[This Transcript is Unedited]
Department of Health and Human Services
National Committee on Vital and Health Statistics
Quality Workgroup
February 23, 2006
HHH Building, Room 705A
200 Independence Avenue, NW
Washington, D.C.
Proceedings by:
CASET Associates, Ltd.
10201 Lee Highway, Suite 180
Fairfax, Virginia 22030
(703) 352-0091
P R O C E E D I N G S
MR. HUNGATE: The technical question for the corner over here. Since everybody is sitting in the same place they were this morning, do we have to go clear back through the introductions again, or can we proceed forward and just add those like Susan Kanaan who was not here for the first discussion? I think that is all we need to add. I would like to spend the time we normally would have spent going around talking and welcome Carol McCall as Co-Chair. As you may have sensed from my remarks from the closing piece of the last session; quality agenda and continuous improvement are near and dear to my heart.
I think that there is a clear need for the Quality Workgroup to follow through on the visioning process that we started in a retreat, and we have been working on. We have gotten a sense that what we are trying to do is very highly linked to every other piece of NCVHS and that we need to have what our work plan evolves to be much better tied to the work plans of the other groups so that we have a good sense of the things we can expect from each other and achieve in that regard. This is a Good morning, everyone. If people could take their seats, we will begin this long and productive day.
I am at this point a lame duck in my time at NCVHS and feel that my personal contribution has to be whatever knowledge and viewpoint that I have got, and transferring that to other who will see the fruits of success. So it is especially in that context that I welcome Carol as Co-Chair. And in that context as well—wishing to be a more vocal and free participant than simply a leader of discussion—I have asked Carol to take over as the primary meeting manager. And with that, Carol.
MS. MCCALL: Thank you very much. I want to thank Bob for acknowledging what for me is going to be a very fun role. So thank you very much for the very kind comments. I also want to say have fun with your new degrees of freedom and I hope to lead us in a discussion today that I hope will be very fruitful not only for today but for some upcoming work that we talked about earlier; whether it is the retreat or the work of the Quality Workgroup, or work with some of the other subcommittees.
To start us off today I wanted to make a couple of opening comments just to kind of lay a context and a framework for what I hope will be a very lively discussion today. Some of those opening comments are that recently we had—probably six or seven months ago now—asked ourselves as a Quality Workgroup a question: what is the focus of the Quality Workgroup? We had just finished a set of recommendations to the Secretary and we talked about it today. We are in a time of transformation, not just transition, with respect to health information technology. So asking the questions, we began some systematic discussions and systematic surveys. Our goal in doing this was to make sure that we could make recommendations that were specific, that were relevant, that were timely to the atmosphere and to the need, that they built on prior work of the NCVHS; but that they also considered other work both within NCVHS that was taking place and other broader efforts of HHS—research organizations and the industry in general. And to that end we endeavored to exploit work we had done, to extend previous work where possible, or identify places where it was no longer applicable or maybe even wrong in its initial recommendation, avoid reinventing things that had already been invented, be it a concept or a vision, and identify any shift in the relative priorities that had taken place.
In doing so, we began with the two documents 21st Century Vision for Health Statistics and Information for Health and looked at their concepts. And then we asked ourselves today whether and to what extent the ideas therein were either complete, on target and implemented or being implemented, inaccurate or wrong, or incomplete and in need of revision. So what you see in front of you, you should have had the opportunity to look at two documents. One is a summary that is of the minutes of a hearing that was conducted on November 18th. The second is a shorter document that is Widening the Vision for Quality that is essentially a synthesis and a summary of that hearing along with what we have seen in prior work and what is happening in the industry. I would just say that we are on the eve of creating what is an agenda for the Quality Workgroup, which is partially an agenda for the NCVHS for the next two to three years. So what you have in front of you is not an end; it is a beginning. Before we begin, we felt it necessary to pause to check our path, to engage our colleagues, to augment or to change our goals, and to make sure they are consistent so that when we create our plan it is relevant.
So, the timing is excellent given our conversations today regarding our retreat and a refocusing and re-centering of what we do not only for ourselves, but in the concept of the other work. What you also have in front of you is an agenda that walks us through a discussion. That is really what this is intended to be, a discussion around the document itself. Hopefully people have had a chance to consume both of these documents: Widening the Vision for Quality and some of the greater depth and detail and feedback that are in the minutes from the hearing on the 18th; as well as a series of discussions that are not meant to constrain but only meant to guide the types of things that we really want to get on the table and get feedback from people. With that, I would open it to the group with the first being a discussion and a review of the document Widening the Vision for Quality.
MR. HUNGATE: Did everyone find that Widening the Vision. Did everyone find that?
MR. BLAIR: The first thing I think of when I think of the intersection between the Quality Workgroup and at least the Standards and Security Subcommittee, I think the first area of leverage is the secondary use of clinical data. Is that one of the items showing linkage?
MS. MCCALL: It may be. Justine has just reminded me that it may be vital to walk through the high level bullet points, if that would be valuable, that are contained in the Widening the Vision for Quality document. So I think that there are a couple of sections here in the high level bullet points. The first section is what the two NCVHS vision documents suggest about quality. The first is those created a very broad vision for health and healthcare. It included multiple dimensions, including personal health as well as providers and population health. It stressed a very broad view of a system actually even going beyond Aziz(?). There is a concept that we got down to the point of calling Figure 2. It is a figure that was in the 21st Century Vision for Health Statistics document that talked about more than EHRs. More than the context of disease it went into functional status. It talked about environmental context. It talked about even broader context yet; saying that to understand and create a vision for health, you need to have all of that context. That really is the first bullet point.
In the second bullet point they describe the role for health information and health statistics as poorly articulated yet crucial to realizing this vision, so that it was not merely about understanding EHRs and that type of things. It was to understand the health of individuals and populations you had to include its influences. You had to create fundamental knowledge. You actually had to guide the assessment, development, and evaluation of health policy and a research agenda; and be in a position to inform and evaluate the impact of programs and interventions as well as personal health decisions. Those were the elements of influencing and the role of health information and health statistics.
These papers also cautioned against thinking of this infrastructure as purely a technical undertaking and identified the initial requirements as being of equal, if not greater, importance. So they went to great pains to define what the called the health statistics enterprise as including this ability to aggregate data. So I believe that is what you are talking about from a variety of sources and to infuse other information from outside the health information delivery system in particular. They talked about the significance of activities and processes, about governance and some of the things that we heard our speakers talk about today that create new knowledge including the means to continually prioritize the things we want to do, to transfer knowledge that is created, and to integrate it perhaps through decision support into the very care processes that we have. This enterprise also includes the infrastructure that supports it—the values, the practice, the relationships, the resources, the laws and standards, as well as the systems and applications. They also portray the NHII as an important foundational element in realizing the vision for health and healthcare.
