The way to reduce errors in health care is to change systems, says this Harvard educator. Punishment encourages people to cover up.
Lucian L. Leape, M.D., is a health policy analyst whose research has focused on error prevention and appropriateness of care. Prior to joining the faculty at Harvard, he was professor of surgery at Tufts University School of Medicine and chief of pediatric surgery at the New England Medical Center. He is a leading advocate of the nonpunitive systems approach to preventing medical errors and has led several studies of adverse drug events and their underlying systems failures. In addition, he has directed research into overuse and underuse of cardiovascular procedures.
Leape was a founding member of the board of directors of the National Patient Safety Foundation and a member of the Institute of Medicine's Quality of Care in America Committee, which in 1999 released the report "To Err is Human: Building a Safer Health System." He also has served on the Agency for Healthcare Research and Quality (AHRQ) Health Services Research Review Committee and the Physician Payment Review Commission's Access Advisory Committee. A graduate of Cornell University and Harvard Medical School, Leape trained in surgery at Massachusetts General Hospital and at Boston Children's Hospital. He has written more than 180 medical articles, book chapters, and monographs. He spoke recently with Senior Contributing Editor Patrick Mullen.
MANAGED CARE: In its report "To Err Is Human," the Institute of Medicine asserted, "Health care is a decade or more behind other high-risk industries in its attention to ensuring basic safety." What caused that gap?
LUCIAN L. LEAPE, M.D.: There are two reasons: Until recently, health leaders didn't know how bad the situation was, and health care was locked into the wrong paradigm for ensuring safety. People seem to think this issue has been around forever, but the first research results were published just 10 years ago. The Medical Practice Study, the first large population-based study that tried to get a reliable fix on the extent of serious medical injury caused by treatment, came out in 1991. That study showed 3.7 percent of people had an adverse event or injury caused by treatment, nearly two thirds of which were caused by errors. But that was just a single study, so it didn't receive a lot of attention. It wasn't until 1995 when a series of events, starting with the Betsy Lehman case and other high-profile cases, brought more attention to the problem.
MC: She was the health reporter in Boston who died from an overdose of chemotherapy.
LEAPE: Right. Then somebody in Florida had the wrong leg amputated and somebody in Chicago had the wrong side of the brain operated on. These cases came out just about the time we started talking about a different way to look at safety. That raises the second point. Until recently, we in health care thought that we had an effective way to ensure safety. The concept was that if you're well-enough trained and careful enough, you won't make mistakes. If you do, we'll punish you and then you'll be more careful the next time. People had never really questioned that approach. Eventually that was called into question and we said, "Look, industries that are much safer than we are don't do it that way." The concept that errors are always with us but can be minimized by looking at systems rather than just focusing on punishing people who make mistakes was a brand-new idea in health care. That approach has been adopted only in the last six years, so I think we've moved very rapidly, all things considered, during that period.
MC: Why weren't errors measured in health care? Was it because it's such a diffused system? When an airliner crashes, it's big news. A medical error or a thousand medical errors spread across the country aren't as dramatic.
LEAPE: There are two aspects to that. One, the difference has never been clear between complications that are part of the disease, complications caused by the treatment, and complications caused by errors in treatment. We had to realize that there are certain complications that we couldn't do anything about, others related to our treatment, and still others specifically related to treatment that wasn't being done the best way we knew. Bringing that into focus was a new way to frame the issue. The other point you make is also relevant. Physicians very rarely have patients die as a result of a mistake. It's only when you add them all up that the numbers get scary. Of course, when an airplane goes down, the numbers are all added up.
MC: What's the biggest cultural obstacle to improving patient safety?
LEAPE: Many of us think that the punitive mind-set is the biggest obstacle that still exists in most health care institutions. It's very hard to overcome. The theory behind a nonpunitive approach is very straightforward: It's inappropriate to punish people for making mistakes because very few are due to misconduct. Errors are almost always caused by systems failures, and those are not under the control of the individual who makes the error. Punishing people is counterproductive, because if you punish people for making errors, they will report only the errors they can't hide. Several studies show that when there is a punitive environment, 95 percent or more of errors do not get reported. We also know that when the system changes, reporting goes up dramatically. We've seen that happen in a number of hospitals. If you're serious about safety, you need to know what's going on, and you're not going to find out what's going on if you punish people. The two cornerstones of safety are, one, creating an environment where it's safe for people to talk about their errors and, two, leadership.
MC: To motivate their leaders to focus on safety, some health systems are starting to incorporate patient-safety standards in their executive review and compensation plans. What do you think the potential might be for that?
LEAPE: I've heard several health care leaders recommend that. It certainly makes sense to me to put your money where your mouth is. If safety is important we should pay for it just like anything else. The problem is having good indicators, because it's not easy to measure safety. We do not measure safety. We measure failures of safety, and that measurement is undeniable because of the punitive environment that still prevails in most of our institutions. As long as we have that kind of environment, reported error rates are very unreliable as a measure of either where you are or what progress you're making. When we talk about compensating people according to performance and safety, we'd better have a very clear idea of how we're going to measure that performance, because it isn't easy.
