window. At the right-hand side, arrow, Meet our speaker!, points to a Speaker Bio window. Arrow, Survey -- we want to hear from you!, points to Survey window. Text, 3M M*Modal. 3M M*Modal Fluency Direct for mac O S. Timothy Ruff. Chief Product Owner. (SPEECH) Good afternoon and welcome to our webinar today where our Chief Product Owner, Tim Ruff, is going to discuss how Fluency Direct is integrating with macOS. (DESCRIPTION) Disclaimer. Slide text, The information, guidance, and other statements provided by 3M are based upon experience and information 3M believes to be reliable, but the accuracy, completeness, and representative nature of such information is not guaranteed- Such information is intended for people with knowledge and skills sufficient to assess and apply their own informed judgment to the information and is not a substitute for the user's own analysis. The participant and/or participant's organization are solely responsible for any compliance and reimbursement decisions, including those that may arise in whole or in part from participant's use of or reliance upon information contained in this presentation. 3M disclaims all responsibility for any use made of such information. No license under any 3M or third-party intellectual property rights is granted or implied with this information. 3M and its authorized third parties will use your personal information according to 3M's privacy policy. This meeting may be recorded. If you do not consent to being recorded, please exit the meeting when the recording begins. (SPEECH) Before we get started, I want to go over a couple of housekeeping items just so you can become familiar with the interface. So we do use ON24 as our platform. This is a web based platform, so we recommend using Google Chrome. Closing out a VPN or multiple tabs, that will just help with bandwidth. If you are having any audio issues, excuse me, check your audio and speaker settings and do a quick refresh. We do not have a dial in number or a call in number because this is a web based platform. So again, a lot of those-- if there is a glitch, do a quick refresh and that should help with that. We have engagement tools for you as well. We have our media player where you'll be able to see our speaker, but also we'll be playing a short demo for you as well that will actually be playing the sound. So if you do need closed captioning, there is that availability in that media player. We have the Q&A section. So we encourage questions throughout. So we'll get to as many as we can. So please feel free to ask questions. And then we also have a survey at the end. So please let us know how we did within that survey. And we also have a resources section that has the presentation for today. So without further ado, I'm going to go ahead and pass this over to Tim to go over how Fluency Direct is interacting with the macOS system. Tim? (DESCRIPTION) Slide title, 3M M*Modal. Slide text, 2002, First M*Modal products introduced to the market. Focus on closed loop, concurrent clinical documentation and decision support based on a unique set of technologies. workflows and processes that help to transform words into action. (SPEECH) Thank you. So I think what we're going to do at first is we're going to dive into a little bit of background information. Those of you that are aware, so 3M or M*Modal has been in business for over 20 years now. And so for the last 20 years, we have concentrated on providing products and services to reduce the amount of time that physicians spend documenting and to allow them to have more meaningful interactions with their patients. And so one of the taglines we routinely refer to is creating time to care. But over the last 20 years, that scope has grown to service over 300,000 physicians in various ways with various products and cover over 4,000 separate facilities. (DESCRIPTION) Slide title, 3M M*Modal speech understanding. M*Modal products point to a cloud labeled, Speech understanding: Fluency for Imaging Reporting, Fluency Direct, Fluency Align, Mobile S D K, Fluency Flex, Fluency Mobile, Fluency for Transcription. (SPEECH) So the way we've done that is traditionally is through supplying different products to fit different needs. So think about transcription. So transcription is, frankly, it's pretty easy for a physician to pick up the phone and rattle off a document and have somebody else do the work for him. And so what we've done as an organization is we wanted to supply all different types of products that are available for different workflows and various needs in different locations. So a phone is more available than a computer is sometimes. But if you want to use front end speech, you use a product like Fluency Direct. And so no matter the product or the workflow that we use for documentation with our products, it all goes back to a core cloud based platform. And that core cloud based platform shares user identities across-- or shares a single user identity across all of our products. So if you're using Fluency Direct and Fluency For Transcription together, it's a single what we call an author or user behind the scenes and so on. And so no matter how you're using, say, Fluency Direct on a desktop, all of that activity goes back to our cloud. And so once it's in the cloud, then it becomes very portable. So not only is it portable across applications, but it's portable across, in this case, operating system. So truly if you think about fluency direct, the vast majority of our users use it on Windows based systems. So ECs or maybe virtualized in Citrix or VMware. But with the Fluency Direct client for macOS, we do give you the ability to take that one step further more places to use that technology that's housed in the cloud. (DESCRIPTION) Slide title, 3M M*Modal Fluency Direct for macOS. (SPEECH) So let's talk a little bit about what Fluency Direct for macOS is. And so I'll probably be pretty redundant about this, but we've worked very hard to give our users a consistent user experience across whether it's Windows or macOS. And so what that means is that Fluency Direct on macOS allows you to speech enable web based applications that are housed in, say, Google Chrome. So there's a lot of examples of those types of systems. But you know, Athenahealth or Medhost or Paragon. There's dozens and dozens of web based systems. And so if you're using those through Google Chrome today, you can use Fluency Direct in conjunction with them on Google Chrome on the laptop or on a MacBook. Support for Citrix based applications. So if you have an application like Epic or Cerner or maybe Sunrise that is delivered via those types of systems, you can access those and use Fluency Direct on a workstation. Then we also have support for native Mac applications as well. There's less of those out there from a medical perspective, but they do exist and people do use them. Have a single user license no matter how you use Fluency Direct. And so what that means is you've licensed to us one user license or many user licenses right now. You can use that single license across windows or Mac. So you don't have to worry about extra costs incurred by using this technology. Just like we share a single identity across Windows and macOS for Fluency Direct, that same identity is shared across multiple applications or modalities for using our speech recognition system. So giving an example, if you are an Epic Haiku user. So it's the same speech profile that you use for Epic Haiku that you're using on the macOS. And then also just in Windows, we support over 66 different medical specialties with various optimizations. So if you're familiar with Fluency Direct today, you pick your medical specialty and the system optimizes for that type of documentation to give you a higher accuracy rate. (DESCRIPTION) Slide text, Single user identity, commands and customizations shared with multiple products (3M M*Modal Fluency Direct for Windows, 3M M*Modal Fluency for Transcription. 3M M*Modal Fluency for Imaging, 3M M*Modal Fluency Flex, 3M M*Modal Fluency Mobile, Epic Haiku\Canto. 3M M*Modal C D I Engage One, 3M M"Modal H C C Management). Slide text, Microphone support. (SPEECH) In terms of microphone support, so we do have a wide amount of microphone support. So headsets, Philips speech mics, Olympus mics, and things like that, handheld microphones. So that's all supported through the application. (DESCRIPTION) Slide title, 3M M*Modal Fluency Mobile microphone. A smartphone app reads, Dictating. (SPEECH) Same thing with mobile microphones. So we have support for the mobile mic and Android. And again, single application, single user experience. So if you've linked your mobile mic to your user identity once, maybe that was years ago, that same linkage will now work on macOS for you as well. (DESCRIPTION) Slide title, mac O S and hardware requirements. Slide text, At least 500 MB of available storage. If your hardware supports Catalina, it will also support the needs of 3M Fluency Direct for mac O S. (SPEECH) Hardware requirements. So on the Mac side, I like to think of it in terms of support as almost operating system level support. And what I mean by that is if you have hardware today that supports macOS Catalina, then you have the required or necessary hardware and operating system to support our macOS client. What that means is if you look at eight gigs of memory, if you're doing local recognition or maybe four gigs of memory for remote recognition, you want a certain amount of storage, so a certain amount of disk space available. And then obviously in order to get access to our cloud based user identity or user profile, you need to have an internet connection. (DESCRIPTION) Text, So, how does it work? (SPEECH) OK. So what I'm going to do is I'm going to go off camera and we're going to see a demo of what Fluency Direct looks like on the Mac. And you'll hear me reiterate some of the basic points. So go ahead and do that. (DESCRIPTION) A mac O S desktop. (SPEECH) OK, let's get the demo started. So I think one thing I want to point out at first is maybe some obvious things. The way you install software on macOS is different than if you were to install on Windows. So it's a different installer. So typically we install on Windows and MSI. Your organization might push it out or you might run it manually. On macOS, it's called a PKG. That's the file type. So you download the PKG, double click on it, and go through installs. For macOS, macOS install is a little bit different than Windows because you have to give Fluency Direct a lot of permissions. Once you give it permissions, those permissions will be remembered. And every time you launch Fluency Direct, you won't have to worry about that again. But a typical permission would be permission to use the