Session 8: How One Pioneer ACO is Improving Healthcare Performance Through Analytics and Cultural Transformation [Kurt] Keep an eye out just before we get started. Everybody should see a "Lessons Learned" sheet in front of them. We encourage you to use that to take notes and write down one thing; a question I'll ask at the end of the presentation. Hopefully everybody is in session eight. You've got all your apps connected to the session eight because you’re going to use the application to ask the presenters questions. We'd encourage you to make sure you're on there. As you'll see, there's already a polling question that's up. We'd ask you to take a minute, look at that polling question, and respond. Couple of other things to note: you have the ability to submit a question through the app. We'd encourage you to do that. For those who have submitted questions, we'd ask others to look at those questions and vote on them. Remember: all these things earn points. You can press applause at any time; we’d encourage that; especially if you hear something that's caught your attention and you think is interesting. Remember: you score points when you participate in the poll, choose applause, or submit a question (especially one that's voted on by your peers). If anyone is having a problem with their app, we've got technology folks in the back (they're raising their hands if anybody wants to turn around). We encourage you do one of two things: head on back to see if they can help or wave your hand and we'll have them come to you. Looks like we may have one person in front that has a technology question for our group in the back. Okay. We've got somebody coming to the front. Anybody else have a question? Oh, very good. Good afternoon, ladies and gentlemen. I'd like to extend a very warm welcome to all of you. Thank you for taking some time today to participate with us. I am very, very pleased to introduce Mark Hohulin, Senior Vice-‐President of Healthcare Analytics at OSF, and Roopa Foulger, Executive Director of Data Delivery at OSF who'll be speaking on the topic, "How One Pioneer ACO is Improving Healthcare Performance Through Analytics and Cultural Transformation." Before we get started, I want to acknowledge the analysts in the back of the room. Susan, Chris, and John, if you want to raise your hand; these are the folks that will help guide us this afternoon. With that, let's go ahead and get started. I'd like to see if we can see the results of our initial poll session. Let me turn that over to the analysts in the back. [Male Speaker 1] Thank you, Kurt. The question was: does your healthcare organization currently have an effective system for using data consistently? Let's go ahead and look at the results. Mark,
Roopa; looks like you have your work cut out for you. 62% said no. Very good. Without further ado, let me turn things over to Mark.
How One Pioneer ACO is Improving Healthcare Performance Through Analytics and Cultural Transformation [03:53] [Mark Hohulin] Can everybody hear me okay? Sounds all right? Okay, thank you. Good afternoon. Before presenting and talking with you about one of our cultural transformations and journey around data and analytics, I'm co-‐presenting with Roopa Foulger who's been a big part of our team in developing the technical side of our data warehouse and a lot of the analytical applications that we have across OSF.
About the Organization [04:21] Just some background on our organization for those who are not familiar, we're located in Peoria, Illinois. We're an 11-‐hospital integrated healthcare system. We have one hospital left in the UP of Michigan, but most all of them are all in Illinois. We have 224 locations, over 700 physicians, and 16,000 or more employees across our marketplace. To give you a better understanding of our size, we're about 7.2 billion in growth charges as an organization in Illinois, about 2.4 billion in net revenue, and we serve about 700,000 patients (unique individuals) across our marketplace. The population base itself is much larger than that. We've been a pioneer ACO for four years (currently in the fourth year). From a performance perspective, the first two years we were really within that 1% corridor for those familiar with that methodology and process. It was really no-‐shared savings or no-‐payment-‐back in CMS. The third year, our performance improved quite a bit and we were able to earn about $4.9 million in [Inaudible][05:30] Some of what we'll talk about today relates to our transformation-‐how the ACO effort and work came about, and how we supported that. Another frame of reference: we have about 180,000 covered lives in a value-‐based arrangement today; about 30% of our revenue and 25% of the patients we serve. We are also part of the Healthcare Transformation Task Force now, which stated the goal of having 75% of our business-‐or lives-‐as part of a value-‐based arrangement by 2020. Big
movement is expected in our marketplace, which is what we have to be prepared for from an analytical perspective.
