The Federal Stake in Quality

( View the PDF )

The government of the United States spends $450 billion annually on Medicare alone. No single player in the healthcare industry has as much weight to throw around in the push to promote higher quality healthcare, and lately Washington has not been afraid to use it. An executive order issued by President Bush this summer is the latest in a string of orders and legislation that call for more accountability in how information on quality is disseminated. Beyond the government’s role as a payor and regulator, a key role is to bring together hospitals, payors, physicians, employers and other stakeholders to find common ground in the push for higher quality. HealthLeaders recently gathered a panel of experts in Washington, D.C., to assess where the federal government’s role in quality stands now, and where it is headed.

Panelist Profiles

JIM MOLPUS, editor of HealthLeaders Media, served as moderator

CHIP KAHN, president, Federation of American Hospitals

J. REGINALD HILL, partner, Waller Lansden Dortch & Davis

CAROLYN M. CLANCY, M.D., director, Agency for Healthcare Research and Quality

PETER PRONOVOST, M.D., medical director, Center for Innovation in Quality Patient Care, Johns Hopkins University School of Medicine

Sponsor

Waller Lansden Dortch & Davis
www.wallerlaw.com

Roundtable Highlights

State of Expectations

JIM MOLPUS (HealthLeaders Media): A recent report from the Commonwealth Fund gave the American healthcare system in general a relatively poor grade as compared to other countries. Do you feel the grade was too low, too high or about right?

CAROLYN M. CLANCY, M.D. (AHRQ): Many reports show that we have a lot to do to improve healthcare quality, but we are seeing an enormous amount of progress. In particular, I think we are seeing a lot of progress in the hospital setting. For acute care, we are seeing some dramatic improvements—whether that is the result of simply transparency and public reporting I think is an open question. We are in the midst of a phase where people across the system finally get the fact that we can’t manage what we can’t measure. As an agency that supports a lot of tools or measurement, I am excited about that. But it is going to be a culture change.

CHIP KAHN (FAH): We have an accelerating effort to measure what is done in the healthcare system, to report on that measurement and to have a feedback loop with the physicians and providers to try to improve care. There is an incredible amount of effort going into this. It is really just getting started and is not perfect by any means. But I think to judge us now is to judge us in the midst of a process. That is why I felt it was a little unfair (for the report) to take the shot that it did.

J. REGINALD HILL (Waller Lansden Dortch & Davis): The momentum for change is unprecedented. In addition to measurement, the biggest change is the incentives that are being provided and the interesting developments on the consumer-driven side. I think that is where we are going to see some impact. It is going to take time.

Multiple roles

MOLPUS: The federal government is the single largest payor of healthcare in the country. But beyond that, the federal government has numerous roles in driving healthcare quality. Of all the roles that the government has in quality, which ones have shown the most progress?

CLANCY: There is no question that the government’s power as a purchaser is incredibly important. Quite recently there was an executive order signed by the president that says that the federal government will take its role as a purchaser very seriously. That means that HHS, Defense, the VA and the Office of Personnel Management are all going to work together to use their power collectively as a purchaser of healthcare to encourage improvement in a few areas: transparency in the quality of care; transparency about pricing, which is very difficult; to promote interoperability of health information technology; and then the incentives piece, which I think is very, very important. That executive order was preceded by several very important demonstrations to test new ideas and new approaches that were part of the Medicare Modernization Act, as well as a couple of demonstrations that preceded the bill. So we know that when the government speaks clearly, improvements will be seen.

MOLPUS: What about the government’s role as a convener?

CLANCY: This is a role that the agency (AHRQ) tends to play quite a bit. We don’t pay for care or provide it or regulate it, which in some ways means we don’t have a “dog in the fight.” In this country, in order to see sustained improvement, you have to have alignment between what private payors are demanding, what states are demanding and what the federal government is demanding, or you are going to drive providers crazy. You can predictably increase burden. It is very easy to add multiple, continuing demands and measures. But if you do, it does not allow providers to be able to focus their attention and resources on improvement. The executive order sets the stage, but there is a lot of very important collaborative work that needs to happen between the private sector and the federal government, because at the end of the day, virtually all hospitals, physicians, other healthcare professionals and healthcare settings see some patients who are insured by the government and some who are insured by the private sector. So it all comes down to one final pathway.

