Editor's Note: Sir Liam Donaldson, MD, MSc, is England's Chief Medical Officer, a post often referred to as "the Nation's Doctor" (similar to the role of the U.S. Surgeon General). Trained as a surgeon, Sir Liam has been an inspirational leader in public health and health care quality in the United Kingdom for two decades. He has also emerged as a world leader in the patient safety field, authoring or commissioning dozens of influential reports, and serving as the founding chair of the World Health Organization's World Alliance for Patient Safety. We spoke to him about patient safety from an international perspective.

Sir Liam Donaldson: First, when I was teaching medical students, I felt that the process of medical education and training—the so-called socialization process into medicine—was at times quite negative for the patient. I think we were tending to eliminate too much of the humanity from our young medical students. Early on, our students were indignant about the sort of things that any lay observer would be indignant about. For example, seeing how physicians talked to patients or indeed talked about patients without involving them in the conversation. The students used to come into the tutorials in the first year very passionate about how medicine shouldn't be about that. When I saw them later on, they were much closer to the model of the body being a diseased organism. They were harder, I guess, as far as the humanity of medicine is concerned. So I always felt that somebody needed to try to preach that humanitarian philosophy as something that should be alive and well in modern health care. I don't think it is, but it should be. The second thing was, when I was medical director of one of the National Health Service's regions, I used to hear about a lot of adverse events. Some were not really about the system side of medicine, about patient safety. Rather, they concerned incompetent or poorly performing doctors, whose colleagues were covering up for them. And I did start to challenge these doctors' colleagues, who were usually protecting their associates. It made me a bit unpopular, but I wasn't prepared to see a tradeoff between patient safety and professional loyalty. So I attempted to uncover some very long-standing problems of poor performance or misconduct, which had been concealed or tolerated. Then I also started to see incidents reported from our hospitals to the region. It seemed to me that history was repeating itself, in that I was seeing identical incidents occurring over and over again, and I felt that something should be done about it.

RW: How do you explain the dominant tensions in this field, which are that it's mostly about the system, but we have to acknowledge that there are bad people?

LD: Well, I'm aware that people in the health care safety field are beginning to find parallels with the airline industry a bit tedious, but there are many lessons that we can learn from them. The big breakthrough in airline safety was made when the field came to understand systems causation and implemented use of crew resource management, team-based training. In the 1950s and 1960s, it had all been about technical factors: the design of aircraft, the development of procedures and protocols, etc. But in addition to the important technical work, there is now an emphasis on the training and assessment of pilots. They go in to a simulator every year. They have another pilot flying on the flight deck with them. They have regular medical checks. Most doctors, at least in my country, are not assessed once they've gone into autonomous practice—for the whole of their career through to retirement. We need to have competent, well-trained, conscientious individuals; we also need to focus on system issues. The danger however is that we're addressing the problem of unsafe care by saying, yes, we've ignored the system too much, so we'll swing the pendulum toward systems and forget about the individual for a while. And I think that's probably a mistake. We need both.

RW: When you're working in a given system and culture, there's a tendency to think that many things are determined by that system. As you think about patient safety, what issues really are determined by local culture and organizational structure, and what seems to be invariable across national boundaries, organizational systems, and payment methods?

LD: Well, culture and attitude tend to vary around the world. However, I don't think you ever see a whole health care system whose culture is oriented toward being truly patient-centered and truly focused on safety. Probably the most important question to ask is: what would it take, or how would we be able to get there? And how do we make every physician, every nurse, think through the prism of, "How can I make that patient's care as safe as I would want it to be for my mother's care?" Only a proportion of health care professionals think like that all the time, but it is probably the minority. That doesn't mean that the others are bad practitioners, it just means that (a) they're very busy, or (b) they see themselves as a cog in a machine rather than as human beings with feelings that can be tapped to create more profound interactions with patients. As far as the incentives and the flows of money, I don't think that there's much evidence to tell us what impact they have. In the UK, we do have an adversarial tort system of litigation which is like yours, although within our society there isn't a comparable urge to litigate. So our premiums aren't as high and our settlements aren't as high. Patient opinion surveys seem to tell us that when somebody suffers harm, their motive is to find out what went wrong, to have an explanation, and to make sure it doesn't happen to anybody else. Getting financial compensation comes lower down the list. Whether that will always be the case, I don't know. I produced a  a few years ago to try to introduce an element of no-fault compensation into our system, hoping that would create a less litigious system and ensure greater emphasis on learning. Because the implied philosophy of the tort system seems to be that if you punish somebody by awarding a claim against them, that will convince others to practice more safely or the same will happen to them. The irony of this philosophy is that, given that most litigation is settled out of court with secrecy agreements, the learning is concealed within that legal transaction. It is not actually exposed, and so nobody learns from the "punishment" of the person who had the settlement awarded against them. I think a no-fault system would ensure less adversarial litigation and more incentives for learning. I also think it would be interesting if a proportion of the damages were given over to patient safety. Five percent would make a lot of difference, and again it would be an incentive for people to do better.

RW: What are we learning about how well reporting systems improve patient safety?