Pursuing Clinical and Operational Excellence [06:23]
To give you an idea of how analytics began, we look at how we can pursue clinical and operational excellence. One thing we've been fortunate with is that our leadership (CEO) has prioritized that we have to develop an effective and price data warehouse and analytical capability to derive insights from the data that are key for our future success. With that in mind, we knew that we had to do analysis in search of clinical business insights needed to derive improvements. Again, the leaders prioritized the analytics as a component of that strategic plan. Part of our development of the plan involved strategic goals that were set by our leadership group. Again, as a frame of reference, the 11 hospitals that used to operate fairly independently as their own operating unit separate from the system perspective; over the last four to five years, we operate more as a central organization. Basically, the term that we use now is One OSF. Everything we do is based upon an entire system. Each operating unit focuses on the healthcare system in total. That's been a big change and a big part of our cultural change that we've been able to take advantage of as well. That helps drive everyone in the organization focused on total organizational improvement on clinical, financial, patient experience, and employee experience across our organization. That's been key. We’ve been going through lots of transformations, one of those being analytics. Leadership put our analytics group at the front of the transformations happening across the organization. We brought about five different, independent groups across to construct one analytics division. Prior to that, we were called Decisions Support, Clinical Physician Support, IT Informatics, and Ambulatory Informatics. We've always been different pockets. What we did is we brought those five to six groups together and basically branded ourselves as the Healthcare Analytic. That was two years ago. That's part of our journey.
The Pain Points [08:32] The thing that we experienced prior to doing some of this is we had some pain points the organization realized that we had. We had difficulty finding an effective system for using data consistently. We had multiple locations and silos of data. We previously had some unsuccessful EDW implementations. These efforts were driven without a business purpose or strategic intent. A little bit from a technical or IT perspective, this is something we need to have just because it feels like we should do it, but it didn't have a lot of ownership or accountability from leadership to make any difference. Ultimately, as the cultural change occurred across OSF, leadership recognized we needed to have clear, aligned data and a warehouse and analytics to support that. That's been a big change. Data overload; many of you have the same issues. Data is everywhere. Our organization has EPIC. We are highly integrated epic organization from primary care to specialty groups to hospitals to home care. Being a pioneer ACO we also have now five years of claims data for 40,000 beneficiaries; we have other payors’ claim data as well. We have a tremendous amount of data, which we knew we had to be able to organize and put into a data warehouse.
We have the inability to determine if we were measuring the right thing. There was a huge effort by our Chief Strategy Officer and planning group. Our leadership put together a focus around what our organization should focus on, the strategic goals, and measures that need to be looked at regularly to understand how it's performing. Roopa will cover some of that in a little bit. We had difficulty creating that transparency and clinical excellence. Again, it was in pockets. It wasn't focused. Our cultural transformation has changed that quite a bit where we've made great improvements. Again, that's all helped us from a healthcare standpoint; from an analytics data warehousing perspective; that's helped us invest resources into those tools and the technology.
Our Approach and Results [10:27] From leadership down we wanted to deliver superior clinical outcomes for all of our patients. We need to improve the patient experience across OSF. Again, going back to that One OSF theme, patients go into any of our locations, any of our doctors' offices, any of our hospitals they need to have the same exact experience. That's another focus that we've had. Enhance the affordability and sustainability of services. Provide core transparency of health through performing not only financially but also clinically and try to bring that whole cost of ownership-‐cost of care-‐into the forefront for leadership to understand and then really drive that cultural shift to embrace becoming a data-‐empowered system. As you'll see in a little bit, when we look at the executive dashboard, the leaders look at this regularly and everything's not cascading down to different service areas; different clinical focuses for the same purpose. We're all looking at measures. We're all looking at how we can be more accountable to improve performance.
Our Approach [11:30] If you look at our approach, OSF needed to drive the cultural shift throughout the organization to embrace becoming a data-‐empowered organization. The question was, “Are we making decisions based on data?” Often, we were not. We started bringing trying in a lot of the strategic goals and measures to increase accountability.
