Center for Studying Health System Change

P R O C E E D I N G S

Keynote Address

PAUL GINSBURG: May I have your attention. I think you have gone far enough along in eating to continue and listen to our luncheon speaker. I am very pleased to introduce to you Janet Corrigan, who is director of the Board on Health Care Services at the Institute of Medicine, where she is responsible for projects relating to health care quality, benefits coverage and the organization, delivery, and financing of health care. That is a lot.

She is also director of the IOMs Quality of Health Care in America Project 3-year initiative to design to provide leadership, a strategic direction and analytic tools that will contribute to the threshold improvement and quality over the next decade.

Actually, rather than reading the rest of it, I just want to say that I’ve known Janet for a long time, and Janet is probably one of the people, at least in Washington, if not the country, that those interested in talking about quality and initiatives to improve quality would like to proceed to go ahead and consult with, so I was really delighted when she was willing to come here to give her perspective about the material we’ve been putting up here and discussing about, both this morning and the afternoon. Presumably, this has some implications for the movements to improve quality of health care in this country that Janet is so intimately associated with.

So I want to introduce Janet, and she will be taking some questions afterwards.

Thank you.

JANET CORRIGAN: Thank you, Paul, for that very nice introduction. Paul neglected to say that when I first came to Washington about 13/14 years ago—it’s been a long time, Paul—he was really my mentor. I started working at the Physician Payment Review Commission under Paul’s direction and learned a tremendous amount. And later, after that, I had a chance to work for him again during the very early years of the Center for Studying Health System Change.

I would, first, like to start by just commending Paul and his staff and all of the participants in the Health System Change effort. You have really produced a very rich set of studies and information and food for thought that I found very, very stimulating as I moved through some of it before I came here today, and the discussions this morning have also been very interesting.

It is real clear that there are some very, very important benefits to having a tracking system which allows you to really monitor change over time and also to begin to pull together not only the kind of information we get from quantitative surveys of physicians and household surveys, but with some real rich contextual information that comes from these various site visits and the in-depth analysis of particular communities. It is a real resource we should all be very pleased to have.

Having said that, I think that the picture that these studies paint is not a particularly positive one. It’s not all negative, and I am actually one who is quite optimistic about the future over the long haul, but it’s not a particularly positive one.

What we see is a lack of a real clear direction for health system change. I think we see some important examples of what I would call knee-jerk policy making, making of policy whether it involves the financing or the delivery system or the regulatory systems that, frankly, isn’t well informed and that responds to the emergency of the day or the need to contain costs rapidly.

We also see some real conflict and mistrust between key stakeholders, and these are the kind of people who really have to work together if the health system is going to move forward in a constructive direction. That’s sort of the bad side or the negative of what we’ve heard today.

I think the whole national experiment with managed care—or maybe I should say "managed care lite" because that’s really what it has been—has been rather disappointing, and I am not optimistic that it will get better, especially as "managed care lite" gets even lighter in the coming years, given the response to the backlash.

I think part of that is because these various managed care arrangements have pursued what is much more of a stick approach than a carrot approach. Oftentimes, they were imposed on clinicians and on patients. Both constituencies very quickly concluded that they had more to do with controlling costs than with improving quality. In many cases, they were right—not all, but in many.

The kind of approaches that were used, the tools that were used by managed care organizations, whether selective contracting or prior authorization of referrals to specialists or utilization review, are blunt tools, and they are ones that are usually perceived quite negatively by consumers and physicians. They impose burden, and in the case of physicians, there’s a very real concern about loss of clinical control and a very real concern about perceived immediate threat to income.

In the case of patients, I think they oftentimes disrupted established relationships with clinicians and raised a lot of concerns about the quality of care.

We have also, during the last 10 years, moved through this period of quite significant change in an environment that is devoid of information on quality. In most communities, and this is still the case today, policy makers, regulators, group purchasers or consumers, have virtually no information to discern whether managed care that is provided is of equal, lesser or better quality than the kind of care that is provided in traditional indemnity environments.

They have very little information to understand whether or not the providers in a network are providing equal, better or poorer quality of care than those providers who are not in the network. And I dare say, in most communities, I think the consumer, the end user here, the one that we’re trying to serve, does not have even a scrap of useful quality information to use when they select physicians, hospitals, or choose between treatment alternatives and plans.

There are really two lessons that I hope that we can learn from the last 5, 7, 8, if not, 10 years. The first is that to achieve real widespread change in the health care sector, the metric must include quality, in addition to cost. In the absence of information on clinical quality of care, consumers and others will always assume that the steps that are being taken to control costs are detrimental to quality, and I might add we don’t know that this is necessarily the case. But, intuitively, many of us—most people—believe that there is an inverse relationship between cost and quality.

Now, if we take a look at the evidence base that’s been developing on the quality of care in the last decade, and I’m speaking specifically about a comprehensive review of clinical quality conducted by colleagues at Rand, Mark Schuster, Beth McGlinn, and Bob Brook and others, they essentially reviewed about 80 leading publications that were published in the last decade in peer-review journals on the technical quality of care, and these were major publications that involved numerous settings or several communities, so fairly large scale.

