A blueprint for high-volume, high-quality lung cancer screening that is detecting cancer earlier—and helping to save lives
Donald Berwick, M.D., is president and chief executive officer of the Institute for Healthcare Improvement, a not-for-profit organization that works to improve the quality and value of health care. The goal, like the man seeking it, sounds low key, until its implications sink in.
IHI wants to transform the practice and business of medicine. Berwick believes that health care officials must follow the lead of executives in other industries who have pronounced the status quo dead by declaring, "The way we've done things up to now is over. There is no way we can get through this if we continue to do what we've been doing."
The institute finds and teaches best practices in aspects of clinical practice. Putting those practices together is the mission of a project called Idealized Design of Clinical Office Practice.
Berwick teaches pediatrics at Boston's Children's Hospital and he serves on the faculties of the Harvard Medical School and the Harvard School of Public Health.
From 1985 through 1989, he served as vice president for quality-of-care measurement at Harvard Community Health Plan, and before that, as associate director of the Institute for Health Research, a joint venture of the HMO and Harvard University.
Berwick is a 1968 summa cum laude graduate of Harvard College. He earned his medical degree from Harvard Medical School in 1972, and his Master of Public Policy from Harvard's John F. Kennedy School of Government the same year. He spoke recently with Senior Contributing Editor Patrick Mullen.
MANAGED CARE: How did the Institute for Healthcare Improvement come into being and what were the goals at the outset?
DONALD M. BERWICK, M.D., M.P.P.: The Institute for Healthcare Improvement was formed in 1991. It followed four years of grant-supported demonstration effort and something called the National Demonstration Project on Quality Improvement in Healthcare. NDP had been funded by the John A. Hartford Foundation and was an initial attempt to test the value of what we then called industrial quality control or quality improvement, and not just in health care settings. We found methods developed outside and inside health care, but the applications we tested were only in health care. I was the principal investigator. We put nonhealth care quality experts in health care organizations on a volunteer basis. It was successful, and in 1991, the foundation came back to me and offered us an anchor grant to continue the work by setting up a freestanding not-for-profit organization. Our mission was originally to accelerate improvement in health care systems in the United States and Canada. In 1997, we removed the geographic restriction and began work in Europe and other places.
MC: Is industrial quality control the same as total quality management or continuous quality improvement?
BERWICK: Yes, although I no longer use any of those terms. I mean modern systems thinking and approaches to improving systems.
MC: Is there still a strong flavor of trying to bring to health care approaches that are being used in other industries?
BERWICK: Yes, although over the past decade work on improving health care organizations has increased to the point now where there are a tremendous number of excellent reference points within health care. You can understand most of what you need to within the system.
MC: You have talked about some of the lessons that health care can take from the way NASA went about its business in the 1960s. What are those lessons for people trying to improve health care?
BERWICK: It's a very tough question. I don't want to be elusive, but in a way the answer to that question is my life's work. There are two core notions that only recently, I think, are being understood at the level we need to. First is the idea that performance is a system property. If you want improved performance, you have to seek a new system. The concepts of improvement and system change are related inextricably. That's a deep idea and it guides activity from then on if you really believe it. The alternatives are exhortation, incentives, goal setting, or lots of other things that aren't really systems-minded. The second core idea is that improvement itself requires a system. You don't get improvement by hoping for it.
MC: How does the work of improving health care differ from improving a manufacturing process?
BERWICK: It doesn't always differ. A lot of health care is quite like manufacturing, especially when we're doing something highly repetitive. There are, however, some recurring differences. The products and services in health care are more complicated than those of other industries — not a lot more complicated, but enough that it makes a difference. Also, there are many different products and services, so a hospital would be like a manufacturing plant that's making a thousand different things instead of 10 different things. Leadership structures are not so well developed in health as in the corporate world.
MC: Is that changing, as health systems evolve and consolidate?
BERWICK: Somewhat. Certainly financially it's becoming more centrally led, but there's still a major problem. A chief executive officer of a health system is likely to regard it as very difficult to deal with clinical matters because doctors are in charge of them. Doctors regard themselves as the professionals in charge of patients, while lay management will be stewards of capital and resources. Division instead of unity is highly prevalent.