What we did with tools and transparency to engage leadership was implement the EDW with the Health Catalyst’s help. The late-‐binding EDW aggregated clinical, claims, financial, and other data to create a consistent view of the ACO's data. Again, this has been a journey. This has happened over four plus years and we're still growing. One thing that we've done to that process-‐and again, it includes the strategic goals and purpose-‐is determine what do we need to integrate into the warehouse and make it an iterative process-‐not a big bang-‐build this entire warehouse, have everything developed, and then do something with it. We wanted to determine what we needed to accomplish and achieve: the quick wins that we could develop as a team in analytics was probably the most important thing for us to do: getting quick wins, getting leadership to see the results so that investments and support continue to grow. An example of that ACO data is we had to bring in the claims data. Shortly after we started implementing the warehouse, we started bringing in millions of claims records. We build analytical tools on top of that. When we brought in the claims data we were able to merge it with our EMR information so we could see the patients, who their doctors were, understand the risks, and be able to transfer that information to clinic managers and clinicians. We also had the ability to look at clinical immigration: where those 40,000 people going for care, how much of it is provided at OSF, and how much of it is going to other non-‐OSF locations. That was one big piece we were able to provide to leadership. In one year, we saw that $70 million worth of care was going on outside of OSF. That visibility helped us say, "If you just got 10% movement (to OSF), that's $7.2 million, 20% movement, $14 million." Even though we're in a value-‐based arrangement, we're still getting paid at these service levels. Also, we know that we cannot manage care if patients go to another system or hospital. We felt it was very important to bring those patients back to OSF, and improve that care so they don't have to come back in to the hospital. They can be engaged with care managers and primary care. At the same time, when the care is being done, we're able to make sure we're getting that dollar amount; not someone else. That was a big, quick win and one of those things that I have said, from an analytical perspective, we have th ability to see that information and then take action. That was very helpful for leadership. The other thing that we had to do was form and support improvement teams. Again, with leadership, form a strategic goal and it was imperative that it came down clinically. We needed to focus on specific items like heart failure, supportive care, or sepsis. With that kind of charge in clinical agenda, we're able to put together a team and then start identifying data sources that we're going to need, the appropriate data format, and the analytics we have to put behind that information. That’s been the approach. As I said, it's
been a considerable process. We've grown over the last four years quite a bit. We know we have more to do, but we've had a lot of successes in those regards. One of the items that we'll be sharing shortly will be the Executive Dashboard. Before we do that, I'll turn it over to Roopa. I'm going to ask that you take a look at this polling question and use your device again. Analysts, maybe you can help me out here.
Poll Question #2 [15:10] On a scale of one to five, how would you rate your healthcare organization's ability to leverage data to make clinical and operational improvements? You can go ahead and put those into your apps. We'll go ahead and give you a few moments to respond to that. Okay. Let's go ahead and show the results. Looks like 40% of you said it was moderately effective, 32% somewhat effective, 19% very effective, 7% extremely effective, and 3% not at all. Back to you, Roopa. [Roopa Foulger] Not bad. Now that we had a runway built, a data warehouse built, it was time for us to fly out and start deploying solutions. Data warehousing, when it’s implemented, is usually
abstract. Most people cannot relate to data. Collecting the information should be the next immediate step that any organization should take. That’s what we did. As Mark said, as we were defining, we were looking at our organization’s strategy and we had a new Chief Strategy Officer. The first thing she did was look at our whole road map. She said, “We need to have something that everybody could align to and look at.” We needed a dashboard that leadership and other staff can look at to understand our performance on our strategies. As part of doing that, we made sure the right people were at the table and there were countless discussions on what the right metrics were. Having the business owners at the table was critical because you want them engaged early on. You want them to believe in the data you’re going to put out and that’s what happened. It helped our adoption.
Executive Dashboard [17:06] The next step was not overwhelm them with too much information as well. Finding that sweet spot and deploying a tool that’s easy to use that would give them the information that they need. It doesn’t show in the screenshot, but we have red, yellow, and green indicators so they can look at it and know how they are performing on those metrics. They are doable to a certain level because it is for senior leadership. They could look at the different operating units. They could look at all our goals. They could also look at trends to see if the trending tool is the right direction. This dashboard is now used in most meetings, including cabinet and board meetings. They pull it up and discuss and ask questions. That’s engagement and adoption. It’s difficult to drive initially, but once you get there, people just use that to change culture. That’s what happened at OSF. There were a couple of other projects that we did as well. Mark mentioned the clinical agenda and aligning the deliverables that we do to the goals that derived from our strategy; from our road map. We’ve deployed a couple of solutions, including Care Transitions. The goal of the Care Transitions project was to reduce re-‐admission in our acute care facilities. When there was a counselor the committee put together, it had people from different care teams. It had nursing, physicians, and pharmacists with a clearly defined goal. When IT was at the table, we were at the table as well.
