Dr. Robert Wachter, Editor, AHRQ WebM&M: What got you interested in the law and its interface with medicine?

Dr. Troyen Brennan: I knew that I wanted to combine social research with medicine. While I was in medical school, I sampled a number of different graduate programs at Yale, and found that the law school offered an interesting way to look at social policy and could teach me a slightly unique skill set. So during my second year of medical school, I applied to law school and then went back and forth between the schools for another 3 years.

RW: Was it always your vision to be an academician studying in both areas?

TB: Exactly. I was pretty settled on academics almost from the very beginning.

RW: When most people think about the interface between medicine and law, malpractice leaps to mind. It's obviously much broader than that. Did you think that malpractice was going to be a dominant issue in your work?

TB: Not necessarily. I was interested in a broader set of issues—law comes into contact with medical practice in a variety of ways. In fact, I was not that interested in medical malpractice, because it seemed like a fairly settled area. I was much more interested in the administrative side. Most of my early work had to do with things that were explicitly regulatory in nature, and I was also writing about medical ethics.

RW: From your unique vantage point, what is it that doctors don't understand about the legal world, and what don't lawyers understand about the medical world?

TB: My general view is that most lawyers who deal with medicine and the hospital industry have a clear view of how the fields operate and how different players work with one another within the organizations. On the other hand, I have a sense that to most doctors, and indeed many people in administration and health care, the law is just white noise. It can be deafening, but they can never quite make sense of it—a din rather than a symphony. And they feel like, in some ways, it's unfair and unnecessary. So, they don't take time really to understand it. The level of sophistication amongst lawyers dealing with medical issues is much higher than the level of sophistication of most doctors and health care administrators when it comes to legal issues.

RW: That's interesting. Any sense of why you think that is?

TB: It's the lawsuit mentality. I think that doctors and hospitals see themselves as chronic defendants, always in an adversarial situation. As a result, there's a strong emotional overlay that goes on with almost all kinds of issues that otherwise would be addressed in a more intellectual fashion.

RW: Looking back almost two decades later at the Harvard Medical Practice Study, first, is there anything you would have done differently, knowing how much impact it came to have? Second, talk a little bit about how it has been used and whether you think it's been used correctly or abused.

TB: Well, you know, it's interesting. We basically parroted a study that had reviewed 20,000 medical records for the California Medical Association, which was interested in exploring the possibility of a no-fault system for medical injury. The interesting thing about that study, like a lot of other things in quality of health care, was that the California Medical Association let the study drop. It was never publicized. Eventually it was published in summary form in the Western Journal of Medicine in 1978. So it was well known to people who were interested in injury prevention systems, but it never had any policy impact—even though it was one of the largest health services research projects ever undertaken. And why did they let it drop? Well, they were in the middle of a tort crisis in California, and coming up with numbers that suggested that nearly 5% of the people in a hospital were injured by their medical care would not be a good way to help bring about tort reform.

The same thing was true for us. We began to think about our study toward the end of the tort crisis of the mid-1980s. Most of our questions were motivated by trying to understand the impact of the medical malpractice system on both deterrence and compensation. So we were just as interested in following up on legal claims as we were in reviewing the medical records. Having said that, I think some of us realized right away that it was a very valuable effort in health services research. We published a number of studies from the original data set assessing risk factors for poor quality care. But that study, and that particular line of research, kind of went to sleep. Over the course of the 1990s, people didn't show a lot of interest in it.

Lucian Leape, who had joined us in the Medical Practice Study about the time we started the record review, published probably the most important article in the safety movement tying the ideas in the Medical Practice Study around adverse events to the notion of error, and linking that to engineering-based safety. And that was the appropriate metaphor in which to view these things. But it really wasn't until the IOM [Institute of Medicine] took the Medical Practice Study and another similar study we had done to verify it—the one in Utah and Colorado—and extrapolated those to population figures and then went directly to public in 2000, that people really began to get excited about funding research to understand patient safety.

RW: As you were participating in the study in the late 1980s, could you have envisioned that the 44,000 to 98,000 figure [estimate of annual deaths from medical errors in the US] would become this iconic number to drive the movement?

TB: Because we were interested in medical malpractice, we were always hesitant about that kind of extrapolation. Frankly, from the tort point of view, if you publicized that there was that much negligent injury, then there should be much more tort litigation. Right now all we have is a somewhat paradoxical, almost sort of untruthful or false, approach in which we try to reduce the amount of tort litigation on the one hand, and yet on the other hand acknowledge that there is tremendous morbidity and mortality associated with iatrogenic disease, about a quarter of which is caused by substandard care.

RW: When did you begin thinking that no-fault might be a better way to deal with medical injury? Was it an epiphany in seeing other systems or did it come to you slowly?

TB: It was almost from the very beginning. We supported no-fault back in the early 1990s, but the most important part of our argument was that only by getting rid of the moral blameworthiness part of medical malpractice and thinking more about prevention would it be possible to effect any change. Not only was it a no-fault program that we'd always advocated, but it was also an enterprise liability program in which the liability lies at the level of the enterprise, not with individuals. In our research (continuing until this day), about half the problems are systems problems, but the other half are individual culpability. It still seemed like the only way to really address those issues was to treat it mostly as a systems problem. And we felt like an administrative compensation scheme would be the way to do that.

RW: In medical thinking about the issue of patient safety, we've sort of gone from it's all about the individual to now, it's all about the system. As you say, both empirical data and observations say that both are right. There are some bad apples. There are some people who should not be practicing. Where is the sweet spot there? How do you get both of these paths moving in the right direction?