The themes from the November hearing are summarized also in this paper. The committee’s mandate and history suggest a very broad and long range view that human health should be the focus; that a policy and an analytic level is missing from the quality discussion; that we should avoid the record paradigm and instead use a system terminology in taking a very broad view of that system in a network metaphor; that the role of a trusted and competent aggregator is a pivotal issue that is unresolved; and that there is a very good question of whether or not quality measures actually scale up to the population level. We talk about quality measures, but they may not sum up to be that which we seek to know.
There was another key theme that quality measurement needs to be patient-centered and relevant to consumers. Information from patients is needed to assess outcomes; perhaps via personal records. Patients and families may be the most appropriate or virtual aggregator and integrator of health and data on themselves. There was also a question posited how can the country move from the current institutional environment to a person-centered environment of information and measurement that centers analysis on the person as the central decision maker and informational source. Again, this is all under this section of quality measurement being patient-centered.
A third major theme is that the quality agenda lacks a business case. There was interesting discussion around this, that there is not a business case around quality. Today it is not listed as a high priority. Only 12 percent of purchasers actually list quality as a priority today. It is not a buying value; it is in fact invisible. Disparities are a major factor. Even though according to the Rand reports most Americans are getting only half of the quality of care measures that are viewable today, care is being delivered in that way. It is losing out to other priorities in healthcare. And it lacks a business case.
The fourth priority and theme that came from these hearings is that the EHR adoption trends and quality initiatives are disconnected. The transformative trend towards EHR adoption is underway. This was acknowledged as building momentum and should increase the urgency because of the disconnection of moving forward with the quality measurement strategies. Currently the performance trends are disconnected from the EHRs and the NHIN, contributing to a growing fragmentation. Some of the people at the hearings described the HIT system as being at a critical juncture. The tracks are being laid for interoperability but not for quality. To some the key issue for quality is not EHR adoption but the lack of will to measure important aspects of quality along with a lack of strategy and an overall picture of where we want to be.
A fifth theme that came through from the hearing is titled Standards, Interoperability, and Connectivity. It recognizes that they are central requirements to EHR design specifications and that they need, and also acknowledge that they need, standard terminology and vocabulary including the action words. We had an interesting discussion around verbs. A sub-bullet is that quality measures also need to be standardized. We acknowledge that AQA is making progress on this front. There was also discussion of a vision; that there could be mechanisms in the market for open transfer of discoveries; that there are moves in that direction: efforts to standardize the application, programming interface. Stan Huff was absolutely delightful in describing the things that are possible within a system that is a true system.
I think that an important theme — for me anyway, that came out is that we talk a lot about interoperability. As important as interoperability is, it is necessary but not sufficient to achieve what we seek to know around quality and population health. For example, interoperability does not assure person-centricity, nor does it accomplish aggregation which is in and of itself its own undertaking that is very large. It does not automatically generate population level data, much less information on the influences on a population’s health. It is a distinct undertaking with its own technical and standards creation issues and a necessary process for understanding quality in health.
The sixth theme is entitled Some Ways to Improve Quality measurement That Are Already Understood. Some important points here: most current work on quality is too granular and fails to capture the more complex aspects of health and health care. The existing consensus process does not produce sufficient quality measures. Digitizing current approaches could only serve to institutionalize current silos. There was discussion that quality measures should capture functional status – the live continuum, the disease continuum, handoffs between settings, the interface between a patient’s family and the system, and the relationship between financing and care delivery. The process measures should not be the chief mechanism of quality assessment; other things should be included. Demographic data are needed to assess and reduce health disparities. Information on outcomes and health experience must be included and must come from patients. Some of the best measures are sometimes a by-product of the automated workflow at the point of care. There were questions posed regarding what consumers want and what measures capture it.
A seventh theme was that quality depends on multiple factors embedded in learning systems being the theme. Some of the sub-bullets here are on Intermountain(?) Health Care, and also the Veterans Administration both showed practices on how the use of EHRs and EHR systems and institutional processes work to promote quality at the end of care because they have a means to discuss what they see and mechanisms to decide, change, and learn with quality as a focus of institutional accountability and support. It was talk that quality improvement involves a suite of applications and electronic health records are just one tool in that suite. Quality cannot be captured from the EHR alone, even if it is interoperable. An interface with decision support is a key element. We also talked about disease management models and the VA and Intermountain Health Care cases also demonstrate population level uses of automated clinical data to promote quality.
Then the eighth bullet point from the panel were specific recommendations for us to consider for the Quality Workgroup. They are that the government’s focus, and thus that of NCVHS, is to be assessing and improving the health of the population based on national goals for health improvement. One was a visualization exercise to go not where the electronic potential is today, but to go where it can be; the Wayne Gretzky approach to skate where the puck is going to be. We need to address the capacity of NHIN and EHRs to improve population health. We can contribute to the missing analytic and policy dimension. We can figure out how to implement person-centricity with respect to quality. We can connect individual and population perspectives. We can work through alternate scenarios for the near future based on technologies; the type of scenarios-based planning that uses different paradigms and scenarios to imagine a future. We can also use that to lay out assumptions about the availability of data for public disclosure and identify short-term use cases and use them to drive analysis of the needed IT structure and information. We can use it to look at quality applications that can be inserted into the EHR – that are not built in today, but part of that suite – to help us move beyond that record paradigm. The record is not key. We must examine the requisites of a data aggregator. To drive a national discussion of best practice for using clinical data and improve quality. To get the word out to the public and make it visible to using NHIN for quality. To recommend that the government fund research, pilots, and demonstrations addressing the questions raised that we heard.
That is the second section of Widening the Vision, which is a summary of the hearings that we held. I am going to ask Justine to summarize the third section of this paper, which was really the progress and the challenges in all the emerging quality issues. I am telling you, there is a lot of stuff going on out there.
DR. CARR: This version that we have distributed is not the final version. I am just looking at my notes here. It is not substantively different, but my parts were edited so maybe we can get that out at some point. I apologize if it seems cryptic in the first draft mode. I think that one of the things that struck us was when we had our hearings in May and September of ’04 it was early on in terms of our understanding of what were the various quality initiatives. Within a year the proliferation of initiatives, activity, definitions, and so on was extraordinary. I think that in our early hearings we thought that there was territory that needed to be defined. As we saw today, IOM and others have been very actively trying to coordinate the same body of information we were.
As the third part of this report what we wanted to do was acknowledge the areas that we were interested in particular, but also to demonstrate that there was a proliferation of activist going on. As recently as June of ’05, the Quality Workgroup’s fact-finding pointed to the absence of attention to the interface between NHII and quality measurement. That had been an area of concern of ours as we moved towards EHR adoption. Although EHR can improve the quality of care in many ways, little attention has been paid to leveraging the EHR functionality to meet the proliferation of quality measurements required by payers and regulators. Currently, some of this data is derived from administrative data; other data must be extracted through chart review. The workgroup remains concerned about these disconnects, but heartened but the proliferation of activity in the public and private organizations and joint collaboratives. So we applaud the fact that some of the changes needed in the health care arena are beginning to happen.
