[Applause]

Moderator: Who’s next?

Speaker: Could I emcee?

Moderator: Sure.

Speaker: We are the group that was tasked with looking at comparing the various providers and then stating their outcomes, the first thing we get is trying to identify the various providers around. And obviously, the hospitals were already listed, the physician and their volumes or procedures are currently available.

And we kicked around one of the other things that we needed to potentially think about, obviously, ambulatory surgery centers are coming soon. The potential of looking at their pharmacies. We did it and I noticed that those cheating a little bit, looking ahead on your border for goals, or things we had to talk about were potentially rehab centers, out paper, psychiatric facilities in that. I think that’s something, obviously, we need to come back and to look at.

00:53

One thing we wanted to try with you those make it sound it was actually it’s something meaningful both for the customer but as well as all the consumers that are out there, the physicians, the hospitals and the rest of the potential folks that are going to be compared.

It is very difficult to get reliable data across 40, 50, 60, 70 different diagnoses that you feel is accurate. I think one of the areas where a lot of folks are wrestling with the accuracy of what’s out there right now, is of sort coming from the discharge data which is completely coding driven. And I think most of us realize that the coding can be very variable whether your coders are looking at trying to do it extremely accurately or they’re trying to optimize billing, or collecting whatever it may be.

So, we thought the better idea might be just to narrow it down, look at some, maybe the most common diagnoses, the most common procedures, and we’re starting with those that are currently out there. There are already eight procedures that are listed on the CMS website, the surgical care accruement programs got some various procedures that are out there.

So, we thought maybe the ideal thing is take 8 to 10 of the most common in-patient procedures, 8 to 10 of the most common in-patient diagnoses for the medical side of the house, do the same thing, break it out for pediatrics because obviously pediatrics is a different animal, of course, my advice is on the pediatrics side so I thought pediatrics is clearly a different animal, but then with the 8 to 10 common diagnoses.

And then one of the things we need to do is come up with a good risk stratification system so that when you always go—especially if you’re talking with physicians, their comment will be, “My patients were obviously sicker than his and that’s why my outcomes were maybe a little bit worse. My length of stay is longer. My costs are higher.”

So, we needed to look at the things that would truly affect that risk stratification, whether it’s going to be infectious, the volumes that you rank and looking at legs stake, infections and outcomes is what we're at, but looking at mortality, potentially someone in the emission labs are now with the art group is now looking at potentially having folks submit a lot of the laboratory data that’s coming in, sorting a lot of number of new clinical indicators to try and improve the risk stratification.

But we clearly need to come up with various stratifications so that when folks were saying, “Are you really comparing apples to apples, one facility to another, or from one physician to another physician?” With reasonable security, we can sit back and say, “Yeah, we really are. Your patients are just as sick as the individual is listed after me.

Try to come up with those both for the surgical side as well as the medical side that accounts to pediatric groups, we could be exactly the same thing. NAFRI already has a risk stratification tool that I think a lot of folks accept although it is relatively complicated formula and that’s something that we need to look at carefully.

Start with this 8 to 10, make sure that we’re comfortable with the data collection system being right, that we’re able to go out and educate the coders on how to code properly so that the appropriate information is getting in there, so that we can risk stratify preferably in New York state. When they went out and look at the cardiac surgical services, as they were listing those, they spent a lot of time educating the folks that were doing the data collection.

And maybe this is something that I think we still need to work through. Do we have the coders or the ones that are actually abstracting this data or do we need somebody else? I think the hospitals and clinics might push back a little bit if we say, “Let’s find a second group of people to abstract the data and submit it but it’s certainly something who we can consider”

But once we identify those 8 to 10 procedures are, the essential data points that we got to add if we were to generate the appropriate risk stratification, then we can go out and educate the folks appropriate on submitting the data, gathering data that we need so that we can make it useful and that everybody would always accept what’s out there.