TB: Let's presume you and I are good doctors. Sometimes we run into a problem and, had we a system in place—such as an electronic medical record that reminded us of the patient's allergies—then the system would have saved the day. The lack of such a system is the problem. On the other hand, you have the individual who repetitively makes mistakes. And that's the so-called bad apple. In any system, you need a regulatory approach that addresses the bad apples; for the other physicians, you concentrate on the systems problems. As your systems get better, it will become harder and harder to make mistakes that otherwise might be considered to be individual errors.

RW: How much progress have you made in gaining traction in terms of no-fault at the political level?

TB: It's very difficult at the political level, I have to tell you, because there are very well-placed constituencies who really don't want change. There's obviously no interest in change on the part of the plaintiff's attorneys because, their role would be narrowed significantly in an administrative compensation scheme. As well, defense attorneys are not interested in it either. Most malpractice insurance companies are in a relatively stable situation right now, and I think are not interested in much change—certainly not radical change that might jeopardize their particular line of business. But also from the point of view of being conservative institutions and doing the responsible thing, I think they wonder if we move to a system based on no-fault or preventable injury, they're not sure how much it will cost to insure. They think like actuaries. What they really want to know is how much it will cost to insure. A change in system really makes it more difficult for them. Whereas if they stick with medical malpractice, no matter how many warts there are associated with it, they at least can predict from year to year what the premiums will be, and they can actually move forward in business.

The other thing is that most doctors have been pretty happy to do traditional tort reform, which sounds like something that's neutral but really it's just a matter of trying to make it more difficult for people to sue. I would go so far as to say that I find tort reform in that format to be unethical, because it really doesn't do anything to help with the deterrent effect of tort law. It does very little to help with prevention of medical injuries, and it does nothing to address the inaccuracy of the system. So you have lots of people who should be receiving some form of compensation for the iatrogenic injury they suffered, but who are receiving nothing.

RW: Is the practical argument that you will not be able to find a neurosurgeon or obstetrician, is that credible or is that hyperbole?

TB: There's probably some credibility to that. The credibility is always surrounded by a good deal of hyperbole. We've just published a couple of papers about defensive medicine in Pennsylvania, where there was a significant crisis. The physicians report that they are not doing things they used to. That they're considering moving and that they're not taking care of certain kinds of patients. The empirical data on rates and procedures does not reveal that is happening in Pennsylvania, at least not yet. So it's very hard to take clear stands on defensive medicine because the data associated with defensive medicine is never going to be that clear.

RW: Do you think there is a "malpractice crisis"? Or is that something that is manufactured?

TB: Well, malpractice crisis is a nice euphemism. It sounds, I think, worse than it is. When you break it right down, the malpractice crisis is just a rapid increase in the cost associated with one of the components of the physician's overhead. It's a threat to physician income, but it's not the kind of crisis like we would talk about a terrorist crisis or a crisis around an epidemic like HIV. I think the language is a bit histrionic. But it is clear that over the course of the last 5 to 8 years in a number of states, numerous insurance companies rapidly increased their malpractice premiums because of the high losses they were experiencing. What are the factors that could be causing that? Well, one factor is just that everybody—the patients, doctors, plaintiffs, and plaintiff's attorneys—are all now aware that there's a lot more injury in hospitals than they thought there was. That's one result of the IOM report. Another thing that clearly happened was that the malpractice insurance companies broke into new markets, either state-based companies moving into other states, or doctor-based companies moving into hospitals. They also became more aggressive in marketing their products, and they weren't careful about setting their premium appropriately. That is called cycling in the insurance industry. We probably were in a hard cycle in 2001 through 2003 or 2004. In many states today, the latter aspect is beginning to go away. Claims rates are now relatively steady, and I think that most companies are writing appropriate premiums at this point. Many doctors, while their premiums have risen significantly, are not seeing very hectic increases in 2005 and 2006, as they did in 2002 and 2003. So some aspects of the crisis are over. However, the real bottom line when you look at the empirical information is that the number of negligent injuries that occur in hospitals suggest that there could be five to eight times as much reasonable litigation as there is right now. A lot of medical injuries are caused by negligence. I don't think anyone feels like the system could accommodate a five- to eight-fold increase in malpractice premiums. So that's just a foundational problem. Until you begin to address that by reducing the administrative costs associated with the system, compensating the people who are most significantly injured, and appropriately triaging the rest of the injuries, then you won't have an accurate and well-functioning system.

RW: What do you think about the general issue of the business case for institutional and individual investment in patient safety, and what is the role of the malpractice system in moving that along?

TB: I do like the example of anesthesia. They definitely reduced the cost of their malpractice claims by adopting end-tidal CO₂ monitoring and continuous oximetry, as well as by participating in some of the team training programs. But we and a lot of other people are looking long and hard for other silver bullets associated with patient safety, and there aren't a lot of them out there.

From a business case point of view, most of the costs associated with medical injury are not being absorbed by hospitals in terms of their malpractice premiums. They're being absorbed by the health insurers. If the insurers and the employers were smart, they would search for a different approach that accurately identifies when these incidents are occurring and sets up reasonable enterprise-related, experience-rated insurance premiums for specific hospitals. That would give the hospitals a significant incentive to try to reduce their malpractice costs. There's no executive in health care today who is thinking, "I can really reduce my malpractice costs by undertaking significant patient safety initiatives." You undertake patient safety initiatives for a variety of other reasons, but it's not because right now there is an economic business case associated with it.