Just to summarize some of these initiatives, we struggled in our first hearing about what is quality. Perhaps we have not answered it yet but certainly the IOM and Crossing the Quality Chasm as we heard today proposed this health system for the 21st century which is safe, efficient, patient-centered, timely, and equitable. The second question is how is quality currently measured. And there we have the initiatives by HRQ, CMS, JACHO, and NQF that have identified evidence-based practices that are quantifiable and capturable either by administrative data sets or medical extraction. Secondly there are data registries, particularly on cardiac programs, that have been adopted or developed to provide a more detailed source of data. Some are proprietary and others are being developed in response to initiatives from CMS – particularly the defibrillator registry and the carotid stent registry.
The challenges to quality measurement include the uniformity of data elements, data aggregation, and public reporting. As we heard today, the IOM and also the Ambulatory Care Quality Alliance have recognized these gaps and have called for a national quality coordinator role. The third question: how is quality achieved. Two areas we pointed to are: pay-for-performance incentives that are focused attention on target areas, leap frog collaboratives of health care purchasers. CMS has linked hospital payment to submission of quality metrics. Private payers have linked incentive pay for physicians and hospitals to the achievement of benchmark performance levels. The challenge to pay-for-performance as we saw in the NQF report from March 2005 is that there are gaps in knowledge and that there is potential for unintended consequences of inappropriately designed reward structures. We are seeing that every day. I will not bore you with Massachusetts’ latest misadventure in that area, but it was painful.
Third, there is a need for a single set of guiding principles. In addition to pay-for-performance there is the electronic health record which facilitates timely availability of information as well as real-time decision support. We have heard today about the role of a national coordinator, AHIC. I think we saw the challenges articulated by Carolyn Clancy in the November ’05 AHIC presentation. One is that there is limited collaboration between quality assessment and health IT communities at present. Second, there is a need for clarification of the role electronic health records can play in moving toward population health and management. And third there is a need for a minimum data set of formatted data elements and field to produce an agreed upon set of measures. Finally, in current state of health care quality we have two reports: National Health Care Quality Report and the Healthcare Disparities Report. In the quality report, we know that healthcare quality continues to improve at a modest pace across most measures of quality. Second, healthcare quality improvement is variable with notable areas of high performance. Third, healthcare quality is improving but more remains to be done to achieve optimal quality. Their fourth point is that sustained rates of quality improvement are possible. On the National Healthcare Disparities Report disparity still exists. Some are diminishing. Opportunities for improvement remain, but our information about disparities is improving. I will turn the summary back to Carol.
MS. MCCALL: I will just summarize fairly quickly and hopefully in a pithy way. Basically the review of our visions and the work to date hearings highlighted the following messages. The first is that there is obvious and clear momentum toward EHR adoption and that creates some opportunities for efficient approaches to quality measurement and improvement. What it also does is highlights a disconnect between various initiatives be they pay-for-performance and EHR adoption trends or what it actually means to measure quality. It gives a sense of urgency to the task of optimizing the environment in which this tool is going to be used because it is a tool. The second is that EHRs and EHR systems – even an interoperable one – are necessary but not sufficient as a tool for health quality improvement. In other words, implementing an interoperable EHR will not give us everything we need for that purpose. I will paraphrase John Halunka(?) who has an extraordinary system in Massachusetts. He has a wonderful system and he is moving transactions and things are getting better on that perspective, but he still needs someone to guide him on what to do on a quality agenda.
The discussion about the committee’s most important contribution may lie in providing the context or the container for the work that is now underway on quality. Two areas that stood out as important and unaddressed in health information policy were understanding what it would take to measure, assess, and improve quality at a population health level and understanding what it would mean to have a person-centric approach and how that links to population health. I would add that subsumed within these and not in the document is that it does require an analytic and policy framework in order to provide both of those bullet points and to actually describe what it means to go from an individual level to a population level.
We found that strengthening the dimensions of person-centricity and population health which are critical for a comprehensive governmental approach to quality may be the most appropriate focus for the workgroup’s unique vision and voice and the committee’s unique vision and voice. Clarification of these dimensions could aid in the shift from provider and procedure-centered quality measurement and an increased attention to population health attributes that are not captured today in our clinical measures. With that there are a series of questions. Before we begin to talk, let me read them quickly and then I want to get people’s reactions.
The quick questions for discussion were what short and long term objectives does this analysis and summary suggest for the workgroup? I would also extend that to NCVHS and the full committee broadly. Other knowledge that you see may be needed to move ahead in these areas and recommendations on who could provide it? How the NCVHS subcommittees could collaborate in addressing what could be a broad quality agenda? And any short-term recommendations to the Secretary that you think could merge from these findings outlined above? It is not meant to be in a final form, but there could be some derivative, specific recommendations before we go into a broader work agenda.
DR. TANG: I think that is an excellent distillation of the issues about quality and I think it is nice and timely to receive the IOM contribution of this latest report in December to say here is a lot of the issues and it has to do with standardizing and making consistent these quality measures. It may have to be a top-down process or something faster than just percolate and hope they coalesce. I think that one of your key conclusions is that there is a total disconnect, perhaps even in NCVHS, in the quality issues/quality measurements and the EHR and NHII or NHIN. Maybe that is a perfect contribution to the whole field in conjunction with what the IOM is recommending. One of your pleas is that the NHII workgroup works with the quality workgroup to try to nail these two efforts back together.
One of the issues I did ask and I do not think came up in this recent report was that they recommend some starter sets of quality measures – let us say the AQA for ambulatory care – to my knowledge the denominator is still based on claims data or administrative data. Would it not be nice if we could try to – in some timely way because the pay-for-performance is going to come out in June – at least get the issue raised about how do we either incent or not disincent the same perverse incentives again, use of clinical data in clinical quality measurements. Right now we will continue to incent people to use administrative data because that is where the money is, literally. It actually punishes, or does not credit at the least, the early adopters who have gone out, spent their own money without pay-for-performance to put in clinical systems only to turn around and try to dig information out of them. That is something that is timely and urgent because of the report that is about to come out and because of all of the initiatives that are ongoing; one of the biggest of which is in California. That seems like a very worthy goal is right consistent with your findings and possibly something we could address. We would have to do it very quickly in order for it to have an impact.
MS. MCCALL: Jeffrey and then Mark
MR. BLAIR: First of all, that is the best summary of quality issues that I have ever heard. I was not being cute or facetious when I said applause. That was just outstanding. You asked for short-term and long-term recommendations. Here is the one that you might put on the list for consideration for short-term. I read through the doctor’s office quality program which the QIOs are implementing with CMS funding and it appeared as if many of the missing linkages that you were referring to are there. Conceptually that program is winding up saying let us have EHRs. The mandate on it is that it will use SNOMED-LOINK and RX-Norm there as the standards. There is the clinically specific terminology that you wind up getting it through to performance measurement; that you wind up having workflow improvement; and they wind up having the folks that will wind up assisting so that you can have the workflow and clinical process improvements. Then it winds up saying we are going to have the data to measure improvements so it leads to continuous quality improvement. So on a short-term basis it appears as if there is a program which addresses so many of the links in small practices like the additional funding, the additional support. You can find out the areas where they need the additional help.
