Preparing for the Future: How one ACO is Using Analytics to Drive Clinical & Operational Excellence

(Dr. Scott Hines, MD, Co-Chief Clinical Transformation Officer)

Thank you for giving us the opportunity to talk about our story here today. We are going to go through a little bit of our journey over the past two years and our transformation here at Crystal Run as well as the changing landscape and national trends of ACOs. Then we are going to review the process of our data driven decision making, the progress we have had with our key challenges which we will highlight throughout the presentation, some early results and then we will be happy to answer some questions.

Before we dive into that we have another poll question we would like to ask. Again to give us an idea of the audience here so if you can answer which best describe which the organization you belong to.

Obviously we are a physician group and I am going to start by talking about Crystal Run Healthcare, who we are, where we have been and where we are going. Crystal Run Healthcare is a physician-owned multi-specialty group that was founded in 1996 by our original eight partners in one location in Orange County, New York. It is in the lower Hudson Valley, about an hour northwest of New York City. From that one location and eight original partners in 1996 we have grown to over 325 providers over 25 locations in Orange, Sullivan, and Rockland County and we have started to integrate a few small practices in Manhattan as well. We have pretty much everything you need under one roof to provide the outpatient care for patients. We have a joint venture with Surgical Center with our hospital in one of our main sites in Middle Town. We have multiple urgent cares, diagnostic imaging with everything from clean film to MRI to nuclear imaging. We have a sleep center, high complexity lab, our own pathology department, physical therapy, and occupational therapy. We were a pool of doctors of the electronic medical records. In 1999 we began NextGen® and we have remained on that platform ever since. We were the first Joint Commission accredited private practice in New York State in 2006. We were one of the one of the first Level 3 NCQA Patient Center Medial Homes in the state, initial certification 2009 and subsequent recognition in 2012. Our newest accomplishment is that we now have an approved health plan in the state of New York which was approved December 31, 2014. So that is a little about Crystal Run Healthcare, The Practice.

Now a little bit about Crystal Run Healthcare, the ACO. We are a single entity ACO meaning that the RACO comprises solely of Crystal Run Healthcare as a medical practice. We were one of the first 27 ACOs to participate in the Medicare shared savings program with an April 1, 2012 start date. We were one of the first 6 ACOs in the country to be accredited through NCQA. We have approximately 35,000 lives at risk with about a little over 10,000 of those in the MSSP program and another 25,000 commercial lives at risk across our various commercial contracts. One fact that is interesting that we will be highlighting later, is that approximately 82% of the primary care services that are attributed patients have are here at Crystal Run so we have the benefit of having very little leakage. I think the national average is somewhere around 60% – 65%, so the majority of our patients remain with Crystal Run and with our providers for their primary care services.

We also wanted to list here our mission statement because we feel it’s important for two things. First of all that folks realize that even though we are a for profit organization, we are very much a mission driven organization. Secondly, to show that our mission statement really aligns with the triple aim which has been our true north metrics for our transition into value. The mission at Crystal Run Healthcare is to improve the quality and availability of, and satisfaction with the health care services in the communities we serve. To accomplish this goal, the practice emphasizes both traditional medical excellence as well as responsiveness to consumer needs through service excellence and patient empowerment. And what is interesting and what we stress to our employees both at new hire orientation and also our employees who have been here for many years is that our mission statement really is peppered with mentions of the triple aim, quality, patient satisfaction, medical excellence, service excellence, and patient empowerment. Even though our mission statement was written in 1997, years before the triple A was invented. So when we have undergone this transformation one of the things we like to remind everybody, is that we are not really asking anybody to do anything different. We are really continuing to practice and continuing to follow the mission that we have had since our onset.