Start with those 8 to 10 and then add, who knows, 5 to 8 each year over the next few years. I think one of the things we all agreed to, relatively agreed to, towards the end of the discussion was we probably don’t need to have 40, 50, 60 things that are out there listed. There are a lot of these that can end up being reasonable search against for a number of other procedures and number of other diagnoses that were in the hospital.

And so if we can look at those, take the most common ones, take a representative sample that might cover a whole group of various ERG or diagnoses. It might be all that we need but we can come up with the right groups of diagnoses, the right risk stratification, get that out there, or we might not get as much—hopefully we get as much pushback from the physicians of the hospital groups that we need to deal with. 

And then obviously, as we go forward, are there other groups that we need to add in such as the annual choice of surgery centers which were—will be coming in that.

So, on your board it says, ‘cheating earlier’, also looking at psychiatric facilities, rehab facilities and the others that the folks may be thinking. We opted not to look at comparing pharmacies. You already got the Florida Rx out there as far as comparing the costs but we weren’t sure if it was really something we needed to go chasing down to compare Walgreens to CVS to whomever. And that sort of where we are at the moment.

Male Participant 1: You had a reference to an assessment to whatever on that for you?

Speaker: Yeah, the National Association of Children’s Hospitals already has a risk stratification case and index tool that they use. I couldn’t give you the formula if I had to but it is one that a lot of folks accepted within the National Association of Children’s Hospitals Network that will adjust for that, that I think most of the institutions are fairly comfortable with.

Male Participant 2: When our governor in Tennessee gave a keynote at Hemps, he made three points: 1) He said built version 1.0 before you build version 6.0, we kind of meet those criteria; 2) Stay focused on just a couple of problems, don’t try to do everything, we kind of do that; 3) Worry about some real adoption in the Heartland that will show you got something right.

What I wonder about this is there is not very much evidence that people do make decisions based on this. There’s a lot of evidence that anytime somebody throws a metric and every faction starts throwing dirt and every other faction about a degree complexity that just creates a whole new industry, to aggregate and lump and classified, it does nothing other than create more complexity for more coders, for more staff.

Is it possible that if you take a few simple things, knees, heart surgery, things where there is concern and you just say, “This hospital—“ It’s said we have to hospitals because that’s where the idea is because it’s staying out where healthcare ought to be focused.

But this data, this many heart surgeries, they do this many people die. They have this length of stay. This is the average cause, period, and they say, “Of course, there’s complication and of course there’s complications. Go talk to them and let them worry about it."

In other words, is it necessary to standardize and risk adjust and stratify and all of that everything. It seems to me if you do that in a way that almost—if there almost like a magical law that said, “You can’t be punished by this data.”  Because again, as soon as there is a thrust that you can be punished on an output data and a quality process, the process stops. If there’s some way to protect it from punishment or adverse judgment, you can make this pretty simple, it seems to me, and get a start somewhere. Am I being too naïve there?

[Inaudible]

Speaker: They're shaking their heads. Yeah, yeah, yeah.

[Laughter]

Female Participant 1: First of all, I have a couple of questions. One, as Morrie pointed out, this is a really great, condensed, we need to do this in the next year. When we’re looking at five years out, when I saw—when our group looked at comparisons in providers by success with specific conditions or procedures, we look at different types of providers and we outline the necessity for additional data collection.

If we want to delve into recording on physician outcome outside of that hospital base, where outside of that inventory surgery center, what type of data collection would be required, how long would that take to implement from the AHCA perspective. What type of budget would we need?

So, my first part of my question is, did you give to any consideration beyond you want for comparison and providers success for the specific condition and procedure? And my second question is, we’re already hosting average length of stay, we’re already recording volume infections, mortality readmission. We’re redefining our readmission in what we're dealing with 3M or a beta site right now on how to calculate preventable readmissions in hospitals.

And we’re also risk adjusting that with the 3M risk adjustment system. Are you suggesting that NAFRI risk adjustment system should be applicable to the pediatric population and not to 3M?

Speaker: We just—we threw that out there as an example of one thing I know that certain groups are comfortable with. I didn’t get the sense from at least the discussion from within our group that everybody is completely comfortable with 3M.