microphone, permission to interact with other applications, and things like that. (DESCRIPTION) Clicks the Finder application. (SPEECH) But once Fluency Direct is installed, you can launch it in a number of different ways. So you can find all your applications that are installed in your Applications folder. So you could find the Fluency Direct icon and double click on it down below. (DESCRIPTION) Minimizes the Finder application. Hovers over a black icon with a white microphone, which reads M Modal Fluency Direct. (SPEECH) What I did is I put Fluency Direct into my dock at the bottom of my screen. (DESCRIPTION) A window appears, Fluency Direct. Launching. A rectangular bar with a microphone appears and reads, Timothy Ruff. (SPEECH) So if I single click on it, it's going to launch Fluency Direct. So once I've launched Fluency Direct, you'll notice a couple of things happen. So one, obviously the program started. Two, you'll notice my name here. So the system automatically authenticated to our data center to get information about me as a user. So I have my client set up here to remember my Fluency Direct credentials. So every time that I launch Fluency Direct, it automatically signs me in. Typically MacBooks or Mac Minis, they tend to be personal devices and aren't shared between users. So setting it up in this way is convenient for the user and it's not necessarily detrimental to somebody else that might use the computer, because frankly, it's a personal computer. But once Fluency Direct has signed in, what it does is it goes out to our data center, gets information about me as a user, my speech profile, user customizations, things like that. So one of the things you'll hear me talk about over and over and over again as we go through the demonstration here is that we've worked very hard to make Fluency Direct a consistent user experience no matter where you access it. So whether you access it here on a Mac device or on Windows, the training all applies. It's the same user interface. The customizations that I've done to my user account, so my templates and commands and dictionary items, those are shared across the platform. So we've done a great deal of work to have a consistent user experience no matter where you use Fluency Direct. So now that I have Fluency Direct launched, another thing you noticed is that when I launched, I don't know if you caught it, but it actually moved it to the top here. That's because that's where the last time I used the control bar, that's where I had docked it. (DESCRIPTION) He moves the bar to the left of the desktop. A translucent vertical bar appears. The Fluency Direct bar shortens to a microphone icon. (SPEECH) So you can move it to different areas of your desktop. (DESCRIPTION) He moves the bar to the top of the screen, where a translucent horizontal bar appears temporarily. (SPEECH) I prefer to maybe always on top. I prefer to dock it up here in the top middle. But all of that is remembered. So the next time you sign in, wherever you left it last, it is remembered. So if I'm using Fluency Direct on macOS, there's basically three types of applications that I would interact with. One is a native Windows or native macOS application. There's not that many of them out in the clinical space. But maybe you want to dictate into Microsoft Word running on macOS. So that's one use case that we handle with Fluency Direct. Another use case that we handle on macOS is even though there's systems like Epic or Cerner or Sunrise are Windows based systems, most organizations publish them through something like Citrix. So you can access Citrix public applications from your macOS device as well. And so it's the same user or it's the same Fluency Direct connector technology that you set up for use with Windows also applies here. So there's no extra set up for your organization to get something like Epic to work with Fluency Direct on macOS if you are deploying that via Citrix. And the third type of application you would tend to access through macOS is web based applications. (DESCRIPTION) In the dock, he clicks the Google Chrome application. A table, Electronic Health Record, with column titl4es, Time, Name, Sex, Age, Reason for Visit, Visit Type, Room. (SPEECH) So something like through Google Chrome, for example. So here is an example of a demo system that I'll use to just demonstrate basic functionality. But think of web based applications like Athena or Medhost or Paragon. There's a number of [INAUDIBLE] different web-based systems. What you can do, no matter the application, if you go to a Fluency Direct documentation site, which is docs.mmodal.com, you'll see a list of all the applications that are certified to be used with Fluency Direct. And so if you want to access the systems from macOS, I would just suggest that you validate that it is supported by our application. But we can do that all in one place on our doc site. So what I'm going to do is walk through some basic functionality of the system on macOS. So I have right now, I know you can't see me, but I have right now a handheld microphone in my hand. So we support on our macOS like we do on Windows Olympus and Philips mics and Grundig mics and so on. The advantage of having a handheld mic is, one, there's specific hardware specifically for speech recognition. You get good quality audio. The better quality audio that we have, the better we're going to do from a speech recognition perspective. The other advantage of using a handheld microphone is that we have a number of different buttons that can be programmed. (DESCRIPTION) The Fluency Direct microphone briefly turns green. (SPEECH) So for example, I don't know if you saw that, I just hit the Record button on my microphone. (DESCRIPTION) The microphone and a thin bar below turn green. Sections titled, Problems, Vitals, Medications, Documents, Discharge Summary. (SPEECH) And so that enabled Fluency Direct for listening. So it's just ease of use. So I can program buttons. I can hold it in my hand and do a number of different things all at once as well. I can also use headsets. So USB enabled headsets and things like that I can use with Fluency Direct on macOS as well as my 3M mobile microphone. So if you're used to using that, again, it's paired with your user account. You link it one time and then it's available to be used here on the macOS or on Windows, vice versa. As you move around, it's one configuration. So what I'm going to do is walk through some basic functionality so you can get an idea of the things that you can do on the Mac. (DESCRIPTION) He clicks a row in the table. Web page, McCray, Robert. (SPEECH) Again, this is just a demo system. So I can do something like go to Documents. [INAUDIBLE] Go to Documents. (DESCRIPTION) At the left, he clicks a Documents icon. Table titled, Documentation. Rows of documents. At the top right of the table, he hovers over a button, Add New Note, before clicking Free Text Entry in a context menu. (SPEECH) Add New Note. Free Text Entry. (DESCRIPTION) Text field, New Note. (SPEECH) Insert SOAP note. (DESCRIPTION) In the text field appear headings, Subjective, Objective, Assessment, and Plan. As the Fluency Direct microphone turns green, text types below the Subjective heading. (SPEECH) The patient is seen at the request of Dr. Phillips for evaluation of renal insufficiency period. So a number of different things happened there very quickly. I did actually three different commands. So those are simple commands, simple one step commands. But I used my voice to automate the navigation of EHR. I inserted a template. And that template had fields. And then I did a quick doc dictation into the system. So if you're familiar with Fluency Direct on Windows, it's all the same. It works the same. You can dictate directly into applications. Again, what we're doing here is doing things through Google Chrome. So anything that is available through Google Chrome, we can interact with and use. OK, so let's talk about, again, super easy, quick functionality. I can navigate around, automate certain tasks, and then obviously document directly into those applications. (DESCRIPTION) He clicks the Fluency Direct bar. A context menu appears. He clicks Speech Box. A window with a text field appears. (SPEECH) If I don't want to dictate directly into those applications, I do have what's called a speech box available to me. So think of this as a scratch tab where I can do documentation and then maybe move around in EHR and do other things. (DESCRIPTION) He moves a slider. The window becomes translucent. (SPEECH) We do have transparency mode here. So if you wanted to dictate the patient is a 66-year-old man. (DESCRIPTION) As the Fluency Direct microphone turns green, text types in the window. Window appears, Warning. He clicks the Close Anyway button. (SPEECH) If you want to dictate directly into the application and so on, you can do that. (DESCRIPTION) He clicks the X at the top right of the window. (SPEECH) Or dictate directly into speech box, if that makes sense. (DESCRIPTION) He clicks the Fluency Direct bar and selects Commands from the context menu. (SPEECH) OK, so some of the other functionality that we have available to you is let's look at commands. (DESCRIPTION) Window appears, Manage commands. At the left, pane, Command groups. At the right, pane, Timmothy Ruff's Commands. (SPEECH) So you'll notice the consistency in user interface between Windows and the macOS here. That's obviously all on purpose. (DESCRIPTION) In the right-hand pane, clicks Insert soap note. Window appears, Edit the command. Step 1, Insert the text. 2, wait time. 3, Run this built-in command. (SPEECH) So I can do something like, give you an example, I went through and dictate insert SOAP note for a basic template. The same basic steps are available to me. There's some differences between macOS and Windows. And we're not going to talk about that right now. (DESCRIPTION) At the top left, text field, Name, reads Insert soap note. (SPEECH) So with a command, you give it a name. (DESCRIPTION) At the left, buttons. (SPEECH) We have a concept of a text step. And a text step is the insertion of maybe a normal or a template. Rich text is not supported on macOS. And so you can see here that is disabled. If you have a command that you created in Windows that is rich text, what will happen is it will automatically insert it into on the Mac side as a plain text insertion if you've created that type of command. Keystrokes. (DESCRIPTION) He clicks a space between step 1 and 2. (SPEECH) So give you an example. Maybe you wanted to simulate a key press. (DESCRIPTION) In the text field, Command + S. (SPEECH) And a good example of that might be Command S on the Mac. You want to save something like in Microsoft Word. So keypress steps work the