HILL: There is another role the government plays that is not the one that is perhaps as forward-thinking as what we have been talking about, and that is the enforcement role. The government is in a position to direct the quality of care through what is required as a condition of participation. What happens if you don’t comply? It can be a False Claims Act claim, or exclusion by the OIG. Those are things that are certainly the last-ditch efforts to enforce quality, but it is an important role.

KAHN: There is another factor, too. When we got started with the Hospital Quality Alliance, we had 10 measures out there very rapidly, and those measures were going to be reported to HHS on the Web site for Hospital Compare. About the same time Congress came in and said in the Medicare Modernization Act that if you don’t report on these measures that the Hospital Quality Alliance approved, then you lose 0.45 percent of your update. Congress came back in the Deficit Reduction Act and said you lose basically two-thirds of your update. Right now there are something like 140-plus hospitals that will not qualify, but that means well over 4,000 hospitals are qualifying. There is sort of a carrot and stick here where the government is clearly using its buying power to get everyone’s attention in terms of reporting, and that has happened with the HQA measures. Obviously I would have preferred if we could have done it voluntarily, but at the end of the day this just makes it happen. Since the measures are being developed by a collaborative, I don’t think we can really complain.

MOLPUS: What are some of the issues going to be as these measures evolve? Could we see some of these measures reflect more of a stick and less of a carrot, so to speak?

KAHN: However you want to define pay for performance—rather than what you have now, which is pay for reporting—has to be a concern. The measures we have right now from a clinical standpoint are basic blocking and tackling. They are a minimum set. As we go to more complex measures, and as some of those become more ambiguous in terms of how they are read and the conclusions drawn, you have to be cautious. I’m confident, at least in the current administration, that Secretary Leavitt is approaching this issue with a collaborative notion.

CLANCY: I think Chip’s point about basic blocking and tackling is correct. Having said that, however, the reports on Hospital Compare (www.cms.gov) make it clear that some hospitals weren’t doing so well. For example, one paper showed that early on, academic medical centers did terrific at cardiac catheterization and not so well at community acquired pneumonia—pretty straightforward stuff that wasn’t happening. I have colleagues at multiple academic medical centers who now have the resources to make sure that this happens the way it should every time to get the right care for the right patient every time. I think where it will get challenging and far more important is when we begin to tackle issues of chronic illness care. That requires a degree of collaboration between the outpatient and inpatient settings. We don’t really have the accountability structure for that most of the time.

Transparency ahead

MOLPUS: Let’s talk about the executive order in a little bit more detail. In practical terms, putting transparency in an executive order is one thing, but how does that work in the actual healthcare system?

KAHN: First, pricing, payment and cost calculation in healthcare is not just arcane and complicated, but in a sense, rightfully so. It is not like buying a commodity off the shelf. It is generally much more complicated. Second, when we think about what kind of pricing transparency policies we want, we have got to think about the audiences. We have uninsured people—some of whom can pay something, some of whom can’t pay any price. Then we have insured people who have great variability in their coverage. Then we have people covered by Medicaid or Medicare who have different kinds of coverage. So we need to develop strategies for at least those audiences.

CLANCY: At a very basic level, people deserve to know what they are going to have to pay. It is astonishingly difficult in certain circumstances to find that out. That feels, in its essence, a bit unfair. The executive order says that people deserve to have this information. And taxpayers in some ways deserve to know what their contributions are paying. What is really surprising is that we know very, very little about what anything in healthcare costs. It will take a lot to figure out how to present this information to people in a way that is fair and transparent. I am encouraged that providers see both the problem as well as the opportunity to be part of the solution.

HILL: The pricing is the hardest part of this both for the provider to decide what they are going to publish and the consumer to make any sense of it. As Chip said, unless you know what your insurance plan is going to do, the data is not all that useful, anyway. It definitely comes down to what am I going to pay, and I don’t think any provider can answer that question for all groups, except maybe the uninsured.

MOLPUS: Is it a goal of the executive order to exert enough pressure through transparency to shake out the pricing system?

