Interview with Arthur Levin

Hedrick Smith: We’re talking to Arthur Levin, director of the Center for Medical Consumers. The first and most basic question is what do American health consumers know about quality of care?

Arthur Levin: I think American consumers know almost nothing about the quality of care that they get. They haven’t been privy to information about outcomes or performance. Medicine has been a very secret profession over the years.

Hedrick Smith: Why?

Arthur Levin: Well, I think the attitude was that doctor knows best, sort of professional elitism, and that consumers couldn’t possibly understand the intricacies of medicine. And the best thing to do was sort of trust your doctor and do whatever he or she told you.

Hedrick Smith: What do we need to know as medical consumers?

Arthur Levin: The first thing we need to know is this tremendous variation in medical practice both as to process and to outcome. I don’t think most consumers understand how important it is that they research not only the condition they may have but who’s going to take care of them.

Hedrick Smith: How does the health system compare say with issues of airline safety, traffic safety, product safety and so forth, anywhere where life and death is involved? What do we know as consumers in health as opposed to travel or using cars or something else?

Arthur Levin: I think the American consumer knows less about the safety of the medical care system than they know about the safety of any other sector of our economy. There’s just no comparison; there’s no industry that’s been able to be as secret over the years as health care has.

Hedrick Smith: How do you account for that?

Arthur Levin: I think somehow the public wasn’t concerned that there was an issue of safety involved in health care. I don’t think the public was aware that there was such variation in outcome and performance. I don’t think the public ever demanded the kind of accountability from medicine that it demanded from airlines or the auto industry, for example.

Hedrick Smith: How serious is the problem with safety?

Arthur Levin: The problem of safety is enormous. The Institute of Medicine reported, just this past November, that medical errors kill between forty-four thousand and ninety-eight thousand people a year in the United States. That means at the low end, forty-four thousand, it’s the eighth leading cause of preventable death in the United States. That’s a significant public health problem.

Hedrick Smith: When we’re talking about medical error, what are you talking about? Doctors’ offices? Hospitals? And what kind of error?

Arthur Levin: We’re really talking about what happens in hospitals because that’s all that we measure at present. We have really no idea what goes on in doctors’ offices [or] in other outpatient settings. We believe that it’s at least as significant there as in hospitals. But the numbers we’re talking about are errors that occur to patients who are in the hospital.

Hedrick Smith: In terms of the way medical profession itself looks at quality, how do its measures of quality compare with what the public is talking about?

Arthur Levin: I think for many consumers quality is sort of a service quality issue. How was I treated when I called up for an appointment? Was the office nicely furnished? Did I have to wait a long time to see the doctor? Was there parking? Those kinds of issues are important issues — issues of convenience. But I think when people in medicine and health care research look at quality, they’re talking about the practice of medicine. Is it best practice? Is what the doctor is doing most likely to lead to a good outcome or not? I think most consumers don’t think about those issues at all.

What I think of when I think of quality is the variation in outcomes, the variation in performance. Is a patient going to survive their surgery? Is a patient going to return to normal function after their surgery? Is a patient with a chronic condition getting the best possible care so they can live the best possible life available to them? Those are the measures of quality that I think are the important ones.

Hedrick Smith: Now, let me ask you about the importance of New York State and its health reporting systems, specifically about open heart surgery. Is it important?

Arthur Levin: New York State has actually been a leader in many ways. It began some time ago to report on the outcomes of cardiac surgery in New York State and was the first state to do this and set an example for the rest of the country in what could be done in terms of getting information that was publicly accessible about the performance of surgeons and hospitals.

Hedrick Smith: Why is that important?

Arthur Levin: It is important to consumers. It’s important to providers of care to know how they do in comparison to others. If consumers are going to truly make an informed choice, "Am I going to have this cardiac procedure?" And then having made that decision, "Where am I going to have it and who’s going to be my surgeon?" They can’t possibly make an informed choice unless they have that kind of information.