same as if you were using them in Windows. So that's a good way of automating a number of different applications is through keypress events. You can do something like a wait event. And so maybe you want to insert template. You want to do a Command Save. And then you want to wait a fraction of a second. So you have the ability to do that. So you can wait for time or you can wait for a particular window to be displayed. So maybe you were navigating, doing a multistep process, and you want to make sure a particular dialog is displayed before you do a keystroke to it. You can do that as well. (DESCRIPTION) Scrolls down. (SPEECH) You have the ability to control the mouse. So left click, right click, and so on. Now, one thing to be a little careful about is when you are doing documentation definitely across operating systems or automation across operating systems with a mouse. So there's probably some differences there and xy coordinates and things like that. So you should be careful about that, how you create commands. For the most part, my experience, especially in web browsers, they work the same. But there might be some differences because of screen resolution and things like that between Windows and macOS. So we talked about the mouse. We do have the ability to control windows as well. (DESCRIPTION) At the left, clicks Window button. (SPEECH) And again, what we're doing is we're building a command of multiple steps. And any one of these steps have the ability to control the system in a certain (DESCRIPTION) Step 6, Simulate window action. (SPEECH) way. So if I want to activate a window or minimize a window or restore a window or close a window, I can do that through this window action. (DESCRIPTION) At the left, clicks Command button. (SPEECH) I can also call other commands. (DESCRIPTION) Hovers over step 4, Run this built-in command. (SPEECH) So an example up here would be run this built in command for first field. So this original command inserted a template. That template had fields in it. And then it navigated to the very first field available. (DESCRIPTION) At the left, hovers over Script button. (SPEECH) Scripting is not available on macOS. So if you have scripting in your commands, that is not available right now. So you should be aware of that. So design your command in such a way that maybe that isn't necessary or you might have the difference between Windows and macOS behavior. (DESCRIPTION) At the left, hovers over Comment button. (SPEECH) And then sometimes commands get very complicated. (DESCRIPTION) Drags the Comment button to below step 2. A comment box appears between steps 2 and 3. (SPEECH) And so what I can do here is use my comment step to document and give some insight into what's going on in the command. So when I come in here in the future and maybe make a modification, I have some notes available to me so that I understand what's going on. So that's basic command functionality. Again, the reoccurring theme you'll hear from me is that it's just like Windows. (DESCRIPTION) At the left, hovers over Rich Text button. (SPEECH) Here with commands, it's just like Windows with a couple of differences. So one, we don't support rich text. (DESCRIPTION) At the left, hovers over Script button. (SPEECH) And then two, we don't support scripts. But you can create these commands. Again, you can create them on Windows. You can crate them on Mac and then they'll apply and be available in both environments if you so choose. (DESCRIPTION) Closes the window. In the Manage commands window. (SPEECH) So those are our basic commands. And then obviously we can add new commands. (DESCRIPTION) At the top right of the right-hand pane, clicks + Add button. Window appears, Add a command. Closes the window. (SPEECH) So if I want to create a new command, I can do that, not just modify commands that exist. (DESCRIPTION) In the left pane, he clicks through different Command groups. At the right, the commands change. (SPEECH) I can put commands in groups. So you can see here I have some basic functionality that's grouped together. I can organize them that way and so on. OK. So let's talk. Let's close commands. (DESCRIPTION) Closes the window. Clicks the Fluency Direct bar and selects Dictionary from the context menu. (SPEECH) And commands. So dictionary. (DESCRIPTION) Window appears, Manage dictionary. Left pane, Categories. Right pane, Person names. In the right pane, clicks + Add at the top right. A window appears. He types in a text field. Clicks a hyperlink. Window appears. Spoken Form tab is selected. (SPEECH) So hopefully you're going to notice all this is very familiar. Dictionary works the same way as on Windows. So what I can do is I can add something to my dictionary. If I want to give the example of my last name, I want to add it to the dictionary as my personal name. I want to change my pronunciation. I can do that. So rough spoken form. (DESCRIPTION) Changes R-U-F-F. (SPEECH) If I want to R-O-U-G-H. So the spoken form is rough, but I want the written form to be R-U-F-F. I have the ability to do that, write it out through my keyboard. (DESCRIPTION) Clicks Record Pronunciation tab. (SPEECH) Or I can come in here and do a recorded pronunciation. So all of that is available to me to use. (DESCRIPTION) Clicks Cancel button. Clicks Cancel button. In Manage Dictionary window. (SPEECH) And then, obviously, to save that off. Once I save it off, all this information persists in the cloud and then is available to me in the future on any device that I go to. (DESCRIPTION) Clicks Close button at the bottom right. In the Fluency Direct bar, selects Abbreviations from the context menu. Window, manage abbreviations. (SPEECH) So we talked about dictionaries. We talked about commands. Abbreviations I think are pretty straightforward. So an abbreviation is a concept of I want to say one thing and I want something else to appear. So I live in Pittsburgh here. (DESCRIPTION) Hovers over a row, A G H. (SPEECH) And so one of the local hospitals is AGH, Allegheny General Hospital. So if I were to say AGH, I would want to automatically expand available to me. (DESCRIPTION) At the right-hand side of the row, he hovers over a lock icon. A hint appears, Published by an administrator via group Dictionary Example, not editable. (SPEECH) One thing I didn't point out is this concept of a lock. So whether you go into commands or abbreviations or dictionary items, potentially you'll see things that have locks. Those are things that have been published by an administrator for your organization. So you can create a library of commands and give it to everybody, a library of dictionary entries, a library of abbreviations. And when you share them and when you publish them to multiple individuals, the user interface shows this lock here. So you know that somebody else has given it to you. It's locked and you can't change it. So again, that's the same concept that we do in Windows. The same administrative tools that you use to administer Windows users also applied here on the Mac as well. (DESCRIPTION) Clicks Close button at bottom right. Clicks the Fluency Direct bar and selects Device Button Mappings from the context menu. Window appears, Manage device buttons. A table of buttons. (SPEECH) Abbreviations. Device button mappings. So I have the ability to map all the various buttons on my handheld microphone or on my mobile microphone, foot pedals, things like that. So that's all available to me. (DESCRIPTION) At the right-hand side of some rows is a lock icon. (SPEECH) And again, I can create organizational standards if I want and publish them and even modify them if I want to modify off the standards. So you can have standards inside your organization. (DESCRIPTION) Clicks Close button at bottom right. Clicks the Fluency Direct bar and selects Medical Specialty from the context menu. Window appears, Select your medical specialty. (SPEECH) Medical specialties. So let's talk about that real quick. So the same medical specialties that are available are available in all platforms. So I think we support over 66 different medical specialties. Anything from general surgery, geriatrics, internal medicine, and so on. Those are all available. Typically you set this up once when you set up your user account and the system remembers it. The reason why we want to specify your medical specialty is each one of these, we have optimizations for each one of these medical specialties from a speech recognition perspective. And that's available to you to make things more accurate. So if I were to pick, so if I'm an infectious disease physician, I would pick infectious diseases. That optimizes the system for it to listen to things that typically an infectious disease doc dictates. And that helps with accuracy. And that also helps with performance as well. So it's a little bit faster and more accurate. (DESCRIPTION) Clicks Close button at bottom right. Clicks the Fluency Direct bar and selects Profile Training from the context menu. (SPEECH) So we talked about device button mappings, medical specialties. Profile training. You have that available as well. Typically that's done the first time you access the system. You need to do that once. So your first time using the system, you don't need to do it. Maybe you should do it is a better word. You can walk through. It's actually more of a training for you as a user. (DESCRIPTION) Window appears, Train your profile. (SPEECH) So let me show you here. It's more of a training for you as a user. It gives you instructions on how to use the system and things like that. But at the same time, we do gather some information about you and your voice patterns and things like that as you go through that helps sort of accelerate the learning process. (DESCRIPTION) Clicks Close button at bottom right. Clicks the Fluency Direct bar and selects Formatting from the context menu. Window appears, Adjust formatting. Fields about numbers, measurements, temperature, percentages, date, time, and paragraph spacing. (SPEECH) Formatting. So there's some basic formatting that we have here. (DESCRIPTION) Clicks Numbers starting a sentence. Options appear, three as a word or a digit. (SPEECH) So maybe you want to start sentences with a number or you want to write out the number. How do you want numbers to be rendered in the middle of a sentence? How do you want ordinals and ranges and measurements? Do you want them abbreviated or not abbreviated? Things like that. Or how do you want temperature represented? So dictation dates and times. So there's a lot of functionality that's available to you here so that you can have things presented in a standard way in the way that you want them to. (DESCRIPTION) At the right-hand side of some of the options is a lock icon. (SPEECH) Now, you also see