That appears to me to be a short-term piece. The long-term one, or longer term, looks to me like a five, seven, or maybe even ten year lifespan. It looks like it has got a great foundation. It is just that the implementation over time would take that long. If it does, then during its lifetime it appears as if another emerging area of quality is emerging. It seems to go by different names. This last issue of Health Affairs had an article which also ties to some other articles which have been released recently. It was called Information-Based Health Care. Are you familiar with it? It winds up saying let us take all of the things that we just mentioned and covered. Let us add proteomics and genomics and wind up taking that where we could being to not only look at personalized care but even wind up looking at it from the standpoint of using genomic information to educate a patient on where they have proclivities for disease in the future. You wind up having wellness programs that are beyond the ones that we have now in terms of quality measure. That is kind of a different dimension emerging. I understand the Mayo Foundation has been working with IBM in this area. The United Kingdom has started to work on this area as well. That might be an additional complement. That is a possible candidate for the Wayne Gretzky of where are we taking this, what is the next step.
DR. COHN: I guess I have some thoughts as well as a number of questions. I almost find myself looking at these suggested steps and I thought the IOM report conversation was a very good program for this. I see that we include both quality measurement and quality improvement in this document, but it seems to not be differentiated very well. There needs to be a slight difference. The first bullet of the summary document sings to me and I was thinking that I do not know exactly what is coming out of IOM in terms of its results. Once they have something out, I guess we could have another letter that says something else about it; it is certainly timely. I guess I was thinking about such concrete things as the CMSG codes and pay-for-performance recommendations. Then the question gets to me of how would you even do those, or map those into electronic health records.
MS. MCCALL: I understand the G codes. You said you expected –
DR. COHN: Did we not talk about the pay-for-performance IOM report that came out in June? I think you referenced that. You were sort of commenting that we should do something fast to talk about this one to meet that time frame.
DR. TANG: Their last committee meeting is in April. I do not know that they have a process like we have.
DR. COHN: I do not think we would be able to send them a letter between now and April. Obviously we are all moving on the same target.
MS. MCCALL: So are you saying there could be a specific short-term action where we would step into this G code process?
DR. SCANLON: I just understood Paul’s point somewhat different which was the issue of how do we take pay-for-performance and use it either to reward people who have adopted EHR powers or to encourage the adoption of EHR. I think that the G codes are somewhere in between what we might call a claim and an extract from an electronic record. In the discussions on pay-for-performance, have been in some respects let us make it hard enough that you would not want to do it without an electronic record.
MS. MCCALL: Or, and I understand what you are talking about, that there is a minimum set of data and if you can actually get a minimum from clinical data from an EHR you are actually even eligible for more. So if there is a set of 100 things that you want – only 50 of which you can get from administrative data – the other 50 of which you can only get from EHR, then by golly in order for you to get full payment guess what you need? You need an EHR. Is it along that line?
DR. TANG: Bill’s look is close to what you said, but you can always do it on paper, it just costs a little more. So the incentive then is to do a more efficient process to get measures we really want. We are not asking you to do something we do not truly want. But, we do not want to pay for claims.
DR. COHN: I guess I was flipping the conversation around because I do not actually know in an EHR how you could even get the G code measures. You probably could, but it is very difficult.
DR. CARR: Just looking at the thing that Jeff just mentioned, the doc thing, they have taken a lot like aspirin. They have gone through and systematically identified what is available through the electronic health record. In fact, in our hospital we are doing that. The G code thing is controversial.
DR. COHN: I was just bringing that up. If indeed the QIO solves that problem then the other question is about EHRs and EHR systems not being sufficient; necessary but not sufficient. I think that is something we need to remind the rest of the country about. EHRs are not a panacea for all that is going on. The final question I had is just a question on the third bullet. We were talking about how we assess quality at the population health level. There is a document a couple of years called Healthy People 2010 which I had forgotten about until I started reading this. I was actually just wondering is it questioned how you drive all those things using EHRs and what do you get when more is needed. Is that some sort of surrogate for how you would measure things at a population health level or is that too simplistic? Is that something we should be looking at?
MR. HUNGATE: We did have Richard Klein come to talk about the Healthy People program in our retreat specifically to try to get an understanding of how that tool fit with the rest of the tools. I do not think we know enough yet. We were told at that time that some of the content that goes into those measures is internally inconsistent; that two pieces will actually be telling you two different things but they are both used in the result. And so I personally have followed that program for 15 years and like the model that they are portraying as a way of beginning to measure population health in a scalable way so that other people can adopt the same tool and measure the health of the population they are dealing with.
DR. CARR: Let me just read you the ten leading health indicators: physical activity, overweight and obesity, tobacco use, substance abuse, responsible sexual behavior, mental health, injury and violence, environmental quality, and immunizations and access to healthcare. So it kind of crosses a spectrum.
PARTICIPANT: I think that is pretty much all the measures that are in Healthy People come from federal statistical surveys.
MS. MCCALL: Russell and then Don
DR. LOCALIO: Carol, I have several points. I will try to be brief. When I looked over this summary of Widening the Vision I was marking in the margins some of the issues that repeatedly come up that I never see other people bring up. I do a lot of work with registry data and registries are supposed to be ways that people are collecting all of this information to follow outcomes as a way to improve quality. Every register I have ever seen is full of missing data. And I am working with people and saying you cannot just disregard it, you have got to use imputation. You have got to drop important factors. It is always a colossal disappointment to work with registry data where you think it is supposed to be and should be great. Second point, this is something that is not explicit here and should be: gaming. How much effort goes to gaming the system. Example, in emergency room care one of the indicators is administration of antibiotics to people with community-acquired pneumonia within four hours. So I found out a couple of weeks ago that at least one institution is gaming that by applying a topical antibiotic to anybody who has pneumonia. They are applying some goo to the skin, and then they check that as administration of antibiotics within four hours. Clearly even to a dumb person like me that is laughably inappropriate. So I would say that when I first read your Wayne Gretzky quote I thought it meant ice skate to where the buck would be.
The next item is sometimes for many types of care you do not see measures on whether something should have been undertaken at all – inappropriate or unnecessary care. It is a counterfactual issue. You can never observe what would have happened if you had left the patient alone. This can be potentially a big problem, especially if the endpoint is a long ways away. Example, today’s Washington Post has a headline that says Medicare to Cover Obesity Surgery. Patients suffering from weight-related health problems will now have three forms of the popular stomach shrinking procedure covered in the federal health program. About a year ago, I was asked to review a review of obesity surgery and my comment was this is the type of surgery that takes 10 to 20 years to figure out if it is working or not. Nobody had done any studies of that length. This is a procedure that is done an awful lot and we never see what is down the pike on that. That is just an example of some of the things.