Hedrick Smith: You mean there’s significant difference when New York published the statistics? Did it show significant differences?

Arthur Levin: When the Cardiac Advisory Committee, which was the instrument of doing this research in New York State, began to look at outcomes they saw what everybody knew existed — wide variation between hospitals and how well they were doing with this relatively new operation. And over time they were able to observe that the mortality rates were going down. They began to publish the data hospital by hospital. And low and behold the rates went down even more significantly. And finally they began to publish the data surgeon by surgeon and saw even a greater drop in operative mortality rates which meant that the procedure was safer in New York State when we were publishing data about results than before we started.

Hedrick Smith: I just want to get this straight from a consumer’s viewpoint, you mean if I decide to have open-heart surgery and if I pick Hospital A as opposed to Hospital B, if I pick Surgeon X as opposed to Surgeon Y, my chances of living and dying are different? When you talk about mortality rates, is this what you’re talking about?

Arthur Levin: What we’re talking about is your odds of surviving the surgery depending on which hospital and, probably even more importantly sometimes, which surgeon does the procedure. There can be a significant difference depending on which hospital or which surgeon does the procedure. And that’s a risk that you can avoid if you know who’s doing a good job and who isn’t.

Hedrick Smith: Now these are all, you’re talking here all about approved surgeons? They’re all certified, we’re not talking about some flight-by-night operation?

Arthur Levin: What we’re talking about here is surgeons that are sort of the elite of their class. Heart surgery is sort of like the world series of surgery. And I think what you’re seeing are highly trained people working in centers of expertise. In New York State particularly where we limited the number of facilities, the number of hospitals that were permitted to do open-heart surgery — we only have twenty-nine or thirty hospitals that are allowed to do this surgery. This was supposed to be a concentration of expertise and excellence. Yet, even in this small number of hospitals -- relatively small number of surgeons, we saw tremendous variation in performance measured by whether patients survive the operation or not.

Hedrick Smith: What was the reaction of the medical profession? How do the hospitals and the doctors react to this idea of giving them report cards and making them public?

Arthur Levin: I don’t [think] they reacted well at all. Again, this was sort of two-stage. The first stage was making available to the public identifiable hospital outcome for cardiac surgery. Hospitals responded with great venom about this system which they felt was unfair. They felt the risk adjustment was not adequate to adjust for differences in patient severity -- what we call case mix -- that is, how many very sick people you have compared to [how] many not so sick people you have. And the people who did badly complain that the data was not good data. The people who did well actually liked the reporting.

Hedrick Smith: Whose idea was it? Where did this come from? Did it come from the medical profession?

Arthur Levin: I think the idea for generating performance measures that were available to the public and were identifiable came from the New York State Health Department and its former commissioner, David Axelrod, who believed that information was empowering to consumers and believed that the health care system had to be accountable to the public it served and to the State, which was acting as the protector of that public. That was a very unpopular notion in those days, it probably still is today, that health care should be accountable to the people it serves.

Hedrick Smith: It was not an immediate decision to make this information public, was it? I mean, there was a battle to get this information out, wasn’t there?

Arthur Levin: In the beginning the Cardiac Surgery Advisory Committee was really looking at this data sort of amongst themselves. They would put out an annual report but it did not identify the hospitals by name. It talked about various cardiac surgeries, it talked about what the average mortality rates were, and then it identified by A, B, and C the variation in, in those rates. But it didn’t actually identify the hospital by name. It took a lawsuit by some New York media to force the Health Department and the Cardiac Advisory Committee to begin to release that information with identifiable hospital names to the public. And that’s how it got out there. It was a tremendous battle, and both the state and the medical profession did not go willingly down this path.

Hedrick Smith: Why not?

Arthur Levin: I think the history of medicine has been secrecy. I think the dirty little secret of medicine, maybe not so little, has been one, that there’s a lot of error and two, that there’s tremendous variation even without error and [in] how well people do in treating various conditions. I don’t think anybody wanted to talk about that publicly. Because if you kept it secret, there was no reason for people to make a choice of hospital or physician based on fact; they would do [it] through referral or other things which really didn’t add up to an informed choice.