here locks. So your organization may have decided for you how they want things to be standardized. So temperature needs to be certain way or abbreviations or [INAUDIBLE] and things like that. That, again, can be published and defined by your organization. (DESCRIPTION) Clicks Close button at bottom right. Clicks the Fluency Direct bar and selects Settings from the context menu. Window appears, Manage settings. (SPEECH) Settings. Let's see here. So settings. Some basic security questions. So if you want to change your password, it makes it a little bit easier for you by giving you hints and questions and things like that and your windows layout. (DESCRIPTION) Clicks Window layout. Options, Same across all computers, Per computer. (SPEECH) So you can have things remembered across computers or per computer. So that's up to you. That's available. (DESCRIPTION) Clicks Close button at bottom right. Clicks the Fluency Direct bar and hovers over Help from the context menu. (SPEECH) That's the gist of it I think basically. So you have your online help. You can change your password. You sign out and things like that. But again, the theme I would like to sort of put in your mind is that it is a consistent user interface across actually no matter where you use the application. So on a Windows device or a macOS, same credentials, same customization commands and dictionary items that are available to me. Administrators use the same functionality to define user accounts and standardize things across the organization. So everything's the same. And like I said, we've worked really hard to give that consistent user interface. One, so the users are familiar with them. Two, maybe if you're a trainer in an organization, you don't have to train it different. So it's the same application, same look and feel, same usability. So that's the gist of my demo. And we're going to go back and take a look at some slides to sort of wrap up the discussion. (DESCRIPTION) Text, So, how does it work? Slide title, Flexible administration. A web page, Manage users. A window titled, Edit a user. (SPEECH) So I guess let's talk about-- so one thing we didn't really go into much detail is about administration. So I talked about being able to publish information across users. It's the same administrative tools that you use for any existing Windows installs that you would use for macOS install. I just realized I threw a picture of our new administrative tool. So we have a number of customers betaing this right now, and it just occurred to me that maybe some of you haven't seen this. But we do have a newer administration tool coming out that has, we think, a better look and feel, better user interface, better functionality. But that will be available to you. But the point I want to drive is the fact it's a single administrative portal tool, whatever term you want to use, through the cloud in order to service your set of users no matter where they're using the software. (DESCRIPTION) Slide title, mac O S and Windows similarities. (SPEECH) So let's talk about to reiterate some of the similarities. So again, we've worked very hard to make the user interface the same. So there are some subtle differences, and we'll go into a little bit more detail about again. But we feel that the application is similar enough that very, very little training is needed in order to go across OSs. A single user license. So that same license that you use on macOS is available on, say, Epic Haiku, which is also available and Fluency Direct in Windows. Same customizations and things that you've added in one place is available everywhere. The support for various applications. So you think about we talked about applications available through Google Chrome. So any application you can access through Google Chrome is automatically available to you through the Mac OS interface. So there's a long list of those. Same thing with think about Windows applications. As long as you're publishing those through Citrix infrastructure, those are available to you as well. So a lot of the large enterprise EHRs are Windows based. But you can make them available too, and you probably already do, to your Mac OS users too. (DESCRIPTION) Slide title, mac O S and Windows differences. Slide text, No foot pedal and hand control support. 3M, 3M M*Modal Fluency Direct, 3M M*Modal C D I Engage One, 3M M*Modal H C C Management, 3M M*Modal Fluency Mobile, 3M M*Modal Fluency for Imaging, 3M M*Modal Fluency Flex, 3M M*Modal Fluency for Transcription, and 3M" M*Modal Fluency Align are a trademarks of 3M. All other trademarks are property of their respective owners. (SPEECH) So let's talk about some of the differences. So we talked about commands are a little bit different. So we don't support rich text or scripting. Again, there's some automatic adjustment for rich text. So you maybe have defined something as a rich tech step. It just won't be inserted as rich text. So it would be more of a flat text. You sometimes have to be careful about how you define commands. So if you're using a lot of automation, particularly maybe mouse or Windows manipulation, a Window might be titled a little bit different on macOS than it is on Windows. So you have to be concerned about that. Or maybe the xy coordinates of a particular button that you want to automate pressing is a little bit different between the different operating systems. So when you define your commands, you just need to be conscious of that and just be careful about how you'd find them. And in my experience, as long as you're thinking about it and you just compensate for little changes in xy's and things like that, then you can still have a single command that services both areas. Device button mappings. So again, single definition. But if you think about the two operating systems are different. So you have the Windows button that a lot of things happen off of in Windows. That doesn't exist on the Mac. So anything that you're doing through the Windows button or automating through the Windows button, you really can't do on the Mac. Same thing with the Command button on the Mac. So the Command button on the Mac doesn't exist on Windows. And so you just need to think about that as you create commands and so on. The control bar location. So the memory for that is we store that. Again, you still have memory. So if I were to go from a Mac OS device to Mac OS device, it's going to remember where I want my control bar. But we do separate Windows and macOS because there are differences there. So just think about the dock at the bottom for macOS or so on. So there's just maybe a different preference you want for each operating system. And so the software accommodates for that. One thing I'm not sure I talked about is importing and exporting of commands. So you can always use importing and exporting on the Windows side to share a command library. And so one of the things with our new administrative tool is you'll have that import export ability through the administrative tool as well. But right now today, I can't export a command from the macOS device. Virtualization. So we support Citrix with the connector today. We don't support the connector even though we have a connector for terminal services or remote desktop, remote app, or VMware VDI, or VMware RDSH. Today that is not supported on the macOS side. So there's a difference there. So as a strategy on the 3M M*Modal side, we have these plugin applications. CDI Engage One, HCC Management, Fluency Align. They are all plugins today for Fluency Direct. So those are products and services that you license separately. Today they're not supported on macOS, but they are supported on Windows. The install process is different. So we support-- well, users can install the software themselves. Behind the scenes, there are tools like SCCM, Microsoft SCCM, where you can still push out Fluency Direct onto macOS devices. So there are automated tools that allow you to at an enterprise wide push out software and push out changes. But it's obviously a different installer. So how you go about installing something is a little bit different. On the macOS side, one of the big differences in the install process is you need to give the user permission to use various things on the device. And I like likened it to your phone. So any time you have an app on your phone, you have to give it permissions to use the microphone or permissions to your contact list and so on. So on macOS, it's very similar. So Fluency Direct needs permission to use your microphone. It needs permission to interact with other applications on the desktop. Once you've given it those permissions, it remembers that moving forward and you no longer need to worry about that. Another feature that's missing that we didn't talk about is, and I think very rarely used, is playback dictation. So if I dictate something in a note, I can't play it back the audio. You can do that today with Windows. That's not available on the Mac. Somebody put in the chat or Q&A about LDAP or Active Directory integration. So that's another difference too as well. So the strategy that we have on the Mac, and I'm up for re-evaluating that, but the strategy that we have on the Mac for sort of a single sign on approach is what you saw. One part of what you saw is, for the most part, we feel that macOS devices are personal devices. And so having it remember the last sign on so far has seemed to work pretty well for us. The other thing is we do support from a connector type install. We do support various SSO approaches there, Single Sign On approaches there. So if you have, say, an Epic install that uses a single sign on approach and launches the Fluency Direct client and signs it in, that will also work on macOS. And then also-- and so on. Let me go on here. And then I think we're at the questions. And so I don't know, Lisa, if you want to state those. I think I answered at least one of them already. Yeah. [INAUDIBLE] question. Absolutely. Thanks, Tim. No, that's a lot of information. And this webinar will be available on our website in the next couple weeks. So if you do want to go back and listen in and rewatch the demo, that will be available. Also, if you are interested in learning how to adopt this at your facility, there's a couple of different ways you can do that. We will provide contact information. But also within the platform in the middle of your screen, it does talk about how to reach out to your adoption specialist. If you don't know who your adoption specialist is, if you click on that Learn More, that will bring up-- that will generate an email for you so you can reach out to the general team and they can help you out. But yes, we do have a couple of questions that we can get to. So the first one is what is the cost per user for this