MC: And physicians may feel that executives are looking at the bottom line — at the risk of patient care — while executives may feel doctors are indifferent to finances.
BERWICK: That's correct. Once the belief structure is there, people stop testing the belief. You end up with mythic beliefs reinforced by ritual behaviors. It's what Chris Argyris calls the ladder of inference [a hierarchy of increasing abstraction that goes from data to conclusion, often leading to misguided beliefs.] Sometimes, after you work your way up to the conclusion, you stay at that level, and never go back to check the data again. A lot of these beliefs are well structured into the system, even though they may not be true, and are at least disrespectful and difficult behaviors. Another difference between health care and the corporate world is the leadership provided by boards of directors. Health care boards, maybe because they come from a voluntary tradition, are somewhat less precise and less demanding, and probably less able in discharging system-stewardship responsibilities. A corporate board can be a little clearer about what it's responsible for: share value and profit.
MC: Are there any other key leadership obstacles to the work that you're trying to do?
BERWICK: Suppose you went to the senior executive suite at Boeing and you listened to what they're talking about. Many of them today would be talking about mergers and Wall Street and capital. But they'd also be talking about airplanes. The leadership is knowledgeable about it, and is cognizant of the actual core product. In health care, when you go into the executive suite, especially on the lay side, you often find that the conversations are not about care, they're not about the core product, they're about deals. Patients may not even be mentioned. That's a disabling characteristic.
MC: A project of the institute that's seeking to improve the core product is called the Idealized Design of Clinical Office Practices. What is it? How do you hope to bridge the gap between folks who are delivering primary care and the leadership of the organizations that they work in?
BERWICK: IHI's first projects were educational, so we have courses on improvement methods and courses on change and on statistics. We also have a large national meeting, our annual forum on quality improvement in health care that has grown very steadily. I think we had 2,600 people last year, and probably will have 3,000 in New Orleans this December. We also have a European meeting, which will attract over 1,000 people to Stockholm this year — our fourth European Forum. In 1995–96 we began what is currently our largest program, which is called the Breakthrough Series, a rolling series of topic-by-topic collaborations for organizations, usually in the United States. We picked topics where there's a strong scientific basis and a lot of opportunity for improvement. So the topic might be asthma care, back pain care, waits and delays, or errors in health care. We form a collaboration of maybe 30 to 50 organizations that will last about one year. They'll work very hard with the best experts we can find to achieve improvements on that topic. These topic-by-topic improvements are good, but we really need to get the full advantage of improvement on a much more comprehensive system level design. This is more than disease management or topic management, this is changing the whole system. We picked the clinical office practice as the system or the minisystem that we want to work on. We searched for the best innovations we could find in a dozen or 15 clinical office-based systems like records, evidence-based care, patient education, scheduling, hiring and firing of staff.
MC: How did you go about that? Who decides what's best?
BERWICK: It's performance based. Through the breakthrough series, we had stumbled on a number of unusual and intriguing and high performing cases. For example, we found Mark Murray, a doctor in Kaiser Permanente of Southern California, who had developed an approach to scheduling patients that is essentially wait-free. We call it second-generation scheduling. We worked with Richard Rockefeller at the Health Commons Institute and Charlie Burger in Bangor, Maine, who has a revolutionary approach to patient records. We have a mailing list of 40,000, so we're in touch with hundreds of organizations around the country and many more in Europe. I won't vouch that in every case we found the very best in the world....
MC: But you found excellent work being done.
BERWICK: Yes. These are outside the current envelope of performance, but they're not system-level. With the idealized design project, we combined these innovations, and made the faculty of about 15 of these innovators. Then we recruited 37 or 38 practice sites that have agreed to join this project for three years and to redesign their own practices with the faculty looking over their shoulders and coaching them.
MC: What are the goals of the project?
BERWICK: The four aims are to improve performance in health-status outcome, patient satisfaction, staff morale and satisfaction, and financial areas. We've been oversubscribed. A lot of sites are being sponsored, as one would expect, by supraordinate organizations. An integrated delivery system or HMO or some large entity is saying, "We've got 27 group practices in our system and we'll enroll these two as the prototype practices and learn about it. Then we'll spread those innovations system-wide." Dissemination internally in the multisite systems has always been a part of the design. When each of the sponsoring entities like Peace Health in Oregon or Mayo Clinic joined, we immediately began working with them on their internal dissemination plans. Public dissemination is also part of the design, through publications and national meetings and conferences.