What we found from that information is that the quality of care is highly variable in all settings. It’s highly variable in managed care settings, it’s highly variable in indemnity settings. The quality and safety problems that we see are systemic. They’re in hospital environments, they’re in ambulatory environments, they’re in acute care and long-term care. They pervade the entire health care delivery system. They don’t appear to be related to a particular delivery system or financing approach.

The other thing that is important to keep in mind is that there is a sizable body of literature that now substantiates that for some types of quality shortcomings, we can actually decrease costs and improve quality. That is not the case for all. But when it comes to medical errors and overuse, two of three major categories of quality problems, we really expect to see that improvements in quality will lead to slower increases in the rate of growth of health care expenditures.

In the case of medical errors, we know that when you make a medical error, in most instances, the patient survives, fortunately, and the way that you treat that patient is with more drugs, more laboratory tests. They typically have several days’ longer length of stay if the error occurred in a hospital and higher readmission rates.

So one thing, when it comes to the medical errors, we do know that we probably can have a positive impact on both cost and quality. We also know that that’s probably the case for the other category of quality issues; namely, overuse. Overuse pertains to services that are provided to patients that expose them to more potential risk than potential benefit.

And the IOM’s quality roundtable concluded in its review about 2 years ago—it was released in JAMA—that about 30 percent of health care services represent overuse. They are a quality problem, as well as a cost problem.

In the absence of good metrics on quality, policy makers, purchasers, regulators, and others simply cannot sort out whether patient and provider dissatisfaction really stems from concerns about clinical quality or whether it stems from concerns about changes in professional roles, status, practice.

A much more sophisticated quality infrastructure is absolutely critical not only to pursuing market-based approaches, but it’s also critical to having effective regulatory oversight processes.

The second lesson that I hope we learned from the most recent sort of experiment with managed care is that not only must quality be a part of the change metric, but the change strategy must address fundamental overhaul of the delivery system.

We have grossly underestimated the challenge of delivering high-quality care. Critical decision making is complex, and it’s growing even more complex over time. The current delivery system is wholly inadequate to provide high-quality care.

Now, as we have moved through this experiment in the last 10 years, I think it is important to emphasize that, for the most part, it did not involve significant redesign of care processes. It did not involve building an information infrastructure to support care delivery. It did not involve altering relationships among clinicians moving towards multidisciplinary teams that are so critical to providing good care to the chronically ill.

Indeed, many providers, purchasers, consumers and others may feel that they have been through a very tumultuous period of change, but in many ways, I think we really just scratched the surface, and the real period of change is the one that is yet to come.

Do we have any reason to be optimistic about the direction we’re going? Well, in the short run, probably not. As we shift more decisions and responsibility to consumers, we have to keep in mind that we’re doing it in an environment that gives them virtually no useful information upon which to make decisions that would help to drive the health system in a constructive direction.

Second, I think we’ve heard about the cost and advancement of medical technology and the potential of a hospital arms race. I think we can assume that in the short run a good deal of the available capital that could be used to building stronger organizational supports for care delivery will probably be used to invest in new medical technology, new clinical services.

Unfortunately, that means there will be less that go into building the clinical information infrastructure that’s so sorely needed to be able to really improve care delivery.

I hope that some of the earlier suggestions today about a possible Hill-Burton, that we carefully think about those, because they may offer some of the best promise. After all, we did put up the dollars for bricks and mortar, and there’s probably no reason we can’t put it up for bits and mortar at this point in time. But it is a policy option that warrants very careful analysis because it might hold some promise.

The other reason that we probably don’t have reason to be too optimistic in the short run is this issue of mistrust and real animosity that I think has built up at the community level. And it’s going to make it hard to get some of those partnerships and collaborative efforts underway that are really needed when you start to think about restructuring the delivery system.

Over the long run, though, I think we should be more optimistic. In short, I don’t think the current situation can go on indefinitely. Indeed, if we don’t take some action, probably the situation is going to grow worse.

There are two major trends, I think, that are likely to put enormous pressure on the delivery system to respond over the next five to ten years. The first is very rapid growth in the science base. If we look at just the area of biomedical literature, there are currently almost 12 million citations in Medline, and it’s growing at a rate of about 400,000 new articles per year.

Now, to put this in a little bit of perspective, if we look back at the mid-’60s, there were about 100 articles being published per year that reported the results of randomized controlled trials. At this point in time, the number is now over 10,000 per year. So we’ve had an enormous increase in the medical knowledge base, and one that continues to grow and will likely grow even more given the investments that we’re making in biomedical research.

Now, that doesn’t even touch all of the changes that are taking place in the molecular life sciences area. Indeed, I like a comment that was recently made by Don Masys, who’s the Director of Biomedical Informatics at the University of California at San Diego. He said, "Current practice depends upon the clinical decisionmaking capacity of individual practitioners for classes of problems that routinely exceed the bounds of unaided human cognition."