software? So again, it's covered in your licensing already. So you license today-- you license a Fluency Direct user and we don't care where you use that. So it can be used with various accounts or various operating systems. So Windows, macOS, on a Haiku, on your iPhone and Android and so on. Next question is, how do I remotely install the software? So again, for the most part, I think what we're seeing so far is users download it and install a PKG. There are some organizations that use tools like SCCM to push it out. And so that's available as well. So I don't know much about SCCM on macOS, but I know it can be done. OK, great. Next question we have is, do you support other types of virtualization other than Citrix? Yeah. And so not at this point. So the focus is just on the Citrix base. Oh, go ahead. No, go ahead. I was going to say the last question that we have is, can I use an Android phone as a mobile microphone or does it have to be an, I'm sorry, be an Apple iPhone? So you can use your Android device or your iOS device. I think there's some more questions coming in here. Oh yes, they are. As soon as I said that, those started coming in. How does M*Modal perform with cloud based EHRs either on Mac or a Windows device? So I mean, I guess it's your definition of a cloud based EHR. So if you're accessing it through a web browser and you're accessing that through Chrome on macOS, then we're good. So you can interact with anything that's available through the web browser. If your definition of cloud based EHR is delivered via Citrix, as long as we have that connector software installed and it's the same software that you probably already have to support a Windows environment, we're good there as well. So we do support that. OK. Another question we have is, does HCC messages and CDI provider queries and nudges function on a Mac? So not at this time. So that's one of those plugin functionality that is not supported. So today I would characterize our macOS support is for front end speech recognition. So speech enabling those applications and not necessarily using our [? CEPD ?] plugin infrastructure today. Will this application support an iPad? So we would have a different approach on the iPad. So it wouldn't necessarily be this software. I'm trying to think. I guess it depends on what you're trying to do on the iPad. So I know that customers have deployed EHRs through Citrix on an iPad. And the way I would approach that is I would install Fluency Direct into the Citrix environment, the Fluency Direct client into the Citrix environment. And then you can-- I know that some customers have had success doing that. And so you're sort of seeing a full Windows desktop right on your iPad and you can hit Record. There is also, if you're talking about speech enabling sort of a native iOS application or maybe speech enabling something through a web browser. So there is something also that we have in the store. You'll see there's a keyboard option that you can install on an iPad or an iPhone. I'll say it's a little clunky, but you can install a specialized keyboard. And that specialized keyboard has a recording write or recording thing. And when you hit record, it will use our speech recognition. And sort of just like the keyboard or Siri would insert text into another application, our keyboard does the same thing. Like I said, it's a little clunky, but that's an option today. Well, that is the last question that we got in from attendees. Tim, is there anything else you want to add before we go ahead and wrap things up? No, Well, I think just that we would love to work with you to expand the use of Fluency Direct into other parts of your organization. And so that's what this tool is intended to do. We think that-- like I said, it's not exactly like we have in Windows, but we believe that the functionality is such that it would be very successful for a lot of your users and meet some of the gaps that maybe you have today. Great. (DESCRIPTION) Slide title, Want to learn more? Contact your adoption specialist today! Slide text, l have a technical question... Fluency Direct Help Desk 855-268-5316. Heading, 3M M*Modal Adoption Management Contacts. Manager, F D Epic Team, Jillian Tomich, j m tomich at m m m dot com. Office: 412-339-2381. Mobile: 412-523-5194. Manager, F D Enterprise Team, Jeremey Ley , j ley at m m m dot com. Mobile: 907-513-6624. Manager, Clinician Solutions, Adoption Services, Tryphena Cockroft, t n cockroft at m m m dot com. Mobile: 724-496-7480. (SPEECH) All right, well, like I mentioned, here is the information to reach out to your adoption services specialist. So if you are interested in learning more. If you aren't sure who they are, you can email us at that website-- I'm sorry, the email address at the top, that hi-fd-newsletter-updates@mmm.com. That will also get you to the right team. So you can always ask the best way to get in contact with us. And like I mentioned, again, we certainly appreciate you completing the survey to let us know. And if you do have any other questions that you thought of, please feel free to put that in the survey as well. There is the ability to add feedback there. (DESCRIPTION) Text, Thank you! (SPEECH) And so again, we appreciate your time today. Tim, thank you for your time as well. We will be sending out just a brief email also after this webinar that will give you a link to our webinar section on our website. So in the coming weeks, if you are interested in rewatching, you'll be getting that link here shortly. So again, thank you all very much. And we hope you have a great holiday season and we will talk to you again soon. Thank you all for joining us today. Yeah, thank you for your time. (DESCRIPTION) He waves.

(DESCRIPTION) A slideshow. Text, 3 M.. Science. Applied to Life. 3 M Virtual Pulse Webinar Series. 3 M, M Modal. Copyright 2022, 3 M.. All rights reserved. Slide, A great company is showing what interesting applications a fantastic product can bring for motivated users. The slide depicts a webinar platform, featuring a media player, chat box, and survey, with a menu bar at the bottom. (SPEECH) Good afternoon. And welcome to our October Virtual Pulse-- Creating time to care. (DESCRIPTION) Slide, Creating time to care. Who is caring for the caregivers? October 2022. (SPEECH) Before we get started, I am going to go ahead and go over a couple of housekeeping items before I introduce today's speaker. (DESCRIPTION) Slide, Housekeeping, a bullet point list, and a disclaimer. (SPEECH) If you've been with us before, you might recognize this platform. But just in case, we are using On24 which is a web-based platform for a better user experience. Because this is a web-based platform, there is no dial in number. Make sure you're using Chrome. Close out a VPN or multiple tabs, that will help with bandwidth. If you are having audio issues, check your speakers settings. You can also do a quick refresh. And that usually does help with any issues that you might be having. You see in front of you a lot of engagement sections. So we do have a media player. So if you do need closed captioning, you can certainly utilize that feature. In the center is our presentation as well as a Q&A box. We encourage questions throughout the session. We'll get to as many as we can. So if you do have a question, feel free to put that in there. We do have a Resources section in there as well where you can download the certificate of attendance as well as the presentation and other resources like our upcoming webinar in November if you want to register for that as well as some other items as well for resources. We do also appreciate feedback. So at the end of this session if you can complete the survey. We appreciate to hear how we did. And if you do have any questions, like I said throughout, please put them in that chat section and we'll get to as many as we can at the end. (DESCRIPTION) Slide, Meet the speaker, a picture of a man. Text, Travis Bias, D O, M P H, F A A F P. (SPEECH) So again, I'm going to go ahead and turn this over to Dr. Travis Bias, who is our chief medical officer here at 3M. If you want to learn about Doctor Bias, you can in the speakers section just to hear a little bit more about him. So I'm going to go ahead and turn this over so we can get started. (DESCRIPTION) Dr. Bias appears on a video call in the top left-hand corner of the slideshow. Slide, 3 M Health Information Systems at a glance, a list of statistics and data. (SPEECH) Awesome. Thanks so much Lisa appreciate it. And welcome all of you. Again, my name is Travis Bias. And I'm a family medicine physician as well as Chief medical officer of our clinicians solutions team within 3M's Health Information Systems Division. And so clinician solutions are business unit, is effectively their Modal business. And we're going to get into a little bit of history there behind our division as well as their Modal business and clinician solution specifically. And so we're going to get into a bit about our ambient clinical documentation solution and all of our capabilities that are built under the hood that support its production and delivery. But first, I want to put in, again, a little bit of historical context around the history of 3M, our Health Information Systems Division and Modal, and why and how we are working on and make clinical documentation solutions as well as the capabilities under the hood. So HIS or Health Information Systems Division has several-- many of you on this call may be very well familiar with HIS's history and the clinical documentation integrity encoding space. That goes back really four decades to the housing and evolution of patient groupers such as the diagnosis related groups. And those that we've evolved into such as the APR, DRGs, as well as clinical risk groups. And other groupers that you may be familiar with. That experience and familiarity with groupers has then built into a lot more clinical documentation improvements and integrity encoding technologies as well as I think, a high level of comfort with supporting a whole range of revenue cycle solutions that really are now growing into abilities to help improve the efficiency of physicians. And we're going to be capturing a complete and accurate record and we're going to dive into a little bit more about that. (DESCRIPTION) Slide, The opportunity for technology in health care, two pie charts, titled Current time allocation, and Estimate with ambient intelligence. (SPEECH) First before we get into some of those efficiency pieces, I wanted to be clear about the problems that we aim to solve. So they're a few major buckets here. So first we get that there are many different competing demands on clinicians time today. So on the left here, you see there's-- this from a study here