I know that they’ve come up with revisions as they were working through it, working through with the folks at ARC. And I think the rest of several opportunity to really look at it, we may say, “Hey, that tool is perfectly fine.”

I think the feeling within the group right now is what we’re dealing with, currently, at the moment, we’re not comfortable with as being the right product that, you know, it’s one of those that we do just want to make sure that whatever risk adjustment, whatever case mix index tool is out there, the folks have had a chance to really look at it and feel that it is at least as accurate, as appropriate as it can be. And the modified 3M tools, they’re great

Male Participant 2: The question you brought up about making it to complex. We talk about and we putting, collecting these measures were put in together so that when it’s reported, it’s fairly simple for the consumers.

If somebody wants to know about a high-volume, low mortality place with reasonable costs having bypass surgery, we can find that very simply. Behind some rating or comparison would be the data that’s going to make up that evaluation. We had 10 different things that are on there and start losing the value of simplicity for the---

Speaker: And certainly, we would add a few things as the years went on. One of the things we didn’t get into was really trying to address the comparing the physician outside of the in-patient setting, the ability to pull that data out of a practice setting. We don’t even know scratch the surface on that.

And one of the things we also wrestled with right now, the way things are reported is I believe is by the discharging physician where if you’ve got a good hospital’s group, put care docs, whatever they’d be, well the whole group may care for the patient over the length of stay, and one guy happens to be the guy on the data that they issue a discharge. You’re kind of like, “Is he really going to—has been responsible for the last seven or eight days? Should those maybe better reported as a group?"

Female Participant 2: A couple of comments whether the risk adjust that and when we go back to what we talked about yesterday and maybe in our group, one of the things that we thought so important for this collaboration was trust.

And I think that the risk adjustment plan that normal consumer, it’s not going to matter anything to them than it’s risk adjusted that the providers can accept the measures, then that’s going to create some distrust of the website which would trickle down to the consumer population.

And so, I think that risk adjustment is key to get back in but I think an eye like focusing on just a few conditions and then grouping the other stuff to it with because right now, it’s all in different places or that’s the way it has been.

Just the way that this website was set up that I think that if we look back to what we heard at the personal health record. There might even a way of going to this higher level and maybe one of these conditions is heart surgery and but above that is heart disease and they can get the other information, manage their care on what are the signs of heart disease when it’s getting worse.

And then when it’s getting worse then get down into the issue of, “OK, I need to have open-heart surgery, where do I go?” And then you get all that information and then down to the level of rehab and eventually to the point of how do they stay healthy again and kind of that whole continuum.

And I know I’m not explaining it very well but it is, let’s focus on the few things you’re doing really well to help a consumer from point A to point B, which A is where they’re getting worse, B is where we get them back to normal functional status.

So, I think that this approach adding the other pieces and that’s not going to happen in a year but I think that that might be a separate, a vision of where we could go with some of these data.

Male Participant 2: It could be that your central premise is that, I think people had the experience that if you don’t do the risk adjustment and stuff, fundamentally the proprietors, aren’t going to trust each other or trust the data, right?

Female Participant 2: The providers will not trust the idea.

Speaker: OK. So, it’s for you so you can trust each other. So, can you tell me then that if you do it this way, you will trust each other?

Female Participant 2: I don’t know if it’s trusting each--it’s trusting that at least that has been accounted for.

Male Participant 2:  OK. I’m just looking at—it seems to me that the last 10 to 15 years, we get found an imossible problem here. And we keep going, "Yeah but, yeah but, yeah but..." And meanwhile, people are picking facilities because of the parking. They’re just making different choices and at some point, is there a way of changing the perception and say, “Yeah, you’re right, this is what we got.”

And I think you’re right. There’s got to be a way up—I fear that if the providers don’t get together and do something simple and imperfect and say, “This is simple and perfect but we back it now and it’s not being used for other things.” That the consumers are going to continue to just ignore everything you’re doing and go through it some other way. I’m just not sure how to ask you move, you said more than a year that worries me.