LD: Many countries have launched their patient safety agenda by establishing a reporting system. It's probably one of the examples where we've too uncritically mimicked the airline industry. We mustn't forget that the airline reporting system is heavily supported by the black box, in that a lot of technical data are downloaded automatically from every flight. We don't yet have an equivalent to that in health care. Our approach of introducing reporting uncritically has quite a downside, in that relatively little attention has been given to the overall purpose of the reports. We need to address a number of challenges; first, we haven't made our mind up on the element of reporting that needs to take place purely at the local level within the hospital or on a national or an international level. So in a lot of systems, including the one in the UK, the view is: report everything, to both the national and local levels. The results is that, in the UK, we have a million reports through our system so far. The problem is that at the present time we're swamped with data and we haven't got enough information out of it yet—in particular, we don't have actionable findings. Ultimately, the challenge for reporting is how much of the data you gather can be rapidly translated into actionable findings. So often the incident report is of an outcome or some unsafe process that's been observed. Then the in-depth, root cause analysis comes later. It's not clear whether that information should then be added into the reporting system, or how it should be aggregated and analyzed. Then you also have to think about whether to wait for a full-blown incident or whether there are pointers to things that might go wrong that you ought to be collecting information on. That's where we can go back to the lessons of the airline industry. Air crashes are so rare now, because they do tend to focus on casual and poor practices, which may signal a potential unsafe situation and action can be taken. In short, the big picture points on reporting systems are how do we decide what to do at local level versus at national level; what analytical tools do we use; and how can we recognize that reporting systems are only one way of yielding actionable findings. Proactive hazard analysis, which has nothing to do with reporting, is another technique that is not really used in health care. It's simply looking at the environment of care and identifying the sources of risk. It reminds me of the pictures in children's magazines that I remember from my childhood—the ones where you have to spot the missing objects or the randomly hidden objects. Similarly, maybe we could walk into an operating theatre and say, there are 30 sources of risk there, go find them.

RW: Talk a little bit about the World Alliance for Patient Safety, if you will. What was it that inspired you, and what should people know about it?

LD: I was the United Kingdom's representative to the World Health Organization (WHO). I still am. I found myself sitting on the WHO's executive board asking myself, what can I do to raise the profile of patient safety to a worldwide level. After giving the board a presentation on patient safety, we developed a resolution, which was passed by the World Health Assembly. It called on all countries to really take patient safety seriously and to put in place programs of action in their countries. WHO then established a patient safety program/Alliance, and each year the Alliance team sets out the areas of work that will be addressed in a "Forward Program." We are very proud of our engagement with patients and the work which is being delivered under our "Patients for Patient Safety" initiative. Another activity focuses on research and development. We are developing a WHO taxonomy for patient safety so that reporting systems in different countries could interact. We wanted to develop a program for evaluating good solutions to reduce risk, so we have a "Safety Solutions" initiative. We have produced guidelines on reporting systems so that people didn't make the mistakes of some of the earlier pioneers and could learn from their successes, and Dr. Lucian Leape has led this work. Every 2 years, we pick one subject in the field of patient safety—a topic, where the world can unite to try to make progress quickly. We called this the "Global Patient Safety Challenge." The first topic we chose was health care infections, and our first campaign is "clean care is safer care." Each initiative or activity is led by one person who is asked to provide the overall leadership, usually an expert in that particular area of safety, and then an advisory board, and a project team is put around them. So we've been running formally for about 3 years, and we're just starting to think about evaluating the programs that have been in the first phase of activity and then setting up some new programs, including a second Global Patient Safety Challenge.

RW: When you think about shared learning and the value of a world alliance, in some ways it's conceptually easy to think about what France could learn from the UK or what the US could learn from Spain. It's a little harder to think about what countries in Africa could learn from developed countries. What have you learned about the commonalities and differences as you try to share patient safety learnings across that divide?

LD: It's very important that the agenda does embrace both the developing world and the developed world, because that's very much the ethos of the World Health Organization. We've launched the alliance in most of the WHO's regions now, including the African region. We were concerned when we went to Africa to talk about patient safety that they would simply stop the discussion cold by saying, "We haven't got any resources; how can we even think about safety?" But that sentiment was not expressed. They certainly drew attention to the impoverished circumstances as far as health care resources are concerned. But they tended to take the view that "we may not have very much, but we want to deliver care as safely as possible, and that when more resources do come in, we want the service to be built on safe foundations." So we found a lot of enthusiasm and commitment. We can guess the nature of their problems, but we don't actually have any evidence. One of the tasks of the Alliance's research program is to do prevalence studies of medical error and safety in the developing world to try to establish the precise scale and nature of the problem there.

RW: When you think about beginning your crusade, particularly as a focus on patient safety, what about change has come easier than you expected and what has been harder than you expected?

LD: The change has been much more difficult for me on the national front than on the international front. I've been amazed at how easily people have come together on the international front. I really don't know why that is. It's possibly because change poses less of a threat. In the UK, when you're trying to introduce change, you rather feel that you're in a goldfish bowl. All the difficulties that people focus on, all the criticisms, all the reasons why we shouldn't do things. On the international front, there really isn't such a focus on difficulties and impediments to change. People seem to be very enthusiastic and committed. I would have guessed it would have been the other way around before I started.

RW: What's been harder than you thought?

LD: Probably the hardest thing has been to get safety information into the public domain. When you're in a political environment, releasing information, documenting the scale of harm, even though it's available in the medical literature and in expert reports like the Institute of Medicine report and An Organisation with a Memory, it's very scary for politicians to see a news release saying this number of people have died and this number of people have been harmed, even though the number may be no different than what's well documented and established. I don't think that we can make significant progress until we have that sort of information regularly in the public domain, and people are watching the trends and the improvements. Then we can start to highlight areas where things are getting better. Until we're tracking it on a regular basis, it's always going to be a scary subject when it pops up, and then is buried in yesterday's news. And most people, apart from the enthusiasts, forget about it until the next news release.