CLANCY: Our current system does not always pay for what we want in healthcare. I think the executive order sets the stage for us to define what it is we want out of healthcare, and then to think about how we pay to get there. Just to give you an example, the New York Times had a terrific series of articles about diabetes as an epidemic. They made the point that we pay, and often very well, to treat the complications of diabetes. What we don’t have a good payment model for is preventing the complications to begin with. Our current payment system actually derives from a time when the biggest problems facing the healthcare system were acute. We are coming to the realization that we don’t have a payment system that is aligned with the needs of people with chronic illness, and we need to actually reward providers for helping patients maintain the best possible controls in the early stages of disease. What that means is that most providers end up being a little schizophrenic in having to constantly shift thinking about what gets paid for and what is the right thing to do. Converging those two paths is where we all want to go.

MOLPUS: So these will ease the perverse incentives out of the way?

CLANCY: It at least puts us on the right path to deal with them. There are all kinds of perverse pressures. Gannett News Service published a study recently showing that hospitals in very poor areas tended to have the worst performance on some of the hospital quality measures for people with cardiac disease. That on its face is not terribly surprising. The difference I thought was very surprising, as did a number of my colleagues, is that most of these hospitals are in the lowest quintile of performance. The big policy question is how we structure pay-for-performance programs when we get there. Is it about an absolute level of performance, is it about improvement, or a combination of the two? Finding the right mix is going to be pretty tricky.

MOLPUS: Beyond the executive order and the government’s influence on transparency, will consumer-driven care be an effective driver of transparency?

KAHN: I think the jury is still out on whether we can move the system to a point where individuals or their families are making economic decisions frequently about the care they receive. If you wake up with chest pains, you want to get to the closest hospital as fast as you can. When you are sitting on a cold table with one of those aprons on is not the time you are going to want to negotiate with your physician over the charge.

PETER PRONOVOST, M.D. (Johns Hopkins): The literature would say that patients are price-sensitive, that they will consume less care as they incur some cost for purchasing healthcare. We could influence their spending by having them purchase more. That trend of putting more of the burden on them seems to be continuing and will likely increase. What is concerning and unknown is whether those purchasing decisions they are making are truly in their best interest—they may be compromising care that in a longer cycle time will benefit their health, but because the potential benefit is delayed, consumers won’t make decisions in their best interest.

CLANCY: If people have to pay more, they will go in less. No question about it. The problem is that a lot of them haven’t had the training, so they may delay or defer care when it really matters. That is a concern in the short run. In the longer run, for us to actually get value out of better care, patients have to be a very important part of the equation. For the most part, people are pretty passive. Studies show on average that patients ask about 1.4 questions per encounter, including questions about parking. So they are not actually pestering clinicians much at all. We’re not going to solve the challenges of chronic illness care and shift that to an earlier stage without activating the engaged patient. Transparency sets the stage for people to play a more active role in their own healthcare, and ultimately that will show up as a much different kind of concentration of health expenditure than we see right now with 20 percent of the patients incurring 80 percent of the cost.

Setting priorities

MOLPUS: One of the biggest levers that the government can pull on quality comes on the research side. Dr. Clancy, what are your agency’s research priorities on quality?

CLANCY: We have had a huge opportunity to invest in projects that examine how to make healthcare safer, in particular. Quality is not something that fits into a bumper sticker. Patient safety people get right away. They understand harm, and they don’t understand why we can’t fix it, and they are right. Our portfolio is pretty substantial because we have been investing $50 million or more a year for the last five years trying to identify particular problems and risks inside and outside of the hospital, and in developing tools to try to help hospitals where a lot more is known about harms.

MOLPUS: What are some specific quality tools the agency’s research is developing?

CLANCY: Next year the hospitals will be reporting on the patient experience of care in a systematic fashion. That will be of enormous interest to the public. We now have a tool that many hospitals are clamoring for to assess whether they have a culture to improve on safe healthcare. It is wonderful when the CEO says, “I want safe healthcare and we are committed to it.” But if the people who work for him or her are getting the message that it is not OK to speak up about reporting harms or possible risks to patients, hospital leaders and boards need to know about that disconnect. This is an anonymous survey that helps them assess that.

MOLPUS: Dr. Pronovost, you headed an AHRQ-funded study to improve care in ICUs across Michigan. How have you been able to take the lessons from that project to a wider audience?