We have seen an explosion in the growth of ACOs nationally since the beginning of the NSSP Program in 2012. There are now a total of 626 public and private ACOs in the country and it’s estimated that ACOs now cover over 20 million lives. With the recent announcement from HHS that their goal is to have 30% of Medicare beneficiaries in some sort of alternative payment model by 2016 and 50% by 2018 we expect these numbers to continue to grow. Not surprisingly, there has been strong physician involvement in the growth of ACOs. More than half of ACOs are led by physicians with another third led jointly by physicians and hospitals. This should not be surprising as it has been commonly said that the most expensive and most powerful tool that exists in the health care system is the physician’s pen as the majority of the health care dollar is being spent by physicians. It’s not surprising that physicians are really grabbing the reigns of these alternative payment models and wanting to lead the charge and remain in control of the health care system. That’s very much been our mantra since day one that Crystal Run is physician run and physician owned. We feel strongly that physicians should remain in control of the health care system and we feel that this is a positive development as far as that is concerned.

This graph shows covered lives.

This map shows, based by hospital referral region, the percentage of lives that are in some sort of ACO contract. So you can see here that where we are in New York State, we are not penetrated as other regions in the county, but growing quite rapidly.

Not surprising that the changes, as well as with most changes, the majority of reimbursement is still set in fee for service, but we are beginning to start to see a shift towards value. As I mentioned before, HHS is really pushing to move this along.   In order to succeed in this future world where physicians are rewarded based on outcomes and not nearly based on transactions, it’s important that practices can continue to grow. In order to better manage their risks it is also important that we align physician reimbursement with the outcomes that we are looking for as far as improving quality and improving patient experience and lowering costs.

I am now going to hand it over to Dr. Spencer.

(Dr. Gregory Spencer, MD, Chief Medical Officer and Chief Medical Information Officer)

A lot is going on and the key challenges I think to execute on all these newer ways of physicians, anyway on the East coast, is to try to reduce clinical variation, continually improve and enhance our operational efficiency, which is always a good idea, and maybe even more so now as the margins are being squeezed, and try to position ourselves to both be able to accept risk-based contracts and also actually manage that risk effectively. We want to be able to give decision makers information so they can be efficient, effective, and always have access to the information that they need to act on, both clinically and from a business standpoint by pulling information from all of our various data sources that were very difficult to connect, that have been in the past, islands of information, into one place so we can connect and advance together. To have a single source of truth is really important with all of this, too, we have been trying to develop ways of having self-serving analytics. Instead of asking a question to a group of people who have to go back, do a report, give it back to you, and then have the person say, that was good but that wasn’t exactly what I was asking, and have this endlessly interpretive process, give more health analytics to the decision makers and clinicians to avoid delays in being able to make decisions.

So, these are four, there are multiple, but these are the most important focuses we have. As a practice we are growing and expanding. We currently are in Orange, Sullivan Counties and Rockland County in New York. As Scott said in the beginning, we have some offices now in Manhattan and we have added about 58 physicians last year and we are already at the 20 doctor level of adding providers this year. So we need to be able to support that, knowing what we need and where we need it. Population health management is no longer sufficient. It’s necessary and will always be necessary to take excellent care of the patient in front of you as an organization and provider, but it’s no longer sufficient. You also know that you are also responsible for the people that aren’t in front of you that come and seek care from you. So to have the tools to manage populations in addition to the person in front of you is very important and not something that we as medical practitioners and systems have done well. Along with that is risk-based contracting; to be able to manage populations with both their health risk and also the financial risk. The money and expenses that are incurred as a result of that health risk requires a whole new set of systems, data and people that know how to use it. Finally, the physician compensation and trying to align the compensation of the provider is not just with doing things, but being responsible for those under their care and the outcomes. So outcomes based, the whole value-based care, is to have the physician be rewarded for doing a good job with that in addition to taking care of the people in front of them.

Data is a prerequisite for many things and particularly value based care and our data driven future. So with that knowledge, we thought several years ago, although we had a long term data warehouse, we needed to get something much more scalable and manageable. Some key foundational applications that were developed as part of this was Health Catalyst and EDW and the four that you see here; Key Process Analysis, Cohort Builder, which I will go through some of these examples, Comorbidity Analyzer and Population Explorer. They are four of several applications or a suite of capabilities that are hooked and land on top of this data warehouse.

A poll question before we proceed further.