MC: How much commonality do you expect in the way that physicians run practices, and how much of this project's work will evolve so that people do things uniquely in their own communities?
BERWICK: There are two levels to the answer. There is — and should be — a set of basic design principles which would be relatively invariant across settings. For example, I think it's a design principle for the 21st-century practice that patients make the decisions, that the practice exists to give people information, on the basis of which, they make choices about their care. That concept is somewhat present in health care today, but it's a fundamental principle for practices of the future. Another one might be use of technologies to get people to the medical literature fast so that they can benefit from the evidence instead of having to search around too hard. Another principle would be Mark Murray's second-generation access systems. I would suspect almost all practices in some form or another that are worth calling modern will use these principles.
MC: What kind of feedback mechanism will there be so that as more learning goes on around the country, it gets spread through the system as well?
BERWICK: At a national scale, I don't know. In organizations that are part of IDCOP, part of the future is performance measurement that's used for improvement, not for judgment. HEDIS measuring is okay for accountability, but we're talking about self-conscious learning systems in which self-knowledge is the aim.
MC: Let me read this quote from you that's on the institute's home page. "We envision a system of care in which those who give care can boast about their work and those who receive care can feel total trust and confidence in the care they are receiving." That's your goal, and this project provides a mechanism to try to work toward it. Yet it seems to be in stark contrast to the feelings of many patients and physicians today. Do you feel you're swimming upstream against a strong current of mistrust among patients toward the system and among doctors toward the financial system?
BERWICK: There is a very compelling and positive and inspiring vanguard of people who just are not letting up on productive change right now. That's where I get my energy. There are inspired organizations and leaders around the country whom I get to work with. They're great and it's fun to work with them and they'll do fine. They all face the pressures of the Balanced Budget Act and other stresses. A lot of places are losing their shirts right now. We're 10 years from the vision, but we're getting there. There also is an awful lot of demoralization right now. It's probably worse than I've ever seen it, and I don't know how long that's going to go on. Out of that terrain, people are emerging who are pretty fed up and are change-minded.
MC: What do you think is causing the demoralization you mentioned?
BERWICK: First, it's a dynamic in which, when stress rises, people — including leaders — become more invested in getting through the storm. That means preserving the status quo. It's a leadership paradox. It's exactly when you need to change what you're doing, that defending what you're doing seems to be the path people choose. The great stories of transformation of industry have always involved executives who basically stand on a table in the cafeteria and say, "The way we've done things up to now is over. There is no way we can get through this if we continue to do what we've been doing. It's a new game and we're starting again." This willingness to fundamentally redesign is the only way out. Demoralization happens to organizations stuck in the first mode. When you're defending the status quo, you're really stressing the status quo.
MC: Who do you think has caught on?
BERWICK: The Mayo Clinic has. They are going to make it and do great. I watch Bill Rupp, CEO of the Luther/ Midelfort Clinic in the Mayo System, and I am amazed. Clinic officials are stressed and they have ulcers, but man, they are not going to be in the future what they have been in the past. I just had a wonderful experience last week with Matt Hanley and Mike Stuart and their group at Group Health Cooperative of Puget Sound. The whole organization hasn't yet decided to change, but here is this cluster of doctors around Matt and Mike. I thought, they are creating a future and the institution they're in has the wisdom to support them. I see a few gems like those every month and I'm heartened. I think that they know what they need to do.
MC: It sounds as if you're making a leap of faith that the health system will get past the anger and confrontation that prevails now.
BERWICK: It's not quite faith. It's a matter of extrapolating the evidence. I've seen enough to convince me that this is possible.
MC: Do you see a return or resurgence of physician executives? They have a clinical approach and can bridge the gap that separates them from executives, despite the common mythology that doctors don't know anything about business.
BERWICK: That is mythology. Yes, I do. I don't think it's just physicians, I think it's clinicians. Pharmacists and therapists of other stripes are equally well positioned to do that. Many big systems are elevating clinicians to positions of unprecedented responsibility precisely to build that bridge. I continue to believe that heads-up lay executives can do a great job too. I don't think you have to be a doctor to lead health care, but it's promising and creates a new need. What's the new learning environment for emerging clinical executives so that they can be as good leaders as they are clinicians? We now have physician leaders who have been elevated to very high levels of corporate responsibility who are untrained in core processes of executive leadership.