Hedrick Smith: How did the medical profession react? Was this information that just came out easily from the New York State Health Commissioner?

Arthur Levin: Actually not. For a number of decades the Cardiac Advisory Committee looked at this information but it was blind. It identified hospitals as A, B, C, D, E, and F, and even though they would report back through individual institutions about their results, those institutions didn’t know what hospital, who Hospital F was or Hospital B was or C was. So it was blind for a number of years. Then under Axelrod’s leadership I think the Department made a decision to make that information publicly available, identifying the hospitals by name. The next major step, which was identifying the surgeons by name, took a lawsuit by Newsday to get it released by the Health Department. Over the years it was a very reluctant move from confidential information kept in the belly of the Health Department to information that was publicly accessible to everyone.

Hedrick Smith: Why do you think that was?

Arthur Levin: I think the medical profession doesn’t like to admit that there’s variation in excellence and skills, in outcomes, in performance. They certainly don’t like to admit that they make errors. And yet we know that’s a major problem in medicine. I think it was convenient to not let the public have access to this kind of information because then the public would never demand more accountability from health care.

Hedrick Smith: What do you say to doctors who say the figures don’t account for high risk and low risk patients; they’re not fair to us, to me?

Arthur Levin: Well I think what the response to doctors who complain about the risk adjustment is that in New York State we’ve had this methodology out there for over a decade. It’s been well-vetted in peer review medical journals. We’re never going to have a perfect system. This is probably as good as it gets. And I think it’s just time to get over it and realize that this is a good system of risk adjustment. It can’t risk adjust for everything, but let’s move forward.

Hedrick Smith: Who complains and who doesn’t complain?

Arthur Levin: What’s interesting to me in terms of physician and hospital complaints about publishing data like cardiac surgery data is that people who do very well in these reports don’t seem to have a problem with the risk adjustment or the publication of the information. It’s people who don’t do well who always seem to find fault with either the risk adjustment or the concept that this is information which the public should have access to.

Hedrick Smith: Have you seen any changes, either in the attitudes of the State or in the attitudes in the hospitals and doctors, toward publishing [data collected by] a commission?

Arthur Levin: Since the Institute of Medicine Report on Errors I’ve seen a lot of changes both on the part of the State and on part of hospitals and doctors -- at least a willingness to admit that there’s a significant problem with error and to say that they need to do something about it. The Commissioner of Health in New York State has said that New York State will meet the Institute of Medicine goal of a fifty-percent reduction in errors by the year 2005. I think that’s a big step forward.

Hedrick Smith: Let me just ask you, people differentiate the studies of medical errors and the statistical reports on performance, for example, of open heart surgery by hospitals and doctors, but do those statistical reports on open heart surgery that show the results, do they embrace medical error? Are medical errors embedded in those findings?

Arthur Levin: I think it would be hard to argue that medical error isn’t part of performance reporting. Some of the bad outcomes that occur in a performance report that are visible because we’ve compiled the data and made it public are certainly due to medical error. So I think to argue that these are two separate and distinct things is wrong; they are different measures of, of different kinds of problems in the health care system but they’re interrelated.

Hedrick Smith: So the implication is once you have a risk-adjusted mortality rate, you’re really looking at error or less than optimal performance?

Arthur Levin: I think you’re looking at less than optimal performance. We don’t know if it’s all error. But we know if Doctor A can achieve a really low operative mortality, there’s no reason to believe that all the other doctors can’t achieve the same low mortality rate with their patients in a risk adjusted system.

Hedrick Smith: Now one of the things that’s interesting is your talking about doctors at one point and then hospitals about another. People tend to talk about and think of heart surgery in terms of: I’m picking this surgeon. Why are hospitals important? Why are there variations within a hospital? Or why is there variations from one hospital to another? You see in some instances the same surgeon is operating in two or three different hospitals and the mortality rates, the adjusted mortality rates, are different. Why? What’s going on?