PRONOVOST: Let me share a story of hope in making this happen. Our teams are collecting data prospectively, and it its scaled such that an ICU, a hospital, a system or a whole state can get a scorecard. But it still requires manual data entry. We recently went to some of the large healthcare IT companies and said, “Here are our standards, and here is the data we want, could you automate it so it imports in?” They were, perhaps to my surprise, delighted to partner because they see if they can prove that they add value to their system by getting measures, everybody wins.

CLANCY: We are thrilled about the results that Peter has had where many hospitals have had zero serious infections in ICUs for months on end, which is unparalleled. It clearly took an infrastructure to get there. From what I’m hearing from Peter, lots of hospitals want to be part of this approach. No one is paying for it. This is just, “Wow, if you can show me how to do it, I want to do that.” So a big focus of our work is trying to make the right thing to do the easy thing to do.

A unique industry

MOLPUS: Government creates standards in other industries—banking and airlines among many. There has been some degree of a cry from the industry for the government to set standards in areas like interoperability of electronic health records. Is healthcare just so different?

CLANCY: I think there are technical issues. But I think what the secretary sometimes refers to as sociological or cultural issues are far more important. If you think about other aspects of life, right now getting your boarding pass is the easiest part of flying. The real issue is that healthcare is kind of a cottage industry. We have not had the training in place. There are models that predict in the next three to four years that 50 percent of docs will have electronic health records if the government and private sector do nothing. This is going to happen. The real trick is how we make sure that it is steered in a direction that supports the value and what we want out of healthcare. That is what the executive order begins to put in place.

KAHN: Hospitals are generally large enough institutions that they can handle the transformation. Clearly banks did. Clearly grocery stores did. Those were simpler systems, and they were not necessarily life and death. Second, they were not part of a cottage industry. I just went to my internist the other day. It’s a three-internist practice and considered one of the best in Washington, D.C. And there are just rows and rows of paper records. I can’t see, unless my physician feels compelled, that he and his partners will be spend the resources for electronic health records. Interoperability isn’t even the issue. Paper is interoperable. You put it in a fax and send it someplace else.

HILL: The IOM has recently suggested that the way pay-for-performance programs should work is that providers who have the capacity should be made to go into it sooner rather than later. The IOM proposal was to reduce reimbursement across the board to provide incentives to get people to participate now and provide incentives to pay for the technology as well as the training, because that is another piece. I go to a solo practitioner. He’s about 70. He and his staff would have to do a lot of work not just to buy some equipment, but to completely retrain everybody.

PRONOVOST: And likely not realize a direct benefit to him, and probably a cost. To him or her it would probably be more aggravation.

KAHN: I don’t want to be the skunk at the party, but particularly with physicians we have a couple of countervailing factors in terms of payment. On the one hand we can talk about restricting Medicare payments if they don’t have IT. From the average physician’s standpoint, Medicare is not paying them enough as it is. And as we know, unless Congress acts otherwise, they are going to get a 5.1 percent cut in their fees on average on Jan. 1. Even if that is turned around, they are not doing that well from Medicare. So the incentive is dubious.
MOLPUS: As it relates to creating composite scores for P4P, what are some of the policy decisions that go into them?

KAHN: We are creating P4P, as we do with many things in healthcare, because there is an imperative. There is the presumption that we are not getting value. There is some evidence that that is the case. So policymakers and third-party payors are looking at this as the latest panacea. One of the pities of us rushing into pay for performance is that with HHS’ Hospital Compare we have a tremendous national experiment, and it would be great if we could see how Hospital Compare works with the Premier demonstration, which has a P4P factor to it, and determine whether reporting alone is enough to move institutions. The initial results from Premier seem to indicate there is, but that doesn’t necessarily say the payment did it; it may have been the information that did it.