MC: You served on President Clinton's Advisory Commission on Consumer Protection and Quality in the Healthcare Industry. How much of that was political posturing, and how much of it will result in useful improvements in the system?
BERWICK: I don't know yet. The commission was great, a wonderful group of people. I did trust and I do trust the commissioners, and Health and Human Services Secretary Donna Shalala and Labor Secretary Alexis Herman were clearly committed to making progress. However, like so much else in progressive public policy, it got enmeshed in the other events of the last year. The commission made a number of very crucial recommendations. Three were essential: Number one was a Patients Bill of Rights. I think that's OK. I believe the bill of rights is probably a pretty good idea.
MC: It doesn't sound like you feel like it's a huge solution.
BERWICK: Right. It's not a huge solution at all. The problem here is trust. There's considerable evidence of eroded trust. Individuals may trust their individual doctors, but they don't trust the system as a whole. Unless we get into the business of restoring trust as a primary objective, it's very hard to make progress. Promising the public things they want to be promised is probably a good idea and the bill of rights is a vocabulary of promise. My own view, by the way, is that the leadership of the managed care industry should stop fighting the Patients' Bill of Rights and just applaud it, welcome it, embrace it, and get beyond it.
MC: What were other key recommendations of the commission?
BERWICK: There was a recommendation for creation of two bodies, a Forum for Health Care Quality Measurement and Reporting, and an Advisory Council for Health Care Quality. The forum has gone ahead. There's a forum planning committee that is charged with designing this new public-private entity. Its job is to establish a consolidated public-private partnership and well-funded entity that will specify a set of dimensions and performance measurements that will become more or less standard in the country. The other entity, the advisory council, was supposed to be a public leadership entity that would be responsible for consolidating national will for improvement by designating and then updating a series of national aims. I think it would be great to have a public body say, "What about pain control?" But the National Advisory Council was rejected by the Republicans because of a very low taste right now for anything that smacks of regulation.
MC: Is it fair to ask what effect managed care has had on the quality of health care in the U.S.?
BERWICK: It's a definitional problem, as you know. Managed care has no definition. I have no idea what it means. Whenever you're in a conversation with someone about managed care, you'd better spend about half the conversation defining your terms. Otherwise you'll be lost. In the United States, we have a breed of well-led prepaid organized systems. The traditional ones are staff- and group-model HMOs, the Kaiser Permanentes and Harvard Pilgrim Health Care, Fallon, Group Health Cooperative. In our work in the institute, we are always looking for the best we can find. We search the country and the world for the best care of back pain we can find, the best asthma care, the best intensive care units or the best obstetrical management. It is exceedingly rare that, after picking a topic and searching hard, we don't end up with some of the famous and important managed care systems in the country on our list of the best we can find. I think it's no accident. I think they're the places that had the leadership and information and systems. The best of managed care is often the best we have. Many people don't seem to realize that. The worst is also represented in managed care. I've been horrified by some care I've seen in managed care when it isn't well led and it isn't truly integrated and doesn't invest in quality. So it's a pretty heterogeneous group. I guess we need a new name for managed care. Maybe we should go back to Paul Ellwood and say, "Paul, it's time to coin another term." There ought to be a name for a system of care that is patient-centered, innovative, prepaid (or at least globally budgeted), population-based, managed in the positive sense, meaning that it's integrated so that inpatient, outpatient, specialty, primary care, and home care are all centered in some hub.
MC: Did you come up with any?
BERWICK: No. I tried to. I thought we would call it "banana," just give it a name that has no connotation at all.
MC: And then let the meaning be poured into it?
BERWICK: Yes. There are some exceptions, but I still get thrilled when I go to some of the best of those systems.
MC: I thank you for your time.
A blueprint for high-volume, high-quality lung cancer screening that is detecting cancer earlier—and helping to save lives
Multiple Sclerosis: New Perspectives on the Patient Journey–2019 Update
Summary of an Actuarial Analysis and Report