in the past you can see that throughout the day physicians are spending a ton of clerical and administrative time within the electronic health record that can pull them away from focusing on more complex patient cases or other tasks that we actually went to medical school to learn. And so you can see on the left there, a lot of that is in Inbox. It's on documentation. It's on order entry and billing with really only about a third of that time left for direct patient care. And I think a lot of us have heard the stats about how for every one hour of patient care physicians are spending a couple of hours in the electronic health record. And I can promise you that was not what I originally envisioned back in the early 2000 when I decided I wanted to go to medical school. And so the goal really with our technologies and what we think within the clinical documentation is to shift over to the right where we're expanding the pieces of the pie chart that involve direct patient care, of course, while improving efficiencies and decreasing the amount of time that we're spending on clerical and administrative tasks within the electronic health record. And maybe even-- so freeing up time for more complex cases maybe so taking care of more underlying issues for patients, but also maybe expanding the time available for personal time or for using that time however physicians or clinicians wish. A second issue that beyond simply just sheer time demands, we know that there are numerous interruptions throughout our day that can demand cognitive test switching. And those can be enormously disruptive. Those disruptions can negatively impact the physician stress level as well as the completeness of the chart. And finally, thanks to this transition that we're seeing from volume to value-based care. Physicians are really asked to help capture-- more and more we're being asked to help capture the complexity of our patient population, whether that's for reimbursement purposes or for quality measures for reporting as well as risk adjustment which really impacts all of those things for reimbursement in a value-based world as well as in getting to those expected or benchmark values in quality reporting. And so all of these are critical tasks to the evolution of our health system for sure. But again, these demand a ton of clerical and administrative work in the electronic health record on behalf of physicians. (DESCRIPTION) Slide, Ambient intelligence: Conversational A I, a list of statistics and data. (SPEECH) And so considering these problems and trying to build on our decades of experience in clinical documentation integrity encoding, 3M acquired in Modal back in 2019 with the goal of really plugging that physician facing-- that clinician facing piece. So the front-end speech recognition and natural language processing capabilities to our middle and back-end experience with clinical documentation integrity to really try to create a more seamless and integrated CDI experience for clinicians really from front-- all the way from the front-end to the back-end. And so and Modal's capabilities have been built on a deep experience with across varying types sizes and types of organizations across the country with over hundreds of thousands of clinicians using our front-end speech recognition tools. And also many customers across the country using our natural language processing or what we like to call natural language understanding might hear me call it NLU. Using those technologies across and capabilities across the country. And so this experience with large amounts of data allows for the training of our core machine learning capabilities as well as to strengthen that NLU engine. And so all of this ideally to the clinicians benefit. So in Modal's experience I talked about 3M EHRs, our experience going back for decades. Their Modal's experience goes back 20 years starting first with their transcription business, which then informed the evolution of their front-end speech recognition capabilities. Where we really naturally understood, but came to learn even more that you've got to be highly confident in the technology's output before you can put it in front of a physician. And so it's that approach-- it's an incremental approach to developing and deploying new technologies, but also a focus on accuracy, that's really been behind our pursuit of ambient clinical documentation. (DESCRIPTION) Slide, Outcomes: 3 M M Modal Fluency Direct, a quote and a list of statistics and data. (SPEECH) So our approach first with just-- starting with just plain front-end speech recognition alone. In addition to the chart commands that fluency direct our speech recognition tool allows, has really resulted in a ton of time savings and a positive physician experience as can be seen with our high adoption rates as well as, of course, an increase in productivity. So this improves documentation capture which is really central to sound value-based payment. (DESCRIPTION) Slide, Engage the clinician at the golden moment, a bullet point list, and an illustration titled Closed-loop C A P D. (SPEECH) And so building upon the speech recognition capabilities we also have again our natural language processing or what we like to call our natural language understanding engine and capabilities so that NLU. So the goal of the bulk of these solutions is really to address the time and attention deficit that I mentioned by leveraging our NLU engine to proactively deploy nudges in real time that identify gaps and documentation that need filling. And ideally improve physicians efficiency with filling those gaps beginning with the first time around in the chart. So these are those computer assisted physician documentation or CAPD capabilities. And so ideally this is automating some of those lower hanging fruit CDI tasks for the clinician to free us up to focus on more complex, higher level tasks that we were actually trained to perform and want to focus on for ourselves and what patients want us to focus on. And it also bolsters the efficiency of CDI teams in the inpatient and ambulatory space as we can automate some of their repetitive tasks and free them up to focus on higher level more higher yield targets for their organization and for your practice. (DESCRIPTION) Slide, Real time clinical intelligence at the point of care, a bullet point list, and an example of a digital chart. (SPEECH) And so on the next slide, I'm going to show you a bit about what our control bar looks like. So if any of you are using our fluency direct or speech recognition tool you're going to note you're going to be very familiar with this control bar up here on the top. That control bar for the user interface can be docked on the side where you can just see the microphone in that ribbon there with the number 3 in it. And this is what our CAPD functionality is plugged right into. And so these-- and this is where those nudges and even manual queries can be delivered if needed but where those nudges can be delivered through that same control bar. So getting to documentation efficiency and accuracy, we are proactively deploying nudges in real time during charting So while either I'm dictating or I'm typing, our NLU is reasoning over that screen in real time. And within seconds, we're leveraging our NLU engine to deploy these nudges as we document. And this is again, with the whole goal of capturing a more specific and accurate record to reflect the complexity of your patient population. (DESCRIPTION) Slide, Ambient intelligence: Conversational A I, features of the platform, including a dropdown list and a document with patient data. (SPEECH) And so this is all grounded, sorry, let me build out this slide a little bit. So I might type in, for example just to give you a couple examples here. I might type in that the patient has heart failure in general. And within seconds I'm getting a nudge within that control bar that says, is that acute or chronic, is that systolic or diastolic. I might type in CKD or chronic kidney disease, and I might get a prompt that says, OK, what stage of chronic kidney is that disease is that specifically. And when we layer in encounter-based reasoning, which in the inpatient space takes into account all of documentation throughout that entire admission, these capabilities become even more impactful. And so these nudges we've built over 400 of these clinical rules out already. And we work with organizations to build these and keep these updated. These nudges are highly customizable both in content and in frequency, that allow you and your organization to balance these along with any other notifications coming from inside or outside the electronic health record. And that local customization can really help to strengthen local initiatives that are going on. So these are really meant to not be a hard stop, these are meant to be soft stops and meant to be, again, nudges, something that where we think if we can present the right information at the right time, then physicians are going to naturally document over time in a more complete, and accurate, and compliant manner. And so the whole goal of these nudges too is to, rather than send me a manual query maybe days later after I've been in the chart to where I've got to then go back into the patient's chart, remind myself who is this patient, what was their story, what was that specific diagnosis that might help capture something more completely. I'm getting nudged, again, in real time ideally beginning with the first time around the chart. Many times then satisfying the nudge as I dictate or type in real time. And so that's-- again, the goal is getting that complete and accurate record for the first time around, but then ideally preventing some of those disruptions with manual queries maybe days or weeks later. And so all of this is integrated within the electronic health record operating quietly while I'm dictating. And we have tight integrations with hundreds of electronic health records because this is, again, enabling reasoning over documentation that's occurring in your electronic health record in real time. (DESCRIPTION) Slide, 3 M clinical intelligence: N L U, a bullet point list, and a circular diagram showing treatment and diagnosis options for heart failure. (SPEECH) All of these NLU capabilities are built on this cloud-based NLU engine that is grounded in Google based clinical intelligence. So you might hear about machine learning and artificial intelligence. But this is a rule-based clinical intelligence that is built on internationally recognized ontologies such as SNOMED, which really allows for more detailed and structured clinical insight beyond simply ICD-10 codes. It captures context as well as unstructured information. So again, I mentioned CKD or chronic