As you are responding to this, there have been quite a few questions asked about the slides being available. I would like to say, yes, after the webinar. We are recording the webinar and after the webinar we will be sending out to everybody an e-mail with a link to the recorded webinar as well as to these presentation slides and even answers to the poll questions

(Dr. Gregory Spencer)

That’s pretty good because that means that most of the people on the call are at least thinking about and starting to use claims data, which is good but I think not truly representative of the entire universe of the health systems out there, so this is good of the folks on this call.

The second poll question.


We also received a question here. There was no option for all of the above for either of the questions.

(Dr. Gregory Spencer)

We have been growing like crazy. Back in 1996 I was a second internist and ninth doctor. Now we have about 350 or so providers at this point so we have grown. We continue to grow rapidly and the Population Explorer on the EDW helps as we are growing now and the number of providers are much more numerous at this point. The EDW helps identify growth opportunities, visualize the characteristics of patients in a certain geographic area, how many patients do you have that live in a certain area, that live in a certain zip code, what are their characteristics, their demographic, their gender, the numbers of patients, what are the kinds of problems they have, what kind of revenue you can expect, what special fees are people seeing. Again, by zip code or by different slices. So, this really helps us to figure out what doctors we need to hire for a certain area or if we have a doctor there, why aren’t people going to them and potentially we need to have discussions about that. Also, for marketing, as we look to grow and beat the bushes or look for new product lines in certain areas.

This is a screen shot of one of the initial screens of Population Explorer so you can see demographics. What I would do with this is to be able to look in that patient information. I can enter a county, town, or collection of zip codes. Recently we had an idea how many patients live in zip codes closer to a certain office and what would their characteristics be.

So I enter a series of zip codes and see a view of those patients, what offices are they currently seeing and who is their payer. What kinds of problems do they currently have, are we seeing them mostly as out patients or in patients, what procedures are they having, and now you can even go down to this, what risk profiles do they have, do they have a Charleston Mayo wrist model based on chronic conditions? This is a lot of information about who is going where and what their make-up is. This is really the basis you need to help you make a decision to spend sometimes tens of millions of dollars to place a new office. This is the animation of that.

Another screen involves looking at the breakdown of the ENM codes, and this population is mostly, 80% – 90% office and the rest is hospital. It is the breakdown and the trend over time in that area, is it going up, is it going down? The current month is always the lowest as it is usually a partial month so you can select a location and you can see the trend in that area.

Risk contracting is what we are trying to do much more these days. Payers are reluctant to do this even if they are part of a national effort that has very deep and long experience in other markets. This is not something that is translated well, so this is an incremental process for us. This is something that we really need to have data so we can have informed discussions with payers where you don’t feel they have all the information. Certainly they have the claims information on the patient that they are talking about, but we know what we do, we also know other patients in certain geographies that aren’t with this payer and that can sometimes help us get a better idea as we’re thinking about going into a certain geography with a certain payer what they may not know. By having discussions where we can discuss comorbidities, risks demographics, etcetera on a detailed level, we can start to justify in the early stages where it’s shared savings, to have an additional PMPM paid to us to help support. We have care managers in our practice who work to manage high risk patients in addition to people that close the care gaps, etc. so to pay that forward. This is not some theoretical thing that they are paying us to do this, this is something that we have that we can show data for.

I alluded to this before, a comorbidity analyzer. This uses a 1:3 model so this is showing a Charleston Deo Model which is based on chronic conditions. There are two other models that you can choose. One is Alex Houser, which is for a hospital model as in a pediatric model, but again, looking at different areas, different offices, different regions, zip codes, we can get a very good idea of what’s in store for us there.

Super important for physicians, obviously, but also important to try to align the systems so providers be rewarded for doing the goals of the system which is to provide value-based care currently. The current fee for service model rewards on a per transaction basis so the more times you see somebody, the more things you do, the more you get paid. This is absolutely not the case in value-based care. Here is an example we had recently. There was a general physician on an oncology fellowship that initially joined us and did a lot of general surgery. But over time, as we have grown in size and scope and geography he is doing much more oncology so we base our physician compensation a good part of it currently, and we are transitioning over based on bench marks and then RVU model and the bench marks are based on what you do specialty wise. There are specialty bench marks for general surgery and there are specialty bench marks for oncology surgery and breast surgery. So we were able to look at him and see that over the last two years he encountered well over 80% as an oncology surgeon so we were able to adjust his benchmark with data and pay him appropriately.