CLANCY: The reality is that most of the private sector innovation where most of the experimentation is going on around pay for performance is a very, very small percentage of the overall revenues that are actually on the table. That will be the subject of some debate: How much is a good incentive as opposed to something that might actually distort healthcare in a very non-productive way? So far, no payor has stepped forward with enough of an incentive to make that something to worry about. I think employers and payors are feeling way too much urgency to say, “AHRQ, do the definitive study and please let us know rapidly how to do this well.” But I do think we have an opportunity to learn about this as these programs are implemented. So, for example, on the question that Chip just raised, we were actually able to support an analysis of some of the Catholic hospitals that did not participate in the Premier demo, but they were part of Hospital Quality Alliance and were reporting publicly, and compare that to some of their sister hospitals that were part of the Premier demo. Was performance all that different? In some cases yes, in some cases no. I think we are in a relatively early phase in terms of why we are doing this and what we expect in return.

PRONOVOST: The percentage of incentive in payment is on average lower than the cost to measure the thing. You get a 1 percent to 4 percent, and we are somewhere in the 2 percent to 6 percent range to put the infrastructure in to measure these things, so somehow we either have to increase the incentive or reduce the cost of collecting these measures. From my perspective as a physician and researcher, the numbers of interventions that are supported by sound enough data to measure by pay for performance are frankly quite small. There are just not that many things that you can get great consensus on that you should do. And I see pay for performance as a component of a strategy to improve healthcare quality, but probably a relatively small component because it will at best cover less than 10 percent or probably less than 5 percent of a hospital’s discharges. We need other strategies to reach that other 90 percent of patients and make sure we do things for them.

MOLPUS: Is there a teaching to the test danger? Are we raising the overall level of quality or concentrating on somewhat limited measures?

PRONOVOST: There certainly is a dark side. I had a discussion where we were looking at one of the measures for pneumonia that we were doing marginally on, and our physicians uniformly said, “I have many more sick patients that I should be giving my attention to rather than focusing on this when they come into the ED.” Is that the right thing to do? Reality comes in and says there is a limited amount of time, so I will redirect my attention from this group of patients whom we are not measuring and if we cause harm it will be invisible, so we can focus on this thing because it is being measured in public reporting. There is no doubt we will allocate resources to get those measures improved, which is why we have to choose those measures wisely.

What lies ahead?

MOLPUS: As it relates to quality, I’m curious what each of you think future policy debates will be about?

HILL: It’ll be interesting to see where liability shakes out if a provider doesn’t follow one of these measures. As we have more process measures, where does that come out under liability, which is a state law? I also wonder how all of the activity will wind up creating a standard of liability for hospital boards, which traditionally we could view as a state law issue. Will there somehow become a standard of care that is established in federal law, much as we did with Sarbanes Oxley for public companies? I wonder if we’re going to start seeing that happen more in the area of quality care.

CLANCY: One new dimension will be how to give patients good information to make choices when there’s more than one right answer. Right now, most of the quality discussions are about areas in which we know exactly what to do, and how often we should do it. When we get into how well we inform patients—did patients have a free choice of info and so forth—that’s trickier. But when you think about all the new breakthrough treatments that are likely to come online over the next few years because of advances in genomics and molecular biology and so forth, that’s going to be really important. The other issue we’re going to be seeing over the relatively near term is secondary use of data, because it has both policy and privacy issues connected to it. If I go to a facility with electronic healthcare records, can they sell my data without my permission, even if it’s de-identified? IT will be disruptive in that sense because people who have the capacity to aggregate large amounts of data are going to have a lot of power.

KAHN: I do think for most of us, over the next decade we will have EHRs. The amount of info we’ll get, not just about particular patients, but about populations, could really have tremendous effect on care. So I’m very optimistic that EHRs, as well as genomics with the information availability, if identifiers and privacy don’t get in the way, could really produce a tremendous amount of knowledge that will really improve care.

PRONOVOST: The push to increase the rigor of the measurement is going to go up dramatically. A common example is for infections, where we use surveillance definitions that are much less rigorous than clinical definitions. Many people believe that as you go to public reporting, maybe we ought to go to the tighter definitions. A second theme I see is we’ll be discussing soon how much safety or quality we will be willing to pay for. I think we thought generally of both quality and safety as dichotomous variables, meaning they’re safe or unsafe. Really, there are degrees of how much I’m willing to invest. There is clear evidence that if we put every patient with one to one nursing, we would reduce harm. Well, that’s not a feasible solution, either financially or from a human resource perspective. But I think as the value becomes more transparent—that is, both the cost and the benefit—we’ll have that decision.

( View the PDF )