Improvement Teams [18:41] When it was time to define the processes we needed to make sure patients coming to our facility received high quality of care, we were involved so we could give feedback, asking “is it measurable or not measurable from a process standpoint?” Being quick in deploying solutions, we used a jive methodology. We made sure that we had a goal when we engaged with the business and made sure we didn’t try to boil the ocean. We always tried to re-‐align with the goal in terms of the measures tying back into our purpose. One of the key takeaways from lessons learned, from our experiences in working with these teams, has been be prepared if they rip it out. You want them to do that. You want them to validate. You want them to trust the data. Sometimes it could be like you want to pull your hair off, but it’s good. It means that they’re engaging. Once they trust it, they’re going to use it to drive action within the organization. Besides the Executive Dashboard you saw a sample of-‐a screenshot of-‐the other dashboard that we put out is not just strategic-‐they’re operational too. We focus on dashboards that are more operational that look at not just outcomes that the senior leadership reviews, but also how the different care teams are performing. It can go down to a staff level. That’s giving action back to the team because now they can use that for teaching/educational purposes. Getting the trust I would say is the most important piece of this.
Value Realized [20:20] It’s been four years since we deployed the health catalyst model of data warehouse and it’s been a value proposition; the fact that we could have a data warehouse up and running in 90 days and runway in 30. Don’t try to boil the ocean. Don’t try to be perfect or you will not engage effectively. We’ve got good cost avoidance dollars: nine to 12 million over three years. That’s underestimating. It’s probably more. Just faster access to data and using that to drive clinical operation excellence. When we engage, it includes multidisciplinary teams at the table. That has helped shift and change culture.
Performance Improvement and Cost Avoidance [21:12] Some facts relating to our effectiveness: by deploying solutions that don’t just show your performance but have some deliverable; in some cases it goes down into the details. Thereby, it reduces this churning of single-‐dimensional reports. I want this report. I want that report. We’ve tried to eliminate that when we deploy a solution. Also, there were a number of staff within the different entities basically doing hunting and gathering. We’ve been able to reduce that and re-‐focus their efforts in analysis by taking on data-‐provisioning activities. One example from Care Transitions is by giving them information at the detail level where they could actually act upon it, we’ve reduced 3% in defects within the organization and thereby have 9.4 million possible cost avoidance. That was a good success story for us. We’ve deployed dashboards for productivity which they used mainly gathering Excel files using some automated solutions to capture the data; this helped us in terms of driving change in culture within the organization.
Heart Failure Program [22:35] There are some more slides with examples of projects that we did. Here’s one: the Heart Failure Program. This was the first initiative within the organization that went across different operating units: home care, medical group, and acute care. The CV service line engaged with us and by working with us and using data, they have been able to get 8% reduction in unspecified coding, 15% improvement of patient education best practices, and also increased the five-‐day follow-‐up on appointments. That’s just an example of improvements that we’ve had.
Cardiovascular Physicians’ Dashboard [23:11] Another example is a CV Physician Dashboard. The goal of this was to have physicians take ownership in terms of quality and safety. When we deployed this dashboard, every physician in the CV service see each other’s performance. They’re not used to this kind of a transparency. Enabling the transparency within seven months, you could see every metric 3% increase. The metrics were more balanced. There was patient experience. There was adherence to medication-‐some quality safety-‐quickly available appointments access. You could see that all these metrics went up just because they could see each other’s data. We also provided information down at the patient level just to gain trust; to have them look at what was missing in terms of care for the patients that we serve.
Palliative Care Program [24:06] Another good example is the Palliative Care Program. There’s another breakout session. That’s tomorrow. We’ll go into more detail. Again, here we engaged with and had the business leadership at the table. It was multidisciplinary team with presence from different units: home care, medical group, IT, analytics, and brainstorming through what are the processes and what do we need to provide so we can serve our patients. We had some good successes come out of this as well.