So this is using that app called the Cohort Builder.   We selected this doctor and the time frame and as you can see the encounters he is seeing is a lot of breast stuff, by far the top ten of encounters he is seeing are breast cancer related. And the percent of the total E&M he was about 87% oncology breast surgeon.

Another app is the KPA tool. We use this as an adjunct for various reductions process of use in the practice. It basically looks for areas that are highly variable and costly and so we try to direct our efforts at those first, since a lot of things are worthy. Looking for variation and then by bringing in physicians and at the beginning and saying here’s the variable areas. What are the evidence based ways we that we can address this given your specialties? Suggestions and what are the ways we can help take care of these patients that have these problems that it brings a physician and enhances physician engagement?

Here’s a screen shot of the KPA Tool. Using a petro analysis at 85:15. This just shows on the initial screen the total RVUs and broken down by clinical care process.

Looking at it a little bit closer, this looks at the variation so this is the adjusted coition of the variation on the y axis and how much charges are, the fee schedule amount, so looking for things are in the right upper quadrant or as close to that as possible so things that are quite variable and very costly. Then looking at that and drilling down.

This is a drill down on into the women’s and newborn’s was one of those spheres and then you can then see each of these areas within that women and newborn area. If you were to click on one of the orange areas, inside the sphere there is the number of members of that and if you click into that you would get a breakdown of the positon and be able to drill from there. So it gives you a place to start. That will have a relatively large return on investment relative to other things which may be important, but at least early on you would like to recoup some resources for the system.

Revenue opportunities is another use of the EDW, the KPA tool I just demonstrated as well as the Cohort Builder we used recently. These are just some recent examples. Once you start having these tools in there, they were initially designed maybe for something else, but as you use them you get more tactile with them. There is a program, DSRIP, Delivery System Redesigned Improvement Program which involves Medicaid. We’re not an F2HC, we are private practice, and in order to be a full operatist you had to have 35% of your total revenues be Medicaid recipients and we don’t meet that threshold. However, because we are large, our denominator is large, but if you look at the number of people we have taken care of who had Medicaid and looked at that percentage of Medicaid that we have cared for it was very high. It was over 40% of the county so we were able to ask for and be granted a waiver to be included in a grant opportunity for this program. It is nice being able to find out where people live by county.   It’s not so easy to find people zip code wise by county that is not so easy because that zip code sometimes crosses counties and that could have been a major project for one of our BI guys in the past. This took a couple hours once we figured out what the question was that needed to be answered. Probably even less than an hour.

Another example is budgeting. We have used a warehouse to improve budgeting so that we could manage and predict productivity. This is actually a very sore point for us right now. With snow days we are getting hammered up here in the northeast. We looked at this last year and we looked at observed versus expected last year. The weather last year year produced about 16,000 less visits than we would have had on a certain day that were not decimated from snow. We were able to come up with budgets for people so they could figure out, by knowing what to budget that much for this year, how many patients do you have to about see extra a day during snow season, and it actually only turns out to be a couple, to offset whatever effects from the weather you might have. As it turns out, it is not a good year for us for weather. I’m glad we did something. We will see at the end of the season if it has helped or not. The cold is not as bad as the snow lately, but it does give us the ability to see where we are and looking forward based on the number of providers we have that are nixed and the number of visits they have what we could expect given a certain number of snow days.

We are expanding offices and looking to where we are going to put an office. Siting an office and, as I mentioned before, asking where your patients are currently. Looking at newer areas that are adjacent to our current areas to say, well if we could capture the business that is currently there, what would be the mix and the savings? An office for us now is in the $30-$50 million range. We have had to make this decision recently to figure out who is going to go there and where we should site it. Again, I used that example earlier. Another one, physician compensation where I talked about making sure everyone is going in the same direction. That analysis I that showed you of the surgeon took about five minutes to do, if that. It was mostly making sure that everyone was on the same page about what the question was and was not so much clicking the places that needed to be clicked.