The other trend, I think, that will place enormous pressure on the health care delivery system and force us to think about more fundamental change is the growing need for chronic care. About two-thirds of health care resources are now expended on care of the chronically ill, and that proportion will likely grow, but the pie will likely grow in size as well. The numbers of people in need of chronic care increase as the population ages, as it lives longer.

It is important to note that chronic care, although it is complex, it is a real delivery system challenge, there’s only about 15 to 20 conditions that account for the bulk of resources that are expended. So when we think about some of the earlier comments about disease management programs and that they’re having slow but steady growth, I think in part that’s because they are very responsive to the needs of the population and likely are the beginning of some real restructuring at the care delivery level that hopefully we’ll see a lot more of over the coming decade.

Chronic care requires a very different delivery system from the one that we have now. Our current system delivers care in predominantly two settings: the hospital and the physician’s office in face-to-face visits. And it does relatively independently, not as a part of a well-organized system of care. The patient’s clinical information is often unavailable. It’s unavailable to the providers who are delivering the care. It’s certainly unavailable to the patient themselves and their family, who deliver the bulk of services to the chronically ill.

Chronic conditions require a high degree of ongoing communication among clinicians in a multidisciplinary team approach and a lot of communication between the clinicians and the patient and their family.

Good chronic care is not a string of more tightly packed acute encounters. It’s a very different delivery system from the one that we currently have, and it will require very different financing mechanisms and benefit structures.

The second reason, though, that I think that there’s reason to be optimistic about the long term here is because I think we now understand the problems of our health system a lot better than we did even a few years ago. The evidence of safety and quality problems is sizable, and it’s growing more and more every week. Most of us don’t even question anymore whether these are serious issues to be dealt with. The lay public also understands that these are important issues to be dealt with.

And even though it’s been a year and a half or so since the IOM’s report "To Err Is Human" was released, we still see on a weekly basis articles in newspapers citing problems related to medical errors. This is one issue that just hasn’t died, and indeed, I think the quality and safety problems are issues where the horse is out of the barn. They’re not going to go away. More and more people are aware of them, not only health care professionals but the lay public as well.

The third reason I think that there’s a good chance that the future will be a bright one—and it won’t be the too-distant future—is that we’re also beginning to see an evidence base develop which more directly links the characteristics of the delivery system to the level of safety and quality. And this is no small thanks, I think, to some of the leadership that John Eisenberg has so skillfully exercised at AHRQ, where they have really started to develop a research agenda that contributes to this knowledge base.

Let me give you an example. If we look at some of the pioneering work of David Bates at Brigham & Women’s, he has in the last five years published three, four, maybe five leading articles on automated medication order entry systems. And what you have seen at that institution is where you have developed a good, solid, automated clinical database, automated clinical record, so most of the clinical information is available in automated form, and on top of that, you have seen better and better decision support logic that can be used in automated medication order entry. And it’s now at a point where they’re reporting that they can achieve about an 80 percent decrease in adverse drug events as a result of those, what I call organizational supports, system supports to the clinician.

How long is it going to be before policymakers, regulators, group purchasers, consumers, and health professionals begin to say that it’s simply unacceptable to be in an environment that doesn’t provide many of these organizational supports, whether that’s a hospital or it’s a large group practice or a small physician’s office? It’s becoming increasingly clear, at least in some areas, that certain organizational supports are needed to achieve a minimal level of quality, and what we may see over the coming years is that that floor of minimum acceptable quality will rise.

So, in summary, I think it’s really quite clear that tinkering with the current delivery and financing systems will not do the trick. We now know that we must make fundamental changes in the health care sector, and those kind of fundamental changes are ones that just don’t happen overnight. And it’s not at all surprising that, as we try to bump up against some of those changes, we’re confronting in all of these communities very, very strong resistance from established interest groups.

The change of this magnitude involves a lot of uncertainty and is going to be disconcerting to many different groups. But once the ball gets rolling—and some of us believe that that will happen pretty soon—some very significant change can likely happen quite quickly. And, indeed, I hope, Paul, that when you pull this group together in about nine years from now and you take a look at what the first decade of the 21st century has brought, that the message you’ll be delivering to us will be a very different one from the message that we’ve heard today, and that we’ll be able to look around and find all kinds of innovative delivery system models, ones that we didn’t even think were possible today, that we’ll be able to see much more extensive use of e-health delivery options that have a lot of potential to improve quality, to improve access, and to reduce costs, and that we’ll all be working and consuming health care services in a very information-rich environment, an environment that provides active supports to patients and their families as well as to clinicians, to make decisions about their treatment, and an environment that begins to provide the necessary information on quality that policymakers, regulators at both the federal and the state level really need to do their job well.

Thank you.

[Applause.]

PAUL GINSBURG: Thank you. That was terrific. We probably should just do a few minutes of questions.

JANET CORRIGAN: Okay.

PAUL GINSBURG: Would anyone like to go to one of the microphones?

[No response.]

PAUL GINSBURG: Okay. Well, thank you again.

JANET CORRIGAN: Great. You made my job easy.

PAUL GINSBURG: That’s right.

Next Section
Back to Index