Data Empowered Culture [24:51] Date-‐empowered culture: to get there, the foundation must include transparency and collaboration. Thereby, accountability will automatically come because once you have the CEO and senior leadership aligned, it’s going to drive change across the organization. Making data transparent, engaging, listening to them when they have questions, and being willing to modify the look and feel of a dashboard or even change the operational definition of metrics if it doesn’t align to the goal: I think those are key things in terms of enabling that culture. The fact that we are always aligned to our strategic goals, whether it’s quality metrics or financial, has made us successful in the solutions that we’ve deployed. Stakeholders throughout the organization see the value and clinical impact of the quality improvement.
Future Plans [25:58] Our plan is to continue to expand the analytics solutions that we have. Two years back, there was an initiative within the organization to have portfolio management aligned to our strategic goals. We are aligning ourselves to portfolio management, which is what are the projects that need to happen within the organization and which of those projects need measurements? We make sure our resources-‐our scarce resources are effectively aligned. We also want to continue to expand our infrastructure capabilities as we grow in volume and data. As people have more access to data, there’s more demand. Being able to keep up to that demand whether we are ramping up our data warehouse capacity through better storage or data warehouse appliance. We’re also looking at Hadoop as infrastructure looking at data quality to work. As they get engaged, they ask more questions; data quality questions and data profiling. To enable our staff on that we are going to expand our technical capabilities. We will tie the feedback-‐all these insights from the data warehouse-‐back into the EMR. We’re already doing it but we’ll be expanding on that and doing more.
Lessons Learned [27:25] [Mark Hohulin] Just some lessons learned for us. Again, it’s been a journey. Engaging executive leadership across the organization is key. I think that’s helped us be successful within analytics; having that executive leadership support and engagement. Assembling project teams with the right technical capability has helped us tremendously. Roopa’s skills and her team have been a tremendous asset to our organization, helping us move further down the road. She’s engaged the clinicians, leadership-‐you name it-‐to help understand the importance of the technical aspects. Having system-‐wide engagement that’s not just a siloed IT project is very important. The strategic intent-‐the objectives of our organization-‐drives what we’re developing analytically. That’s much better than saying from a technology side, “Here’s what we can do for you.” It’s working with the business and coming down. Setting up a business plan and measurement up front is something we’ve had success around. We’ve also had some areas for improvement, but we’ve had strong business, clinical, owners, and leadership. We’ve had tremendous success working through the process so that patient care experiences have improved. Where we have not, we get to a point where the data is available, but we have no engagement, no action that follows. That’s a key thing, it’s having a business plan and measurement set upfront.
Achieved rapid implementation of the EDW with actionable insights to demonstrate value. Again, get the small win. Focus on what the system objectives are, the goals, get some of those quick wins that build support, success for the organization, and then again, more investment within analytics. Those are supposed to be some of the lessons learned. I would say one thing I’d touch back on that Roopa ended with-‐one thing that we really need to do more of-‐is improve the clinical decisions forth by time data and we’re having a warehouse back into the EMR. We’ve started doing it as she referenced. That’s the key for the future in my opinion: getting the clinicians’ information in the EMR because that’s where they live and make it more actionable because they don’t want to navigate a bunch of different applications. They just don’t. That’s our culture anyway within our organization. We know that we have to do a better job at trying to get everything back into the EMR and that’s part of our future as well.
Questions and Answers [29:29] With that, we’ll take any questions that you may have. I think we finished a little bit early which is fine. [Kurt] Test, test. There we go. Mark and Roopa, thank you so much for a great presentation. I’d ask the audience to give a great round of applause. Before we jump in to Q and A right away, I just wanted to go back to our analysts in the back to see if they had any insights that they could share based on your responses and based on the information shared. [Female Speaker 1] Apollo, our applause analyst has noted that the pain points mentioned resonated with the audience and also the comments on performance improvement and cost information seemed to resonate. Looking at our session and summit polls, this is pretty much just going to confirm the obvious and that’s that those of you who rated the strength and quality and culture of quality improvement in your organization as being high also rate the health care organizations’ ability to leverage data as quite high. That’s not a truly amazing insight, but it’s confirmed by the data.