Another example is of a Medicaid grant which is different than what we talked about before. Looking at the requirements of Medicaid is again more geographic. So you have a geographic dimension that is not something you normally think about as a business entity, not just where you give care, but where the patients are. Then in the hospital, what kind of patients you are seeing in the hospital in various types of care. Annual budgeting and productivity improvements we’ve again looked at more the seasonal, but also projecting forward with certain specialties. We’ve got a merger with a certain specialty and you feel good when you see your RBs bump, but then how much is it compared to what you expected it could be? Were those physicians seeing more or less on what you expected based on referral patterns, etc.?

Again, referral improved contracting. We’re able to bring impressive information to the table to be able to capture PMPM dollars, to get paid up front for doing things rather than waiting and then sharing that savings with the payer. It’s basically a win-win for them because we are doing a lot of the work they have normally done and we’re only getting 50% of whatever savings accrues from that.

We like to, again as I said with the various reduction, bring in the physicians to decrease the inappropriate clinical variation again. This is not to make everybody do the same thing. This is to discuss the things that are already worked out where there is evidence for the best way of doing things. Since we all cover for each other and are partners with each other, let’s all decide with all things being equal, this patient unless they have something weird going on, let’s all do things the same way. And obviously if there is something weird or different with that patient or if they’ve got an allergy or something you can’t do it that way. It turns out by reducing clinical variation that it does goes a long way into saving money because people are doing less of the stuff they don’t need to do and then sometimes you will have some people doing things they should have been doing at the beginning but that far outweighs in the other direction. The KPA tool helps us identify priority areas. We get a good return on investment for the time that we are putting in to do this and by bringing in the physicians early and identifying best practices. It really has the buy in and again getting people thinking about value based care in the same way. Ultimately, you want to have, and hope that less variation is going to lead to improved care and lower costs.

I am handing it back to Scott.

(Dr. Scott Hines)

So along the lines that goes on the graph up on what Dr. Spencer was talking about, as you can see there is a wide variety of ways that these different applications can be used and it’s even more critical now that we have this information at our fingertips now that we have our own health plan. As I mentioned at the beginning, we have our personal health plan in place. We currently have an Article 42 which is a PPO and EPO plan which is a value based network. So it’s a skinny network and a narrow network which again, for us is quite attractive, because as I mentioned at the beginning, the majority of our patients are quite picky to the fact that there is little out migration and there is relatively little leakage from these sort of health care providers.

One of the ways that when I had a look at these applications that are sitting in the data warehouse and how they can help us. With the health plan I see it two ways, I see it both as strategic as far as planning for growth of the health plan, but also in the day-to-day operations where we will really be focusing our efforts. Again, as I mentioned, we have a lot of patients that are quite sticky, but with these applications we can identify the disease processes or in the areas where we may be having a little bit of leakage or we may be having patients be seeking care at the inter care centers or that sort of thing. It helps us to identify some of areas in that practice that we need to grow on the strategic front. Also on the strategic front, using the populations to identify what are the most rapidly growing counties and areas that we need to potentially think of spreading the health plan to next. That can be quite useful. Then on the day-to-day operations, we use the KPA tool to identify the next diagnosis or topics that we should have in our quarterly variation reduction meeting where we have blocked time for our providers where every specialty does a variation reduction exercise once a quarter. Many times we leave that exercise saying, well what do we want to look at next time? Now we have this tool to be able to say, well what are the areas that have the most variation? Let’s look at those and we will be able to choose those. One other prime example. One of the things we are trying to decide as a practice is whether or not we want to participate in the newly announce oncology innovation model that Jim Uriah announced last week. One of the questions that we had was what is our current population that is receiving chemotherapy services? Then we try to determine what would be the sum of the forever payments and what would that mean as far as working capital to start this sort of program. Would we be able to grow the infrastructure we currently have to be able to allow us to do this successfully? We were literally able to do that as we were sitting around and talking about using these tools where previously it would have meant a few days into our BI team. They do a great job and they give us the right answer, but it would take a few days at least and usually a week or two to get that answer. We literally have that data right at our finger tips. So again I think that you can see throughout this presentation how these applications can be used both for strategic purposes as well for the day-to-day operations of the practice and also of our health plan. So that concludes the prepared remarks and prepared portion of the presentation. Dr. Spencer and I would be happy to take any questions that you may have.