MC: What effect has managed care, broadly defined, had for good — or ill — on patient safety?
LEAPE: From a theoretical standpoint, safety is a management problem. Safety results from good management of care, and errors result from poor management. So we need better-managed care. People mean many different things when they talk about managed care. Some large health care organizations, such as Kaiser Permanente and Allina, are clearly leaders in safety. The biggest managed care organization of all, the Veterans Health Administration, was an outstanding leader under Ken Kizer [M.D., M.P.H.,] and is continuing since he left. It has been a trailblazer in implementing practices that have been shown to make a difference. It was among the first to go systemwide to nonpunitive reporting, to implement computerized records, to use computerized order-entry and barcoding, and now it's working on a highly developed voluntary reporting system. These are not only good ideas, but the VA is doing them in 172 hospitals. It's impressive what the VA has accomplished. A number of managed care organizations of all kinds have shown leadership. I'm impressed, for example, with how HCA has made medication safety a major program at its hospitals. Big payers have also been interested through the Leapfrog initiative to push ahead with some technological improvements and directing patients who need complex procedures to hospitals that most frequently perform those procedures.
MC: How can practicing primary care physicians help move the process toward greater patient safety forward?
LEAPE: Safety is local. It has to happen at the point at which doctors and nurses and pharmacists and others interact with patients. The practicing physician is crucial to the effort, and if the practicing physician isn't safe, health care is not safe. Without physician involvement nothing important is going to happen. That's true in a hospital, an academic medical center, a clinic, or an office. What we are striving for, and we're beginning to see in all kinds of settings, is doctors beginning to question and evaluate their practices. They're asking how they can work in ways that make it less likely to lead to mistakes. The profound concept, one that the Institute of Medicine enunciated loud and clear but that is still hard for people to understand fully, is that safety is a systems problem. When individuals make mistakes, it's because they've been set up to make mistakes. They've been put in situations where they're more likely to make mistakes and they're put in situations where it's hard for them to recognize a mistake and head it off before it causes injury. That idea is a very powerful transforming concept in terms of the way we look at everything we do. As doctors have applied that in their everyday work, we've come up with a variety of ways of changing what they do to make them safer. I'm referring to such simple things as having a good system to make sure that no laboratory report gets lost; that when a lab test is done, the doctor has the answer promptly, doesn't have to worry about remembering it, and gets the information back to the patient. A lot of it is simple stuff, but it's important and it's the kind of thing that only doctors can do at their own practice levels.
MC: How would you assuage the fears of physicians who worry that if they're more open about the mistakes they make, they're setting themselves up for malpractice suits?
LEAPE: The physicians who are moving ahead and improving their systems and making progress are those that have been able to get beyond the worry about malpractice. A common thing I hear from doctors is that they can't talk about their mistakes because it exposes them to risk of a suit. That's not true. In every state there are protections for discussions of errors for purposes of quality improvement. Hospitals, health care organizations, and practices that are serious about safety are going ahead and looking at what's going wrong and making changes. You don't have to say, "I made a mistake" to recognize that there's something wrong with a process and redesign it. We've got to get beyond our concerns about malpractice. We do need to drastically change the way we deal with negligent or potentially negligent care in this country, but we're not going to be able to do that in the next year or two, and we cannot hold progress and safety hostage to fears about that process. I'm impressed that there are a lot of physicians out there who are quietly making their processes safer. They don't find that it exposes them to increased risk of lawsuits.
MC: How large an impact do you expect from information technology tools, such as electronic prescribing and the long-awaited electronic medical record?
LEAPE: One of the recommendations of the most recent IOM report was that we vastly expand our use and application of information technology at all levels, both known and to be developed. No one really quibbles with that. One cautionary note is that any time you introduce a new technology you introduce new possibilities for error, so it has to be done thoughtfully and with monitoring to see what's going on. Our experience is that computerized order entry, for example, dramatically reduces errors. It also has the potential to introduce new ones. Therefore it has to be done with great care. On balance, technology is a positive, and as we work out the kinks it will have very little downside. I'm also excited about the entry of handheld devices. One fundamental principle of design for safety, and a cardinal "human factors principle," is to make it easy to do right and hard to do wrong. If doctors easily and accurately can order prescriptions, they'll quickly shift to new uses of technology. There are a lot of exciting things on the horizon, and we need to grab them and run with them.
MC: Health professionals have been reluctant to use technology that slows them down or doesn't have a clear value. How close are we to having tools that simplify rather than complicate practitioners' work?