Thank you. Back to you, Kurt. [Kurt] Thank you. I want to thank everybody. We’ve got 26 very, very good questions and unfortunately we’re not going to be able to get to all of them, but we’ll get to them based on the way you voted. What’s interesting is the very first question is how are you dealing with the lag and claims data? [Mark Hohulin] It’s always a challenge. That’s one of the first things we experienced with the pioneer claims data that came in. Are there any other pioneering fields in the room or rather, ACO arrangements where you’re getting claims? Our first experience with that, with CMS probably the first six months, we got files from them, the data layout, file format change almost every single month. Our team really was pulling its hair out every time it figured out the first time how to do it, the next file came in and it had to be changed again. Lot of learning curves around that. Ultimately, they improved. The fact of the matter is it is still lagging. That’s where we can take the golden nuggets that we get from the data in a lagging format and try to make it more transparent and actionable for our leadership. I think linking it with our EMR data has helped us tremendously to get past that, helps us provide the risk information on the patients, the high utilizers on in-‐patient or the frequent visitors in the ED. We’re able to do a lot more with our internal data by just capitalizing on some of the signals from the claims data. I don’t know that’s the best answer or not, but that’s one thing that we were able to do with the lag. [Roopa Foulger] I can add more to it. Yes, the claims is lagging and you cannot drive too many clinical improvements through that because you need it to be actionable right at the clinician’s hands. One of the Chief Clinical Officers had the care teams in the frontlines focus on data that’s already available in the EMR because they feed the data back into our EMR readers stratified being know who are those high-‐risk patients and they make sure that the clinical gaps are meant as BPAs that fire the EMR. This care gap report, they know that an ACO is coming and the visit puts the gold out for the nursing staff to look at so that they can look at the gaps and make sure that those are addressed. Being more proactive has helped us. The claims data we use it more for financial and utilization metrics and UPM BM and things like that. [Mark Hohulin] I have this to add. For those that are involved with it, even having the financial understanding of how a performing is not very clear with the lagging claims that and getting the report so far behind from CMS has been very challenging. To say today we know how
we’re doing in planning a report precisely, I can’t tell you because again, the lag comes, the methodology tells me I must calculate is lagging. We just have to deal with it. I’m thinking of the earlier presentation about the big audacious goal, being part of the Healthcare Transformation Task Force is a big audacious goal our system has because that’s a huge change for us that we know we have to get ready for, but it’s 2020. 75% of our business in that value base is kind of scary actually based upon how the claims come in, how we get information. It’s not very fluid today. We only hope it’s going to improve with our analytics. [Kurt] Beautiful. Thank you. Next question is what data visualization tools do you use? [Roopa Foulger] Excel. Everybody loves Excel. It’s from Excel to, we have ClickView. We also have an SAP BusinessObjects Shop. It’s mainly these three. Of course there’s SAAS and SPSSR for data modeling. Any of our dashboards they mostly deploy on ClickView. We also have Excel CS which is now Crystal Dashboard which are studying the regular usage of that. [Mark Hohulin] We kind of go off, but we have a BI Portfolio tool for different purposes and audiences. These tools are picked based on which one would be best for that audience and a lot of Excel still, lot of BusinessObjects for permissions, and then more and more ClickView. [Roopa Foulger] We have HPM. That’s another tool. [Kurt] Very good. Next question is what specific findings from analytics contributed the most to your cost savings? [Mark Hohulin] A lot of our clinical efforts are really providing some of the insights on how to improve the costs. We do see re-‐admissions. We do see lengths of stay that’s occurred. Again, it’s a little harder understanding those cost improvements, but we know from working on accountable care, improving the population of health aspect, that’s what we have to do. We can see within our results and our cost structure that that’s made a big improvement. We’ve had a three-‐ or four-‐year financial cost goal and we’re moving $235 million from our system. We’re on a strong pace to do that in year two now. Again, analytics is part of that. I’m not going to say we take entire credit for that, but it’s provided having the ability to shine the light, enable the information for our leaders and permissions. It’s helped us remove a lot of cost. [Kurt] Very good. Thank you. The next question is how are you aggregating data from disparate EMR systems?