It looks like we have a lot of great questions coming in. We would like to remind everyone if you do have questions or comments to be sure to enter them into the questions pane in your control panels. Let’s get started. First question


Do your ACOs include dental services in your model? In particular, are pediatric dental services covered under the Crystal Run health plan due to meet to the f requirements? Are dentists part of your Crystal Run plan network or are these services contracted out?

(Dr. Scott Hines)

No, we don’t have dental services within our practice so those would be services that would be contracted out.


Are the providers or practitioners that work with you employees of Crystal Run?

(Dr. Scott Hines)

We are a partnership model, so as I mentioned we have about 350 providers of which right around 300 are physicians. Of those physicians about 110-115 are partners of the practice so we are a partnership model. The remainder of the physicians are employed. Many of which are in their initial two years of employment and some of which have decided, for whatever reason, that the partnership tract was not the tract they wanted to take, but are very happy working here, plan on working here for a long time, and we’re quite happy with their quality and their patient satisfaction so they are long term employees of the practice.


These questions are somewhat related. Can you address the criteria that you look for as more practices and more providers want to join your ACO. How do you decide who to accept?

(Dr. Scott Hines)

We are a single entity ACO which means that our ACO is comprised only of our medical practice, so if you are part of Crystal Run Health Care, LLP then you are part of the ACO. Now when it comes to hiring and retaining positons at the practice, we make it very clear to folks when they interview that we are on the leading edge of health care reform and health care transformation and the innovation in our region. We really want to make sure they get that so it is a good fit for them and for us. As far as ongoing evaluations go, we have what we call a physician metric. It is our physician score card that looks at specialty specific quality measures, patient satisfaction scores, and cost of care metrics. On a quarterly basis our physicians are given a score card so they know what areas they are doing well in and what areas they should be focusing on.


Do you receive clinical data from any of the HIEs in your area or is your EDW just claims data.

(Dr. Scott Hines)

We are starting to receive some HIE information as well as direct information from these PCBs. Our data is not mostly claims data. Our data has some claims data from the shared savings as well as some other claims that are starting to come in more recently. Most of the data is clinical information from our various sources, but mostly from the EMR and the practice management system, but also from our general ledger, and from our HR system credentialing etc.


Are the screen shots showing from a third party software or from internally developed applications?

(Dr. Scott Hines)

The screen shots that we showed for this are off Health Catalyst. These are pretty much off the shelf apps from Health Catalyst.


Where do your bench marks come from? What sources are giving you the bench marks?

(Dr. Scott Hines)

Physician Pay is currently a blend of AMGA and MGMA per RVUs. The reason we did that was to increase the N because some of them. If you use these things the numbers become very small and yes, there is some cross pollination between the two, but it does give you a somewhat narrower, somewhat improved confidence that this is close to what things should be, but I think that is probably the benchmark you’re talking about.


We have quite a few questions around the same theme, explaining more about explaining more about your physician compensation models. How do you differentiate compensation between PCPs and specialists and just other requests to explain more about your compensation?