And then my final example has to do with I did not see anything as traditional as medical education. I would think that if we are talking about quality, this is something that I did not see. I will give you another example. I have to teach tomorrow so I will not be here and my students are mostly physicians. I have about 40 physicians and they have all been through medical school and through residency. When they come to my class one of the first things I do is say I am giving you an example on sensitivity, specificity, and likelihood ratio – in the clinical, decision making sense not in the statistical sense – and I get a lot of blank faces. These are very bright people who had very good education, yet they have never heard of a likelihood ratio. Something was missing somewhere down the pike. These are people that have done their boards and all that stuff. I think there is a wonderful layout of many of the issues but I think when you are measuring quality this is a complex area. I have just named a few of the things that I thought maybe needed additional attention.
MS. MCCALL: Before we go to Don I would like an opportunity just to expand a little bit on the points that Russell brought up. One is that this was not intended to be an exhaustive review of all the current issues. The purpose of this document is to find the place where the quality workgroup as a part of NCVHS will focus and set an agenda for the short-term and the long-term. That is its purpose. So you will find things that are missing, Russell. They are all over the place. The other thing I would say is one specifically that you called out, education is a great point. We went back and looked. NCVHS has made specific recommendations in 2001 which are deep recommendations on what it means to actually educate people, including physicians, on things like likelihood ratios. And we opined elegantly on the need for that. In calling and harkening back to those papers we are essentially imputing all of that into this work. There are a number of recommendations that are similar to what the IOM talked about earlier today that have not been taken up. They have to do with research agendas, a policy and an analytic framework that are absolutely vital. We need to go back and remember what it is that we said would be vital and revisit that. We need to dust them off and bring them forward again and freshen them up if in fact that proves to be viable today and relevant today.
The third point is about specific measures. I actually have a question for the group, whether it is around G codes or whether it is around minimum data sets or whether it is around what do we do with topical application of antibiotics and specific measures around that and gaming. Is it the role of the quality workgroup to set certain measures? The reason I ask and the reason Justine summarized the activities in the space today by AQA, NCQA, and AHRQ, and the whole alphabet soup is that there is tremendous work underway today to define measures – be they what they are – that look at not only effectiveness but waste and try to shine a light on that. So those things have a tremendous amount of momentum to them – more now than ever before. So the question for us is whether or not we want to allow those to naturally play out. They are public-private partnerships. I certainly do not think we will take them over. I have been involved in that work. Or do we want to try to take a significant place at the table? What is our role? It does beg that issue. If we feel that they are misdirected, if we feel that they need support or things that will not happen naturally, there may be recommendations we can make. That, I think, becomes a broader question within any discussion around specific measures. So I wanted to add that. First Don and then Stan.
DR. STEINWACHS: I really like having a piece that talks about what EHRs can and cannot do for quality measurement. And thinking about what can and cannot do, certainly there are some areas like talk about having coded information that could talk about what is needed in order to enhance the capacity of EHRs. One thought is the IOM does provide the six goals. So one could talk bout what EHRs have the potential of doing or not doing in safety, effectiveness – which is more evidence-based care, timeliness. Is that a way to get at it at all? Patient-centered – does it capture anything that is indicative of patient-centered care? Efficiency and equity, and in the equity certainly you would like to think that all of the EHRs capture race and ethnicity so you could do that. However, you go to most EHRs and I do not think that is part of it. I think in many ways we have developed an inflated sense of what EHRs may do. Most importantly I think there is a lot they could do if structured to do so. Maybe part of what the committee might do is focus on where the areas are to be strengthened and that would come back to some of the work in other committees too.
A second point that always bubbles up in my mind as a difficult one mainly because we are the National Committee on Vital and Health Statistics is how do you construct a denominator for an EHR. So most of the kinds of metrics that we talk about – and one of the reasons for using administrative data sometimes – is that you can say you were enrolled. It talks to EHRs having something or a link to something that says you entered the practice at this time and I now say you left the practice at this time. It is not a traditional enrollment system. Somehow it seems to me that we have a responsibility to be some innovative or creative ways to think about how do you create a denominator. And the last part of that is how would you take that and raise it up to a population level. First of all, you would have to have a denominator and then you would have to think about what is it you might be missing and how would you address that. There are certainly some things that are only received through regular medical care processes and maybe you would say you are not missing much if you can capture all of the providers that serve the population at hand.
DR. HUFF: It seems to me that one of the difficulties is that there are so many good things that you could do; so many opportunities that if we do not focus you can be in trouble. So I would just reiterate a couple of things. One is that I hint there is a clear boundary where you could say NHII workgroup is worried about implementation of EHRs. You could say that the standards and security subgroup is worried about enabling and approving standards that would allow for data exchange that would support quality initiatives but it is up to the quality group specifically to say how we want to use that and can we do an experiment that would actually take us through the steps. So I am not sure I can understand exactly what would be done from what Paul was saying but it sounded good; potentially the doctor office quality initiative as well. I am thinking that so many of these issues could be explored if we focused and said let us do glucose in ICUs or let us do hemoglobin A1C in the outpatient ambulatory environment. Let us not try to do it globally because we know that the EHRs. And in the spirit of going where the puck is you say let us go to Kaiser Permanente or let us go to Indianapolis and let us demonstrate what the value of this is in a place where it is being done. It is going to take a long time to being the whole infrastructure up to where you could do that globally. If you can say these people have done it right or at least better than some other people, then implement it in a situation where all the issues will come out where you will have missing data, and you can answer the question of what will we really learn from this. Is it hopeless? Can we get a denominator out of it? And do that in the context of a very focused demonstration project.
MS. MCCALL: That is a great point and that is what David was talking about today. We do not have the expertise today to be able to talk about it and you do a pilot to try to discover it. We have got Paul, then Bob, then Harry, and the Jeff.
DR. TANG: One is there are so many good things to choose from. The discipline is that we need to do the right thing which will produce the next big action. I think that goes back to our strategic planning, which is our roadmap. The next thing we could do that would have a big impact would be this nexus of EHR and quality measures. It is already in our domain. We already have the people and the workgroups even. I think that would be extraordinarily important. One of the questions that Don asked was what is the denominator? There are two aspects of the denominator. One – is that person your patient? Except for Kaiser, most of us do not know. Measures are most accurate and reliable if you need them for multiple reasons, including conducting your business. Why would we, a clinic, want to know who our patients are? Well, we implement something called open access. That means you call today and you get an appointment with your doctor today. Well the only way you can do that is if you have supply equals demand. The only way to get the proper number of physicians on board is to know how many people are in that person’s panel. Again, in the non-complicated world that is not a trivial task, but you need that number. So empirically we determined that if you had one visit with this person in 18 months it is on your active patient list. That is how we measure our demand, our business. It is also the key to who is your patient. The other thing we do – is that patient in the target population with that disease. And for that we use a problem list, not the two encounters that are billing. Those are notoriously inaccurate. For example, if you use the administrative claim data to determine who has diabetes you will miss a quarter of our patients. We have done that by gold standard chart review. You will miss 30 percent of our coronary artery disease patients. What kind of quality number if you have missed that significant a number of people? On the other hand if you look for diabetic using the problem list alone you get 94 percent of them based on gold standard review. So in other words there are some really good data in there if you were to use it and we were to reward you for that.