Arthur Levin: Cardiac surgery is a complicated bit of surgery and it requires a lot of preoperative and postoperative care and management. So the surgeon’s skills may be very important, but if the institution is not very good at taking care of cardiac patients, that surgeon, as good as he or she is, may get bad results.

Hedrick Smith: One of the arguments you hear made is consumers don’t pay any attention to the statistics. Does that matter?

Arthur Levin: I don’t think it matters at all. I don’t care if not even one consumer ever picks up one of these reports and uses it to make a choice. The fact is it changes the behavior of surgeons and hospitals and that’s what we want to have happen.

Hedrick Smith: Why don’t consumers look at it? I mean if they don’t.

Arthur Levin: Well they’re actually studies that show in Pennsylvania, for example, which also now publishes the same kind of data that New York State does, that consumers don’t use the information. I think there are a lot of reasons. I think people are not used to having this kind of information available to them, they really don’t know what it means. And the way they’re referred to surgeons and hospitals sort of mitigates against them using reports like this to make their choices. And their health insurance may limit the amount of choices they have.

Hedrick Smith: How do people pick their heart surgeons?

Arthur Levin: What happens is somebody has a symptom. They go to their primary care doctor. The primary care doctor refers them to their cardiologist that they always use. The cardiologist decides they need surgery and refers them to the surgeon that he or she always uses. The patient really doesn’t get to make any choice in this process.

Hedrick Smith: You get some doctors and some hospitals that say this data is simply not good enough and it’s not fair to me in my, the risk load of my patients. What’s your response to that?

Arthur Levin: What I say to doctors who continue to complain about this information being made public is it’s as good as we’re ever going to get. It takes us a lot of time to even do this. Everyone agrees that consumers and purchasers of health care have to have information on performance. It’s time to move on and do this for other procedures and stop sort of resisting even doing this for cardiac surgery.

Hedrick Smith: What do you say to people who are very interested in health care improvement, quality improvement who say this is the wrong way to go? The best way to get health care improvement is to get the medical profession together to cooperate, let them use the data, but privately.

Arthur Levin: When I’m told by doctors or others that this is the wrong way to go, that publishing data actually will hurt quality rather than improve it, and that the thing to do is to make this information available to doctors and hospitals and let them work on improving quality, I just say that’s not been the history of medicine. What got us to this place in the first instance? Error was kept secret; variation in performance was kept secret. I don’t believe we’ve had any indication that medicine, hospitals, doctors are really willing to do much to improve quality unless they’re publicly accountable, unless the data is published.

Hedrick Smith: Do you see the medical profession getting more open about sharing this quality performance information with the public?

Arthur Levin: I think there’s changes at the margins, but I think for the most part, we see in organized medicine -- in resistance to many of the recommendations of the Institute of Medicine’s Errors Report -- that organized medicine -- medicine as a group, hospitals as an industry -- are still going to resist public disclosure. They don’t see it as in their best interest, and they’re telling us once again leave it to us, we’ll take care of the problem, trust us. I don’t think there’s any reason to trust them.

Hedrick Smith: When we confront a hospital and we tell them we want to talk about this, and they resist, is there a signal there for consumers?

Arthur Levin: I think when hospitals resist being open with consumers, with the press, I think what they’re saying to us is they have a secret to hide. And I think it tells us that they’re afraid to be open. Now whether that’s simply the fact that because the next hospital, the next doctor isn’t being open, they’re afraid they’ll look bad in comparison, I’m not sure. But I think it tells us that they’re still fearful of being open and having the public have access to their information.

Hedrick Smith: Are you familiar with the Northern New England Collaborative, where the six different hospitals got together and have been working collaboratively?

Arthur Levin: I know a little bit about that.

Hedrick Smith: Well, their argument is that they’ve been able to achieve significant reductions in open-heart surgery over a period of pretty close to a decade now. And they’ve done it privately. What’s your response to that?