(Dr. Scott Hines)

So in the first quarter of 2014 we had a pretty radical change in our compensation model. All of our partners and long term employees are on the same compensation model whether they are specialists or primary care doctors and basically it is comprised of three inputs. The first input is a specialty. So a specialty weighted depends again on benchmarks that look at what is the relative weight of a specialty compared to what we used as our benchmark with Stanley practice without OB. If per example the benchmark suggests that a cardiologist makes 2 ½xs that of a FT without OB at the 50th percentile then what we call the RCI or Relative Compensation Index for cardiology is 2.5. So one of those three inputs is your specialty. The second input is your productivity. Currently productivity is still defined as work RVUs for both primary care and specialists but the plan in 2015 is to pilot at least primary care to change the definition of productivity from more RVUs to panel size.   That is one of the nice things about the model is you can plug and play what you define as productivity. The third metric is your value score. That is based on the physician metrics that I previously mentioned that looks at quality metrics, cost metrics and patient satisfaction metrics. It’s really those three inputs of specialty, productivity, and value that feed into the compensation formula and that is what determines comp. There are a few nuances as you go through there as far as non-work RVUs, stipends, if there is excess overhead those sorts of things. We actually have a whole other presentation just on the comp model, but the basics is that it is grounded in those three inputs, specialty, productivity, and value.


Do you urgent care centers as part of your group?

(Dr. Scott Hines)

We do. We currently have four urgent care centers that are located in our bigger buildings. They are quite busy and they have been a great tool when we move into a new area of attracting new patients.


We have 4-5 questions around your enterprise data warehouse. What are the major reasons you selected Health Catalyst, DDW.   Will you definitely be able to accommodate claims data? Basically a lot of questions are about the way in which your data warehouse is set up to handle data. Questions like that.

(Dr. Scott Hines)

That’s a great one. We have done our own data warehouse for many years and ended up choosing Heath Catalyst several years ago. The data warehouse, you can read or listen to some of their webinars about this late binding concept where basically you bring data in pretty much untransformed or minimally transformed, following certain rules to make sure it is clean and you know what it is and where it’s kept and then binding it, that is saying a certain number is a certain thing, a blood pressure or whatever, and it relatively lengthens the process instead of putting all the work in to make it perfect and that thing when you first bring it in so you get a lot quicker use of the data. We were able to get information up and running within 70 days or whatever from the time we started working on it. It wasn’t perfect, it wasn’t everything, but we were able to get useable data out of it very quickly. So again, these claims are an important part of it. Claims are brought in. We don’t have the same capability as say middlemen has and the insight as far as the benchmarking all the different slices and dices and different ways they are able to parce the data. We might be at some point, but right now it’s mostly the basic PMPMs, the different sites of care, utilization that we’re able to get in addition to costs and other things from the client’s information.


It sounds like you also addressed the next group of questions. There are several asking about how long you took to go from data to actual usable insight.

(Dr. Scott Hines)

Again, it’s always growing. It’s something that’s a living, breathing thing as you are adding new source to our systems and developing new marks and things for areas to use. From the time we started the phase, the green flag, until the time we started to get data was about a couple of months. It was about 70 days or so.


We have time for one more question. This question is how you evolve to a new HMO model. How do you anticipate your relationship to existing health plans to change?

(Dr. Scott Hines)

That’s interesting. We are already having those discussions. It is actually pretty collegial. They know where we are. They have had to deal with us. One of the reasons we got big and grew was to be able to have a seat at the table and that we could have a say in the care of our patients. It think it does allow us to speak on the same terms that they are talking and so they know we are an informed partner. In the new era this is not going to be about trying to shake down the system for the most money possible. All this stuff will eventually, probably in a relatively short term, will be widely available the cost of care. We will work with competition. Competition is good and competition is important and choices are important. We hope to be one of the choices in the market that we serve.


Thank you so much Dr. Spencer and Dr. Heinz.


Before we close this webinar, we have one last poll question.   Our webinars are meant to be educational about various aspects effecting our industry especially from a data warehousing and analytics perspective. We have had many requests however, about more information about what Health Catalyst does and what our products are. If you are interested in the Health Catalyst introductory demo, please take the time to respond to this last poll question.

Shortly after this webinar, you will receive an e-mail linked to the recording of the webinar, presentation slides, the poll questions and the results. Also, please look forward to the transcript notification we will send you once it is ready.

On behalf of Doctor Spencer and Doctor Hines and as well as the rest of us here at Health Catalyst, thank you for joining us today.

This webinar is now concluded.

[End Transcript]