Then the comment on the educational piece and to piggyback on something Marjorie said as far as the retreat, communication is a missing element. We had this prophetic report. I am not sure how many people read it or know the output of it. It is one of the things we learned at IOM. The Quality Chasm report was based on data that was a decade old, but it was extraordinarily well articulated and written. It also had a brilliant communication strategy. That is what made it stand on the map. Part of that brilliance is attributed to Janet Corrigan and very rightly so. So we can leverage what we already do just by developing a communication plan. There is some money attached to it but there is still stuff that is part of just talking that comes free. So I think that focusing the roadmap and picking the right thing – we already have the workgroups in progress – and just create the nexus is an opportunity staring us in the face I think and one that no one else is addressing. That is the end of my comments.
MR. HUNGATE: I think that this picks up on another piece of the same issue, and tell me if I am wrong. You have identified the ways of defining population within your plan as a relevant group. We use terms of data pools, aggregator, and trusted aggregators. You are the aggregator in this process, really. You maintain the pool that has the information.
DR. STEINWACHS: They aggregate their own data which is different from aggregators who bring together data from a pool of different providers.
MR. HUNGATE: It seems to me that we are depending on the ability to define a population and understand who is in it and how we have dealt with those. Russell mentioned the missing data in registries. I think those are common themes of quality issue. What data do you have and how do you use it? If comparison to peers is the strongest change element we have got in the system – and I think it is – then how those pools and registries are used if used correctly is very much a pivotal piece of where we go. Am I on the same wavelength?
DR. TANG: You are right and the commentary is if we do not reward Wayne Gretzky with a goal it will not do that anymore.
MS. MCCALL: We have got Harry and Jeffrey and then Justine.
MR. REYNOLDS: I am going to take this hockey game a whole different way. I am working on some future looks at how North Carolina in general looks at quality and how we could do it, not just Blue Cross. We have got a lot of different factions. We have got Medicaid. We have got regular payers. We have got CMS involved. We have got everybody involved. If I sit back and listen to the people that spoke earlier and what I hear right now, I am sitting in the stands watching a hockey game where every player has a puck. The goalie has the net on their back and they are just running around so nobody can score and nobody can do anything because everybody is looking at it solely from their own view. We always use the word pay-for-performance and pay-for-performance means that somebody is going to decide how to pay for a performance. One of the things that I hope happens is if we get the situation where somebody mentioned minimum data set and I will not belabor it I will just use it as an example. For example CMS right now is requiring the submission of quality data in some ways. Comfortably all the CMS has in North Carolina is the Medicare patients in North Carolina. Blue Cross asks the same thing, and I consider your quality as you providers sitting around a table as whatever you did for me and Medicare does the same thing. So consequently you can inwardly think about it but the point is I would rather have a single source of all the situations in North Carolina with a standard structure. I am not going to say minimum data set, you guys are the ones that understand that. If I could then download that to study quality so that if I am philosophically evaluating Duke Hospital I am looking at all the cases at Duke Hospital, not the 100 open hearts they did for us of which 5 had a problem which means they had a five percent error rate; versus they did 10,000 and we had five of the seven. So I hope that there is some way – whether we play off of what CMS has got started – or we look at some things that are in place, group them up in a way so that each of us in states for example could grab it. I have even recommended at the University of North Carolina – because they have a data center that is really independent – maybe all that data ought to be sent there, set up in a single place, all of us pay to download it so that we can really look at your quality based on whatever is set up as goals. That way when we start to look at pay-for-performance, we start looking at some of these other things, we are looking at the entire environment of all the patients that they saw. Obviously the providers and others will get involved in deciding what should be out there.
MS. MCCALL: On administrative data, there is CFP coming online. There is AQA coming online. Are you talking about things like that?
MR. REYNOLDS: I am saying whatever is decided. I do not know all these acronyms.
MS. MCCALL: I understand. They become important, though, because what we do not want to do is be redundant.
MR. REYNOLDS: I am giving you the issue and looking at it through a different set of eyes than I have heard discussed around the table because the point is your quality is all the patients you see not just the ones that Karen pays for or we pay for or somebody else pays for. You have got to be careful that you do not let yourselves be marginalized.
MS. MCCALL: We have got Jeffrey, then Justine, then I think Bill.
DR. CARR: With regard to the privacy discussion that we had earlier, I think an important distinction that we need to think about is if you as a physician choose to have an electronic health record then all of your data elements are in an electronic fashion and are available for you to manipulate or analyze your quality. Are they also able to be submitted for external aggregation? That is different from patients being on NHIN or whatever if they go to an emergency room in Oklahoma and they live in California they will be able to be seen. I think that may be a distinction we want to continue to keep in front of us in the privacy committee because there is aggregation happening now every day every way within a hospital or group practice. This secondary use of data is underlying a lot of these p-for-p initiatives. So I just wanted to highlight that I think we ought to be really clear about what is on the information highway and what is electronic, aggregatable, and submissible to external agencies. Does privacy work on both of them?
DR. SCANLON: This was in part a reaction to what Harry was raising which I think is a completely valid sort of issue. Justine touched on part of it. The point you raise has come up in the context of p-for-p as well as in terms of trying to inform practice, particularly with respect to CMS. If you want to call it the 800 pound gorilla, you can because when it wants something providers pretty much do not have a choice but to comply. The issue of providers’ sensitivity and rights comes up in that context. So we raise it her from the quality perspective but there is also the issue of a privacy and confidentiality perspective. I thought earlier Stan was going to go to what are the population quality implications of the things that all the other subcommittees are doing. Is there a role for a population perspective in terms of the other subcommittees? When Mark was talking today about the various recommendations and you could think about it in terms of the cost and benefit to the individual of what would be the vision on each one of those there was also social cost and social benefits that were associated with every one of those. And the question would be when do the social costs for the social benefits override the individual’s perspective and should dictate the decision? Not having seen the report, maybe they are all being dealt with already but there is that question of from a population perspective how is the work of the rest of this committee taking population into account? Would that be our strength to work as a committee on these issues? One of the things I think the quality workgroup has had difficulty doing is from this long litany is selecting the prime opportunity; whereas some of the other subcommittees have got some prime opportunities which they have been very successful in picking out or they have been dealt to them. They are part of a natural role that is ongoing. There is a question of should there be a population perspective to some of those things.