[Roopa Foulger] The senior leadership made a conscious effort to making sure that most of our hospitals and other entities are on Epic. A couple of our hospitals are on MEDITECH and CPSI. We bring in more the coded data and financial data; not the clinical data, because that gives the immediate bang for the buck. They’re all going to come on Epic sooner or later within the next two years so we are not going to focus our efforts. Most of these other hospitals that are on other EMRs are smaller. [Mark Hohulin] Last five or four years, about five hospital acquisitions or affiliations have become…they’re really critical hospitals, CPSI or MEDITECH. We’ve taken a little more of a conservative approach of saying, “Let’s wait till they get put on Epic,” because if they’re on Epic, all close into the warehouse. We haven’t gone after…they have a source data. [Kurt] Thank you. In your care transformation journey, did you restructure efficient physician compensation models and if so, what are the key metrics there? [Mark Hohulin] I’m not going to be able to answer that the best as far as specific metrics, but I know within our work around accountable care we’re actually doing a large-‐scale transformation effort right now with our primary care group. The physician compensation models are being framed. Again, how aggressively and what specific metrics? I can’t answer that. I potentially could get some of that information for you after the fact if you give me your business cards, but I don’t know the details on that one. We know it’s important. It’s going to drive accountability and ownership in this work. That’s definitely something that we’re doing. [Kurt] Thank you. Next question is how are you showing the result and value to frontline staff and providers? [Roopa Foulger] There are a couple of ways. One is they Leverage Epic and they’re pushing data into Epic. They’re able to get to it. The other is Epic Hand Interface to other BI tools. We try to interface. We tried to reduce the amount of clicks outside of Epic that the frontline caregivers have to go and that’s definitely helped. In some cases, we have automated PDFs that are shipped to physicians. It’s mostly physicians because some of them do print outs and that’s helped as well. [Kurt] Very good. Next question is what have been the main challenges in shifting culture to be more data empowered? [Mark Hohulin]
Prioritization. We have a huge demand for analytics. It’s been a challenge just what do we resource and prioritize, if anything. Thankfully, I guess, using our strategic goals and the priorities that have been set as well as the clinical agenda, that’s been tremendous for us because then we have something to look at. When requests come in, we can first go to our strategic road map-‐our clinical agenda-‐and say they are either of those. If it’s not, we’re able to say, “You know what? That’s a good idea, but we can’t do that right now. We’re focused on the position itself.” That’s one way we’ve been able to do it. [Kurt] Next question is how do you plan to send EDW data to the EMR and how will the EMR actually consume and display that data? [Roopa Foulger] As I said, we haven’t gone full-‐fledge on it. We’ve just started on that. It’s been a year and a half. Because we’re on Epic platform, we leverage some of the interfaces that they have, whether it’s through their Bridges Interface or loading some of the claims data that we get from the Pioneer ACO to fill the missing gaps in the EMR. We go through that. The other thing that we just finished up is through Datalink. We would prefer Epic coming in and pulling in the data rather than us pushing it. There are nightly processes where the Epic Datalink feature comes in and pulls the data from the data warehouse and pushes that into cache. [Mark Hohulin] The most recent one she references is we have a predictive re-‐admission model that we’ve developed in our warehouse that puts predictive risk on every patient. Just as of last two weeks, we’ve now been through Datalink pushing that back into Epic. It’s visual within every screen on the patients for the clinicians to see. That’s a start. We think we’ll be doing that same thing for a more value-‐added element that we can bring from the warehouse. [Kurt] Awesome. Unfortunately, we’re down to time where we’ve got just one more question that we can share with the group. This one is what are the trade-‐offs you experienced with the centralized analytics approach? [Roopa Foulger] It’s probably the operating units where they’ve had an analyst and they could ask the analyst to do anything that they wanted. It being centralized, we align ourselves to what our goals are and some of the smaller things that they couldn’t get done sooner is probably not happening. [Mark Hohulin] I would say we’re five out, 85 to 90% centralized. We still have some of our larger operating units that analysts have capability to do some of the work. They did not come within our transformation into our group primarily because they do multiple things. They wear
multiple hats. We aren’t just analytics so we didn’t feel we could reach in and say, “You’re part of analytics solely.” There are still so many visuals out there, but that’s probably some of the change points we have. We’re still working through that, understanding and then having facilities rely on a central resource; not individual resources at that site.
One More Thing [46:07] [Kurt] Very good. Before everyone takes off, I will just ask for you to fill out the “Ask one more thing” and then if you would please give one more great round of applause to Mark and Roopa.
Thank You [46:26]