LEAPE: It's a mix, but there are many factors driving doctors to become more comfortable using more sophisticated technology and sharing it with their patients — a change that is coming in the nick of time, as far as I'm concerned. It has been a long time since any physician could keep in his or her brain all the relevant medical knowledge about everything that he or she may see. We have to become much more comfortable with managing rather than simply knowing information. Doctors every day now are seeing patients who look up their symptoms or condition on the Internet and, at that moment, are better informed than the physician about the disease. They just read it; the doctor may have read it years ago. We have to be able to deal with that and realize it's not a threat to our autonomy, confidence, or professionalism. We have to turn it around and use it to the benefit of the patient. Good physicians believe that the well-informed patient is the best patient and that the more he or she knows, the better. We have the opportunity now to have patients who are very well informed, and then we can build on that and carry out our responsibilities to help them with their condition.
MC: The system is moving from one built around treating acute episodes to managing chronic diseases. What are the implications of that?
LEAPE: It's a multidimensional challenge. The second IOM report [released in March and characterizing the U.S. health care system as disjointed] gets at some of this. Clearly, chronic-disease management is thinking about care as continuity, not as episodes. Much of our health care system is designed around episodes of care, admission to the hospital, receiving tests, and office visits. We need to get beyond that and develop creative ways of involving patients and their caregivers much more in their care on a continuing, rather than episodic, basis. The second implication is that making available the full array of all that we can do for patients with chronic conditions involves tremendous complexity that requires a team effort. No individual doctor, nurse, technician, or therapist can meet all of a patient's needs, and no one should try. We need to coordinate efforts and work together. It's better for the patient and more satisfying to the workers.
MC: How well are physicians being trained to work in that kind of a team environment?
LEAPE: We haven't been doing a good job at that. We need to change our training methods starting in medical school. Some people have proposed that it's time to have medical students and nurses and pharmacists work as a team around some clinical problems in school so they get used to doing that. Clearly we have very different curriculum needs in terms of basic science and much of the specific training, but we should look for places where we can make common ground and give our young doctors experience working as part of a team.
MC: With the work that the IOM and others have done, is the structure in place to begin to make the changes in patient safety that are necessary, or are we still too early in the process?
LEAPE: It varies a great deal. Some health care organizations have an institutional commitment to quality and safety and a tradition of professionals and nonprofessionals working together and looking at their processes.
MC: We're moving toward a national quality report card. Where does that stand, and how long will it be before we have leading patient-safety indicators that are analogous to leading economic indicators?
LEAPE: It's hard to know. We really would like to know outcomes, but they're hard to measure because of the tremendous unreliability of our reporting methods. That's not going to change very rapidly. For example, if you try to rank organizations by numbers of reported errors, organizations won't report errors. For the foreseeable future, instead of looking at outcomes, I believe we will have to look at process measures, which flies in the face of the great outcomes movement of the last five years. In safety, we haven't yet reached the point where we can judge quality by outcomes. But we certainly can ask whether they have safe procedures in place. The purists will say that's not measuring what they care about, the number of people being injured by mistakes. I agree, but we won't be able to measure that accurately for a long time.
MC: Why not?
LEAPE: Because we are a long way from creating an environment where it's safe for people to talk about errors. Until that environment exists, the numbers never will be accurate. We need to accept that. We can do a lot in terms of measuring the processes we put in place. That's a fair report card.
MC: Pressure, from the media and others, to produce those bottom-line numbers is not going to go away.
LEAPE: That pressure better go away because it's an exercise in futility. The media and the payers and the regulators need to be educated. I'm serious. All the people who are knowledgeable about this issue are of the same mind. There is no way you're going to have a credible "error rate," and it's time to quit talking about it.
MC: How would you characterize the federal government's response to the issue of patient safety?
LEAPE: We couldn't have asked for a better response from Congress. The response has not been punitive or regulatory. Congress isn't saying, "Let's get tough with the bad doctors." They have listened and have tried to respond in a positive way. They've given $50 million a year to the Agency for Healthcare Research and Quality to develop best practices, fund research, and disseminate information. Pending legislation would give protection for reporting for quality improvement. So far, it has behaved in a very responsible way.
MC: How optimistic are you about the progress that is being made toward greater safety in health care?
LEAPE: Change never comes fast enough, but we're certainly on the right track. There's an incredible amount of work going on at all levels, from simple things like medication systems, to more complicated issues such as how physicians and patients interact, work force issues, and house officers' hours. These issues are being discussed and thought about, studies are under way, and action is being taken. Hundreds of hospitals are implementing safety programs of one kind or another. Safety is certainly not a fad. There are 10 times as many things going on now as two years ago, and government assistance is going to make a big difference. AHRQ has issued a flurry of requests for proposals for research on many key safety issues. By doing that, the agency is going to create a cadre of researchers and thinkers on this issue who will be leading the future. There are a lot of exciting things going on. So, while we should not be complacent, I think we're off to a good start.
MC: Thank you.