MS. MCCALL: We found ourselves with kind of a blank piece of paper and needing to fill it in. There was so much to choose from and so much underway that that should actually be a blessing if we seize the day. If we also think about it quality is more like an emergent property. It is the result of – not the sum – but the result of a number of other things that happen. I have heard a couple of people say that we may want to create a nexus or pilot or some way in which we bring a lot of pieces together. We have a visitor that is with us who would also like to join the discussion.
MS. BICKFORD: Carol Bickford of the American Nurses’ Association. There are some issues that I just want to toss out to think about. One is you have been focusing on physicians primarily as providers and reimbursement from the structured payment models, third party payers, or CMS. The majority of us are paying out of pocket for initiatives, for health and wellness components, supplements and the preventive therapies. We have no mechanism for understanding those for quality initiatives. I would also identify that you have not necessarily addressed the need for a standardized terminology. That has not been recognized in the CHI initiative focusing on the concept of outcomes. There has not been discussion on the linkages of the diagnosis, interventions, and outcomes. Nursing terminologies have been struggling with that and working with that for many years so feel that we have some perspective in that arena.
The biggest issue I think is that there are many other providers that are engaged in this initiative focusing on health and wellness, not disease and the pathology. What does that have for the implications for Healthy People 2010?
MS. MCCALL: I think those are great adds. Again, when I said that we went back and read the earlier vision documents, that perspective was brought through very clearly. And so we did not do justice to the depth and the richness, but what she talks about is very real whether it is functional status and the whole ICF paradigm to statistics that are outside the healthcare system and the delivery of care and go into the community and what the resources are that are available. When you try to get the root cause, what is that context like? It is not just SES. It is all different types of resources, everything from health and education to education in schools and all that type of infrastructure. So I think that is a very good point. Paul you had actually wanted to jump back in and say something nasty to Harry.
DR. TANG: On the contrary I wanted to compliment Harry on an excellent idea and I think there may be a new idea to rise from that that may be a very short-term recommendation. There is something called p-for-p, the integrated IHA program in California which I think is one of the largest. I think there are something like six payers that participate. So we had to not self-report because we do not keep good track of who the payer is for each patient. Consequently, with all due respect to Harry even though I like him, I am not a Blue Cross provider.
MR. REYNOLDS: That would be quality at work.
DR. TANG: We accept Blue Cross but we do not think of ourselves as a Blue Cross, Aetna, or Cigna provider. Yet, we are measured as I have to have my Blue Cross measurement, my Cigna measurement. In fact, that is how IHA is doled out. I think Harry’s idea is an excellent one. We are a provider that should be measured against our patient population, each of us individually and as a clinic. So he should look at my score – our score – and then take the proportion of patients that are Blue Cross subscribers and that is how he gets his score to report rather than taking a different slant with a denominator of only Blue Cross patients. That way we can report once on some standardized measure and that is what characterizes us as an individual provider or provider group. They all get their separate percent credit for that, but that basically would redefine quality. I do not think that is something that I know is done.
MS. MCCALL: Yes, that is not done today because data is not aggregated enough today because AQA or CFP are three letter acronyms. They do not really exist yet in that sense of having aggregated data so it is not possible. You are right. The broader question that I would have for NCVHS and this goes to Simon for a thought exercise is p-for-p and those types of mechanisms because we could sit and chat for the rest of the day on different ways to design those from an economic and actuarial perspective and all that. Is that what we do? I hear you loud and clear, but as I try to bring us back to what are our short-term and long-term opportunities, do they include planning and/or making recommendations for p-for-p. I am getting a lot of body language that says no, it is not. However, there may be things that we can do that are within the areas of our domain that make that easier or more possible or less possible if they tend to have perverse incentives. I just wanted to leave that question for you, Simon, for us to address. Don you had a point that you wanted to make.
DR. STEINWACHS: One comment and one point. In Maryland, the Health Services Cost Review Commission is in the process of developing a pay-for-performance system, which is an all payer state. If it goes into being, it will be the first time that all payers including the uninsured going through hospitals would be subject to the same pay-for-performance standards. So it would be an experiment to look at. The other piece of that – which I think is one of the very tricky ones – I remember when people suggested that you pool together your Medicaid enrollees with your other enrollees for measurement. They said we need a separate Medicaid hetis(?) measurement set. I think it was not so much the issue of different measures. There are some rationales for some different measures, but it was because you pool those two together depending on what that ratio is and your performance looks wildly different. So I think that this sort of total performance measure raises a troubling issue many times is do you somehow case mix adjust it? And is that somehow then say that the racial composition, the socioeconomic composition, other factors that seem related to quality performance are adjusted into that? So it would be a great area if you really want to think about it for NCVHS to address in terms of this idea of pooling measures, pooling data, and pooling performance. Then there is a fair comparison issue as well as a standards of expectation.
MS. MCCALL: Why is it different when you pool?
DR. STEINWACHS: Historically it has been that Medicaid enrollees in health plans meet the criteria about half as frequently as middle class enrollees in health plans. And so there was a real disadvantage to health plans depending on whether you had five percent Medicaid or 25 percent in how you would be looked at. I think you get the same thing in hospitals when you do it by payer there is a kind of segmentation that deals with socioeconomic class, it may deal with race and other factors. It is one of the reasons that I think from some points of view it is actually attractive to disaggregate. From the other point of view I would agree totally. I think it ought to be an aggregate measure. I think everyone ought to get the same quality of care but that is not the incentive experience.
MS. MCCALL: There have been discussions about that at AQA and I will tell you payers would very much insist on having the ability to look at those very differently by payer. If it is a three-payer market and for whatever reason the value that is delivered to people who are Harry’s members are different from those that are delivered to Simon’s and somebody else, they are going to want o actually pay out differentially. That does not mean that you cannot start with a whole piece and dial it up and down and there are techniques to do that. The second point is that if you case mix adjust everything away – the AQA had discussions that if you adjust it all away – everything is fine. They are having very robust discussions on how to do that and methodologies to do that. I am not saying they are all perfect but they are in fact taking place.
DR. STEINWACHS: It raises some of those ethical issues. It seems to me that at the heart of this idea of what should America be getting in quality versus why is a provider want to respond.
MS. MCCALL: As opposed to the specifics of actually getting down to it if we back up and look at the policy issues there. This has been very rich discussion. What I would like people to do is look at the specific agenda for today. We have actually addressed a lot of this along the way. We are certainly not going to get to a lot of it. We have talked a lot about the first one – short and long term objectives that this analysis suggests. In and out we have talked about other knowledge that is needed and had some recommendations also about number three – how other NCVHS subcommittees and workgroups could come together. I have not really heard anything regarding number four. Does anybody see any immediate short-term recommendations that would be valuable to come from this work to the Secretary with a little bit more work? I am not talking about being off the rack. Paul?
DR. TANG: Is it possible that recommendations about further study of clinical measures and how incentives can be designed to reward use of clinical measures could be short-term recommendations?
DR. COHN: I am trying to think of what that means exactly. Are you meaning using clinical data? Or, like hemoglobin A1C is a clinical measure.