kidney disease. We have a tight grasp on different ways that we might use acronyms or might describe terms. And we've learned, again, honed that over a couple of decades. And capturing that unstructured information is really key. We know that there are many ways to mind discrete data elements, but getting to that unstructured or narrative information is really where a lot of the power lies. And so whether you're inputting, again, via speech or type or copy paste or the templates that you've already built out, we're able to reason over whatever is on the screen and that's, again, another strength of ours. And again, using all these capabilities to identify gaps and documentation that need filling and doing that in real time. (DESCRIPTION) Slide, Ambient intelligence: Explainable A I, a screenshot of an Evidence Sheet, with a button labeled Generate query. (SPEECH) So our NLU engine is not-- again, I mentioned it's a rule-based clinical intelligence and natural language processing capability. And so all that means is that this isn't a black box. I mean, you're able to-- we work with organizations to see all of our rules, to see the different things that our internal content teams have built to help these rules fire appropriately. And again, to reason over documentation appropriately. And it's that transparency I think that a lot of organizations and physician leaders like because they're able to go in and review these rules before you're deploying anything to your front-line clinical staff and make sure that it aligns with your local and your organization's standards. So these rules then enable us to benefit again from the power not only from the discrete fields but from mining that unstructured or narrative information which, of course, we all know as a primary care physician knows, it's very rich in content. And so being able to combine those areas for a more sophisticated insight into the record is really hugely impactful. Beyond simply what's in the record, we can also identify what's not in the record. And so again, through a long history with clinical documentation integrity, a lot of our goal is centered around capturing a complete and accurate record. And so as CDI teams or as clinical leaders work to capture a compliant record, this is something that can be leveraged to make sure that we're capturing diagnoses appropriately and in a compliant fashion from the beginning. (DESCRIPTION) Slide, Outcomes: 3 M M Modal C D I Engage One, a bullet point list, and a list of statistics and data. (SPEECH) And so we are seeing with our CAPD tools, already a ton of great successes across the country at large and smaller systems, not only through these clinical documentation integrity efficiencies, through decreasing manual queries, but also through accurate patient safety indicators or those PSI-- capturing those PSIs in a more accurate fashion. And so capturing that more specific documentation helps in many instances-- many regards through more accurate risk adjustment so that risk adjustment can then help with reimbursement in a value-based world. But also that risk adjustment helps for accurate quality measure reporting. So getting to those accurate expected levels, so a lot of people are familiar now more and more with how prevalent some of these analytics platforms are becoming. A lot of physicians are really interested in dialed into how they're being judged in these quality measures. And so a lot of folks are a lot of leaders are using that to then help their clinicians understand. Well, the way to get to these expected values or to these benchmark values, that's coming from documentation. So the more that we can get more detailed and accurate documentation up front, the more fairly we're going to be judged in that denominator for a lot of these quality measures. And that's how a lot of physician leaders have encouraged their clinicians to engage with our technology. And that's a lot of the messaging that we help to support them with. (DESCRIPTION) Slide, 3 M M Modal Fluency Align. (SPEECH) And so fluency Align which is our ambient clinical documentation solution really builds upon all of these capabilities. So our proactive nudges that improve documentation completeness and accuracy and specificity. Explicit commands, they're are virtual assistant technologies such as Hi Epic that allow a clinician user to control the electronic health record through, again, the explicit commands. And through 20 years of transcription and now front-end speech recognition experience to help capture even beyond those explicit commands but maybe even implicit commands during the office visit to where ultimately the goal is that the documentation becomes a byproduct of that entire encounter. And so we're getting all those capabilities rolling up under one integrated user experience is what we're calling fluency align. (DESCRIPTION) Slide, 3 M Fluency Align, a picture of a digital device, and a screenshot of a virtual assistant. (SPEECH) So fluency align we think of as being the role of two major technological capabilities. So one is that ambient clinical documentation or that note generation piece. And secondly is leveraging our virtual assistant technology. So again, that capability for the user to explicitly control the electronic health record for various purposes. (DESCRIPTION) Slide, 3 M Fluency Align, a list of definitions. (SPEECH) And I'll get into that a little bit deeper here. So expanding on those ideas. We think of this technology is really enabling three major buckets of action. So number one is capture, again, capturing that clinical record directly into the electronic health record in a complete and compliant fashion. Number two is surfacing insights. So maybe using our virtual assistant technologies and explicit commands to surface relevant information. So maybe the latest hemoglobin A1C, the latest blood pressure, the latest creatinine. Surfacing the right information at the right time and even maybe surfacing proactive insights like our nudges that I mentioned earlier. Lastly, enabling actions. So maybe using, again, our virtual assistant technologies to input orders or maybe through implicit understanding. So what we mean by this is discerning from the clinical conversations specific orders and maybe entries into the record like the HPI or the plan simply from gleaning that from the clinical conversation. So I might be discussing with Mrs. Smith, I think it'd be a good idea to check your blood pressure twice a day for the next two weeks and send me that record. And we'll go ahead and start you on lisinopril 10 milligrams once a day. Well, then let us just go ahead and surface that order even though I haven't said OK, Fluency, start the patient on lisinopril 10 milligrams once a day. So that implicit understanding is a piece that, of course, I see is potentially a high value piece but, of course, the most challenging piece as well. And so taking a big step back really, thinking about what do we have the capability to do right now with that captured piece with M in clinical documentation. What can we really do right now today. So right now what we can do is listen in to the entire clinical encounter, transcribe that encounter, that conversation between patient and physician. Transcribe that, run it through our natural language understanding engine and then surface a protonote or a draft note to a back-end human editor or scribe who then can edit the field to make sure the discrete fields are complete and accurate, make sure the narrative fields are complete. And then return that fully completed note ready for final edits and signoff to the physicians and basket within three to four hours technically. We promised within 24 hours right now. With the goal of gradually over time, increasing the role for technology and automation and gradually decreasing the amount of human life required in that process, of course. (DESCRIPTION) Slide, Path to automation, a graphical timeline. (SPEECH) And so this next slide I really like because it really helps better visualize. I'm just going to go ahead and build it out for us. It really helps to better visualize how we're thinking about that evolution from time 0 on the left where this is effectively an asynchronous drive workflow that I just described to you. As we move from time to zero on the left to the right, as in that green bar you see gradually going down the amount of human intervention and editing required to this point in the middle where you've got still-- there needs to be some human quality review although the technology is handling more and more to then ultimately off to the right where we're getting a fully auto generated note in real time handled by technology, which I think is also our goal. And so as we automate more and more-- and a lot of this is taking place under the hood right. We're automating more and more technology's handling more and more, a human editor is doing less and less behind the scenes. The quality expect-- my opinion is that the quality expectations on behalf of the physician ought not change. Over time we're going to be more and more confident, the output of that technology to the point that we can put it directly in front of the physician. But for now, we're very clear about the need for a human editor on the back-end before that completed note reaches the physicians in basket. And that's because that we know, again, reliability and accuracy have got to be our top priority. And that's for patient safety reasons number one, but also because we know that there's no quicker way to frustrate clinicians than to deploy tech bells and whistles before they're ready for prime time. So again, it's this incrementalism-- this incremental building, deploying, and introducing new technologies, and this focus on reliability and accuracy and quality. That's really informed our approach over the past few decades and will continue to guide our work towards this ambient clinical documentation solution. (DESCRIPTION) Slide, a cartoon depicting an exam room, with a doctor pointing to a line graph on the wall and saying to a patient, This is your I N R trend since starting Coumadin, and the patient saying, Oh, I see. Devices hang on the wall next to the graph. (SPEECH) And so I'm going to leave you with this cartoon, this image that I love that really, in my opinion depicts our overall goal. So you see just to the right of the rotoscope there, there's ambient listening somehow going on in the room, whether it's through our ambient listening device or through an app on the phone without necessarily requiring a ton of explicit direction from the user. And so that's where the recording is taking place. The attention in the room is focused between the patient and the physician building those human