Arthur Levin: I think for those few places in the country where either individual systems or hospitals or groups of hospitals have gotten together to work on the quality problem and actually made progress without the lever of public disclosure, I think that’s all well and good. But that’s far from universal. Also, I think we live in a democracy and I believe this information which is collected with public taxpayer money should be available to the public that pays for it. So I applaud the efforts of those institutions and groups of institutions that are working together to make things better. I don’t think that justifies closing off public accountability.

Hedrick Smith: Is there an important role for government to play?

Arthur Levin: I believe there’s an important role for government to play in both collecting the information, doing the analysis, and reporting it to the public. I don’t know of anyone else who is better suited to do this, who’s more independent and more trustworthy than government. Why? Because they’re accountable both to doctors, to hospitals, and to the consumer in terms of the quality of the work. The problem with internal fixings of what is wrong in the system is who is that accountable to? Here when government does it, doctors can review the information as to validity, hospitals can review the information as to its validity, and the public has a voice in the quality and, and how that information is presented to them.

Hedrick Smith: What’s the role of the public itself, of consumers in improvement of the quality of health care in America?

Arthur Levin: I think the role of the public is to begin to demand more accountability from their health care system — that is from their doctors and their hospitals — to begin to demand that this information be available to them about all kinds of procedures and conditions and to begin to demand that they need to be partners, full partners, in the decisions that they make about their health care and their health outcomes. And they can only do that as informed consumers with this kind of information.

Hedrick Smith: What advice would you have in terms of getting the best quality care?

Arthur Levin: I think if somebody wants to get the best quality of care possible, they first have to really educate themselves about the condition that they have. You really have to do some research about its diagnosis. What are the alternative kinds of ways of taking care of the problem? That makes you a knowledgeable consumer in the sense that you have a way to begin to judge the advice you’re getting from doctors and from other health care professionals. If you know nothing about heart disease, how in the world do you decide that heart surgery is the correct advice for you? You’ve got to have some basic understanding of what’s going on before you can really make informed, intelligent choices.

Hedrick Smith: Do you see much evidence that best practices and that the quality and improvement movement is growing, jumping from one region of the country to another?

Arthur Levin: I think what we see is [that] a lot of examples of interest in best practices and care that’s based on the clinical evidence is spreading but a lot of that’s in the journals and a lot of that’s among people who really have been working in this issue for a long time. I’m not really convinced that as yet this is a widespread adoption of sort of a best practices principle.

Hedrick Smith: And what do you think the resistance is? Why wouldn’t people want to do better if they could, particularly people who are in the, the business of curing illness?

Arthur Levin: I think doctors are educated and trained and socialized to be autonomous professions. What does that mean? That means I’m the doctor; I’m the one with the experience and knowledge, I make the decision. In a very complex system, that doesn’t work very well. People need to work in teams and they need help. I think there’s a resistance based on how doctors get trained and socialized on admitting that one doctor can’t know everything, and that there needs to be a very different approach to making decisions. It has to be based in the evidence, it has to be scientifically evidenced, and that there are best practices that may be different than what that doctor is doing at this time. And that that doctor should give up their autonomy and aspire to best practice.

Hedrick Smith: Is there any historical analogy or comparison that comes to mind when you look at where the medical profession is today on the issue of openness, error, quality improvement, comparative performance data?

Arthur Levin: I think one of the interesting things in the deliberations of the Institute of Medicine committee that published the report on errors was looking at what goes on in other industries and how much further ahead so many industries are beyond health care in terms of their openness and their willingness to deal with safety and quality issues in public. I think one example would be the nuclear power industry, which was very secretive and assured us that everything was fine until we had something called Three Mile Island, and then the issue of safety and quality became very important. The airline industry, of course, is another example. The airline industry can’t keep secrets when they hurt people. Planes fall out of sky. It’s on the news. There’s no secrecy. Again I think the major problem with health care has been the ability to bury its dead.