Female Participant 2: …more than a year to build up that thing. And I know, I think that there has been consensus around—not only necessary on the children side, but even with our workgroup when we met and the choice was administrative data. And administrative data is getting better in terms of having some quality, and of clinical information that you could pull out. It’s not perfect but we’re working that way.

But the risk adjusted system, at least for the purposes of when we started looking at their website, there was one that has been adopted and I don’t know that other than the children side that there’s much argument with the risk adjustment part.

[Inaudible]

Speaker: Just quick comments to close this up.

Male Participant 3: Why do chronicling information that I---

Speaker: OK. Sir?

Male Participant 3: The risk adjustment is very important not just for a variety of hospitals because of certian patients which is already happening. The managed care companies are already rewarding doctors and doctor’s patient’s data, economics and I remember one medical director when we’re discussing the fact that there should be a problem where health risk of patients will not be taken care of because they are going to be costly to take care of—

Male Participant 4: So, they could definitely use that against if it’s not you—

Male Participant 3: They’re going to use against the physician but more importantly its public consumer, they’re going to find that you’re really a sick person and you need a procedure. You’re not going to get it done if we don’t do this. We need those data.

Female Participant 2: Just so everybody knows whether it’s imperfect or not, tomorrow a press conference for CMS is listing 17 conditions for mortality rate by institutions. That will be on there for the risk adjustment which will be on your website. We’re doing the press conference about it tomorrow.

We think we’re ready to do it or not, whether or not they’re doing it. And if they’re doing a press conference in the morning.

Male Participant 4: But just a point of information. The 40 outpatient defined ARG risk adjustment system. Is it adopted by AHCA and is used as risk adjustment for anything.

This was co-developed and is the risk adjustment is a system of choice. It's more accurate—we worked with John Alden and Johnny Houser who is the Medical Director, who 50 years ago, worked with us to develop that site. So that’s part of the same system.

Moderator: I didn't realize that. Did you have a point?

Female Participant 3: I just wanted to clarify on the data collection process. We collect attending physician and operating physicians. So, it may not be what you’re assuming as the discharge physician.

Speaker: But an attending physician - if it’s a group, it’s a hospital’s group, critical care group, neonatal group, whatever it might be - I would bet during the majority of the hospitalization, the attending physician changes within that group. And that’s where a lot of folks do question, "You know gee, is this really reflective of Dr. X?" when Dr. X is really Dr. A, B, C, D, and X that has managed that physician through the entire hospitalization.

Female Participant 3: And we’re looking at ways to change that for a better work---

[Inaudible]

Moderator: Very quick.

Male Participant 5: This is Edward. Generally speaking, how many specialist are there in a hospital? And my concern is, are we really at this point in 10, 12, 15 yezrs that I’d be saying, “OK, that means gastroenterologists was checked out or the pods aren't going to be.” Is that going to be sensitive enough to cover all the different types of specialties to work in a hospital?

Speaker: It will probably not cover all because you have more groups like gastroenterologist that are migrating out into more than outpatient setting.

A lot of the practices is not in the inpatient side. But as you get ambulatory surgery centers, outpatient diagnostic centers, which is where gastroenterologists do practice, you’re going to at least pick up some of the data there but you’re right. For folks, allergists, most allergists do not practice in a hospital. We got to find—

Male Participant 5: You’ve been working in a hospital, are we going to be, at the end of the 10, 15 years there’s three different types of specialty doctors that perform that there is no grade one and getting to evaluate either between the specialty is saying and if that may—

Female Participant 3: ...especially you’re going to be dealing with 10, 15% of diagnoses. So probably, yes.

Speaker: Maybe if you with the majority of the ones that practice in the hospital setting, I think you can ultimately capture everybody. And there's one thing that very consciously and that look at it as it evolved on the first few years, making sure that you added in those ERGs maybe will adopt those groups that you have gotten an indicator on.

Moderator: OK. We need to roll. So, we're going to move on. Thank you.

[Applause]