connections that patients crave, and that make our work more rewarding. Technology as you can see on the wall of the screen, their technology is augmenting their interaction, whether that technology is explicitly directed or implicitly requested to do so. And for documentation again, ultimately becomes a byproduct of that entire encounter. And so that's the ambient future that we're building towards. (DESCRIPTION) Slide, Physician engagement, Communicating the why, a diagram titled Strategic pillars, depicting three interlocking circular puzzle pieces. (SPEECH) We've been learning a lot from our customers through our implementation and adoption teams about the importance of really demonstrating to physicians how engaging with any of our technologies can help drive efficiencies towards more complete and accurate documentation capture. And this is, again, towards-- on the wheel here these areas that I've discussed, but really towards the end of more accurate risk adjustment, again, for potentially reimbursement purposes and/or for quality measurement purposes. And we've seen that those organizations that can tie all those efforts back to their strategic pillars in addition, of course, to framing the value for physicians day-to-day work. And those are the organizations that end up with the most active and engaged physician users. (DESCRIPTION) Slide, Goal of ambient intelligence: Support the quintuple aim, a circular diagram consisting of 5 wedge-shaped pieces, each containing a goal. (SPEECH) And so our-- my final slide here, our overall goal really is to leverage our technologies to support the quintuple aim. So, of course, you've heard a lot about how the goal is really to improve the clinician experience as well as the patient experience by moving technology further and further into the background, especially with regard to a ton of those clerical and administrative tasks. But it's also our hope that through more efficient and complete documentation, that will improve population health outcomes through a few ways through freeing up physicians, to have more time, again, to focus on more complex tasks and more complex cases, as well as through more reliable information for our health system. And that more detailed insight into population health measures and data will ultimately improve efficiencies and reduce redundancies, and ideally help to lower costs overall. And ideally help to direct resources to populations where they're needed most, impacting that fifth aim of health equity. So as we are developing new technologies and thinking about how our technology is going to be used in the real world by large inpatient systems as well as small independent practices in the ambulatory space, we stay focused on how our technology is able to be used as a tool to impact and really move the ball forward on reaching the quintuple aim. And so with that, I hope you've gotten a bit of a flavor a lot of our technologies which hopefully some of you are familiar with already, and how that's ultimately going to be building towards this ambient clinical documentation future which is being kicked off of pilots this year. And so with that, I will stop. And thank you so much for your time and your attendance and your attention. And we can take any questions that you all might have. Great. Thank you so much and we certainly appreciate the information. It was a great presentation. And we do have a couple of questions. So the technology that you were discussing, one of the questions that came in is, what type of facilities and practices are best fits for this technology. Well, there are a lot of-- many things I can say about that I think there are several practices-- facility wise we're starting more in the ambulatory space just because with the ambulatory-- excuse me, with the ambient clinical documentation pieces, just because that's a little more of a controlled environment for recording conversations and discussions. Of course, our CDI and speech recognition is used throughout all sizes and types of systems and organizations inpatient and outpatient. However, this ambient clinical documentation, of course, recording the conversation can happen a little more easily in the ambulatory space as we work to understanding and transcribing the conversation is not the challenge, we're quite good at that. It's highly accurate. The next challenge then becomes, how do we appropriately diarize and make sure we know is that physician speaking, is it patient speaking, is it the caregiver speaking. And sort that out. And understand how to organize that transcription into a coherent note, into handling dropdown fields into a narrative that makes sense for me. That might be different for me as a primary care physician than from the orthopedic surgeon down the street. And so we're also starting with specialties that might have a bit more templated notes because obviously, again, handling discrete and dropdown menus and whatnot can be a bit easier to tackle upfront than understanding what exactly from the conversation really ought to be in the narrative. And so I think that answers that question. And, of course, we have a lot of organizations that have various pieces of our solutions in place already. They already have fluency direct so the speech recognition or they already have Hey Epic or some other virtual assistant technologies that, again, our virtual assistive technologies are underpinning a lot of those different branded solutions out there. And so if you-- the way that we've talked about, that I've talked about is if you've got some of those building blocks in place already, it's going to be easier to naturally evolve into using an ambient solution because it already understands your speech profile, already understands a lot of things that are necessary to just I guess coalesce those building blocks into place for ambient documentation. (DESCRIPTION) Slide, Questions? (SPEECH) And I think that starts to answer that question. But we're piloting with some groups this fall which is very exciting. And our user experience researchers and our teams are really starting to hone in on. What are some things we haven't thought about. How is this being deployed in real time, which, of course, is going to differ widely from a large health system, a large clinic to a smaller clinic are going to vary from a primary care physician who I like my long HPIs and my long plans to show what I've been thinking versus the orthopedic surgeon who have got orthopedic surgeon buddies but their plan might be a little more short to the point. And it might just-- it's going to depend even within specialties how they're using this technology. And so we're going to be learning a ton from that over the next 6 to 12 months. And we're excited to show you what we've learned. Great. I think a good follow up question to that that we have is, how do you protect patient data from this-- basically like you said, it's listening in to the sessions and recordings. So can you talk about that a little bit. Sure. I think that's another piece that we'll be learning because there will be audio capture and recordings and data that we have to manage. But again, we've been doing this for decades. So we've got a high level of comfort with large amounts of data and with protecting PHI and protecting data that ought to be protected in the right way. I think what also people are wondering when that question is asked is how accepting our patients and how accepting our physicians that this technology. And I think patients who-- we have a thought process that we discuss with clinicians, of course, upfront to ask for verbal consent before using such technology. And most patients are very willing and open to use this technology once they see that I'm not going to be dialed in on a computer screen when they're sitting over to my left, when I'm not scrolling and typing at the same time because I have confidence that that recording is being captured, that that note is being captured so that I can then just focus on the patient in front of me and make sure we're having a conversation. And I'm paying more attention to those nonverbal cues. And to what's going on with their emotions and things during the conversation. Most patients when they see that are very much accepting of this type of technology. But, again, we're going to see and we're going to learn from that. And I think we've just been very clear with frontline clinicians about the need for that consent and for that need to be clear about how this technology is going to help them in their patients. And I think cybersecurity and security with data has been the top priority for 3M. And again, we have decades of experience managing that. So. Great. Well. Thank you so much Dr. Bias. This was a really great presentation. (DESCRIPTION) Slide, Resources, a list, and a screenshot of a webinar titled Transforming the experience of health care. (SPEECH) We don't have any more questions that we need to get to at this time. So if you do have any more, please feel free to reach out to us. We do have some resources available in the resources section. The presentation was there as well as a certificate of attendance. And so if you do want to submit that certificate to an accredited association, excuse me, you can obtain CEUs that way. If you are interested in registering for our next webinar, which is on November 10, transitioning to practices to a risk based world, we have our another one of our leadership members that will be presenting on that. As well as, again, just getting some more information, you can certainly look at that resources section. (DESCRIPTION) Slide, Thanks for joining us. Want to connect? Travis Bias, D O, M P H, F A A F P.. Chief Medical Officer, Clinician Solutions. T-B-I-A-S at M M M dot com. (SPEECH) And if you do have any questions, please feel free to reach out to us. (DESCRIPTION) Slide, Thank you! (SPEECH) And, again, we always like to hear how we did. So if you do want to complete that survey we would certainly appreciate you letting us know how we did and we look forward to seeing you in November. And I do know that we are going to be hosting more of this Virtual Pulse series where it really is speaking directly to our clinicians and physicians and how important that specific group is. And we'll be having more webinars specifically geared towards them. And so, again, Dr. Bias, we greatly appreciate your time today. And we look forward to hosting this group on the next webinar. Thank you so much. Yeah. Thank you all for your time. And again, please feel free to be in touch. We've got a whole team of physician and nurse and pharmacists and would love to have a conversation about what your organization is trying to do. So thank you so much for joining. Thank you.