Few publications in recent memory have had the impact of the 1999 Institute of Medicine (IOM) report, To Err Is Human.1 Neither the alarming statistics nor its central message, that errors are caused by faulty systems not by faulty people, was new. Indeed, the importance of systems design has been well recognized by human factors experts in industry for decades2 and promulgated by a few recent converts in health care for 5-10 years.3 Nonetheless, the speed and intensity with which this report from the National Academy of Sciences captured media, public, political and professional attention surprised everyone. And, it is no passing fad-attention to patient safety has not subsequently flagged, it has increased. In this Medical Error Symposium, the American Journal of Law and Medicine has brought together authors from diverse backgrounds, medical and legal, to advance the debate further, raising important questions about the implications of that report and the changes already underway in health care. It is interesting to speculate on the reasons patient safety has become a cause celebre. Beyond the lurid statistics, surely a major factor in the traction enjoyed by this issue is the unique nature of medical injury. While many other human enterprises, such as aviation, building, and military operations are associated with substantial hazard, in no other situation is the harm and suffering caused by the actions of individuals whose sole purpose is to relieve suffering and in whom the victim places a profound and personal trust-doctors and nurses. It is perhaps this very personal sense of affront and betrayal that accounts for the intense emotion surrounding medical injury, particularly when caused by an error. And, the feeling is universal: victims, doctors, lawyers, and the public are all affected. After all, we're all patients at one time or another. Ironically, that unique nature of medical injury, or more precisely, our reaction to it, has been the major barrier to reducing medical errors and injury. Shame, guilt and fear prevent many physicians from discussing their mistakes, being honest with patients, and being able to look beyond their individual errors to correct underlying systems failures. They can only try harder. For many lawyers, a sense of just cause, in some cases moral outrage, similarly blinds them to alternatives to tort litigation. Both are misplaced. And both have been manifestly unsuccessful in preventing medical injuries. We have created a monster. The toll of medical injury is truly appalling. While the oft-quoted figures of 44,000-98,000 deaths per year are horrendous in themselves, little attention has been given to the accompanying estimates that over a million people are injured by medical treatments annually in the U.S. Even this number, shocking as it is, is clearly an underestimate. Without exception, every study that has examined specific types of injuries (such as adverse drug events, injuries related to use of medications) with more sophisticated methods finds rates of injury that are much higher than those indicated by the rather crude record-review and population-based studies that are so widely quoted.4 And autopsy studies suggest that preventable deaths may also be many more than 98,000.5 The second reason the IOM report captured both public and professional imagination was, I believe, that it packaged shocking news with a compelling remedy; that errors can be prevented by systems re-design. Le., we can do something about it. This idea that errors are primarily caused by systems failures and not human failures is a truly transforming concept. It turns on its head our long-held beliefs and assumptions about why people screw up and what to do about it. It is truly a paradigm shift. Early evidence also suggests it works. For several years before the IOM report in 1999, leaders in a number of health care institutions across the country had begun to implement non-punitive reporting and use of human factors principles in the redesign of systems with fair, if modest, success-particularly in medications systems. …