DR. TANG: No, but the way the denominator is determined is done administratively. If we determine it by using clinical measures then you would have a different ratio. You might have to have different goals because it takes all comers not just those that show up more frequently for example. As you now, both HHS through CMS as well as the rest of the community are proceeding on the traditional AQA way.
MS. MCCALL: I think that my initial reaction is that we will take this back to the next conference call. There is enough information to make a general recommendation around its importance. There is not enough information around specific, actionable, here-is-exactly-how-to-do-it recommendations. So I think it could be done with some work. I just do not know personally how much value there is or how much it needs that extra push or we want the positioning as the committee.
DR. TANG: Could we recommend a demonstration project substituting clinical measures?
MS. MCCALL: I think those are already in place so there were some pilots that were awarded by AQA and one of them that is on the books is in fact to look at quality measures that are specifically driven out of EHRs as opposed to administrative data. So again there are a lot of things already underway and it is really important that we have knowledge of that landscape so that we do not double down. Any other short-term recommendations that people saw as potential out of this?
MR. HUNGATE: I am not sure we ever got back to Jeff’s initial comment about secondary uses of data and what the follow on is to that in our activity. So maybe that is just left as an open question. Part of what I wonder about there are the issues that Russell raised about missing data in registries. I think there is downstream some strong expectation that data pooling will result in better evidence for feedback into decision support systems and that there is a feedback loop around a collection of clinical data and use of that data to improve the evidence for future treatment. This is a serious underpinning for quality long-term and quality improvement. I do not know whether that is a topic generically which is of importance to NCVHS first and secondly whether it is appropriate for the quality workgroup. So that is a question that I want to lay on the table as a finite piece of content that I am uncomfortable with the degree of work that has gone in preparation for its presence. I guess in the future if data work is going to take place, I do not know if we are ready for it.
MR. BLAIR: Thank you, Bob. In a sense I originally asked about secondary use of data but then Carol took us through everything and then she led to the questions at the end. I really felt as if the DOQ-IT program – Doctor’s Office Quality with Information Technology for the small practices CMS program – included so many of the things that Carol had reviewed and there is the secondary use. But, it is not just secondary use of existing data, it is secondary use of clinically-specific data which SNOMED-LOINC and RX-Norm. There is the leverage for the future. There is the near term piece because it is an existing program. It was introduced about eight or nine months ago and I think funding may be either cut back or limited for that. If there is anything that we could do to celebrate, promote, or even just protect the funding for that I think in a short-term, near-term basis moves us more than anything else. The other piece, and this is the comment that I passed up, I feel like this is a really important topic area and I sort of feel like you have got to handle on the thing. You have got a foundation and I just envision we are going to build on this and this is going to come out with something that is really good. When it does come out with something that is really good, it is going to wind up going to Congress. And if it goes to Congress my thought is what are there priorities. Is it quality? It is probably access to care and it is the cost of care. The thing about it is that possibly some people could make the argument that the biggest, most effective lever that we have to control the cost of care and therefore have the possibility of increasing access to care is quality. When you get to the point of putting a title on the document and you are doing that first introductory paragraph, maybe it could lead with something to the effect that the most profound effective levers to control health care costs and improve access to care turns out to be improving quality and patient safety. Maybe at that time we could get a more receptive audience to the folks that are in a position to fund this stuff.
MS. MCCALL: I agree. We have a marketing problem with respect to quality and the term. Simon wanted to say some things.
DR. COHN: I think this is an excellent base piece upon which to build. You were asking about short-term pieces and I am not certain I am seeing short-term as opposed to low-hanging fruit, which I tend to think of as a slightly different thing. It has a little less urgency but without a lot of work are things that are doable within six or eight months to a year’s time, especially in the first couple of bullets. I was actually going to also suggest leveraging other things that may be happening. Your agenda seems to overlap a lot with the things Jim Scanlon was talking about this morning in relationship to a workshop, looking at secondary uses of data, surveying instruments, et cetera. I think we need to think of ways to leverage all of those things to help us reach our goals.
MS. MCCALL: I think that is a great comment. I had hoped to have more time talking with Jim Scanlon this morning. His laundry list of things is delightful and I think could be very fruitful in terms of creating some long-term.
MR. REYNOLDS: I just want to make one comment. I sense that all of my comments were taken as a payer. I was stepping back from that. We have one health care bill in North Carolina. I do not care whether you are uninsured, Medicare, Medicaid, or you are a commercial customer. There is one bill in North Carolina and we are all paying for it. I do not mean one bill, do not take that literally. I mean there is one cost of healthcare in North Carolina. So, a hospital has to make up their uninsured and they have to make up their underinsured and everything else from everybody else. We are trying to step above that since a lot of people are working in a single state. So my comments had nothing to do with Blue Cross or a single payer. They had to do with standard protocols. In the overall bill it does not matter whether you are seeing a Medicare person or anything. If we do not make them better they end up in the emergency room and we all pay for it. So I want to make sure you did not just hear my words as a payer. A lot of us are at a whole different level from what our day jobs are on some of these issues.
MS. MCCALL: I think we are about out of time, so we may run about 5 minutes over at most just to wrap up so we can stick the landing. First is really just to say thank you for your very involved and enthusiastic participation in what has been a very open invitation. There is a lot of work that has gone into this and a lot of time that has been involved. You have done it justice and honored that. So thank you again. There was a lot around the terms of aggregation and nexus and talking about the opportunities around the Doctor’s Office Quality program. I think that this could be something that is truly transformative. Think of it this way. We have been successful – and granted I am a freshman here – but so successful in fact that our subcommittees each as they are pillars in their own right, now have an opportunity to come together in a sort of nexus which is in fact the ultimate goal.
If there are some kind of low-hanging fruits, I also see the potential for a recommendation that basically says it is time to bring some things together. To do so in a way that Jeffrey, Paul, and Stan talked about in almost a pilot and say what does it actually mean end-to-end or full circle because it is not a line. It has specific feedback loops, and to understand what it means in a small way to do the whole thing. And that would be the recommendation. It would be specific and pithy and it would not come from the quality workgroup. It would actually come from you, Simon. And it would be something that perhaps we could take on as one of the homework assignments or one of the topics for the retreat. Things are not lining up yet. All of the work lining up today in all of the different places is incredibly vital but is not going to get us there if it is all that we do.
If NCVHS is going to go toward where the puck is going to be; if we think ahead of the curve, this is ahead of the curve to connect these dots. So that is a potential and I think maybe we can talk about that in the executive subcommittee. As one of the other takeaways, I think we should calendar and schedule a quality workgroup to work through many more specifics but I think there are many opportunities to look at partnering and seeing what Jim Scanlon is doing. Looking more specifically at the Doctor’s Office Quality program and finding out exactly what they do need. Trying to find other potential playgrounds where they may be ready or ripe to try to bring some of these other pieces in, including the analytic framework. So thank you very much.
(Whereupon, the workgroup was adjourned.)