May 24, 2001

 Good morning, Mr. Chairman, Senator Kennedy, and members of the committee. My name is Lucian Leape and I am member of the faculty of the Harvard School of Public Health. I practiced as a pediatric surgeon for much of my career, but in recent years have focused my attention on research into medical errors. While I was a member of the Institute of Medicine's Committee on the Quality of Health Care in America whose report To Err is Human: Building a Safer Health System has led to our meeting today, I speak not for the Committee, but as an individual concerned with making health care safe for all Americans.

 I wish to lend my strong support for the Patient Safety Improvement Act. Nearly 18 months ago, the Institute of Medicine (IOM) published its now-famous report, To Err is Human: Building a Safer Health System. After drawing attention to the appalling magnitude of the problem of accidental medical injury, it called for a national commitment to reduction of medical error. Both public and private organizations responded promptly with a flurry of activity and implementation of a variety of safety initiatives. Particularly commendable was the alacrity with which the federal Quality Interagency Coordination Task Force (QUIC) supported the IOM recommendations and government health care systems began to implement the IOM proposals. As the result of Congress's generous funding of $50 million for patient safety, the Agency for Health Research and Quality (AHRQ) in particular has moved aggressively to advance the cause of patient safety.

 Progress in the private sector has also been impressive. Hospitals and health care organizations throughout America are implementing safety programs. Key professional organizations, such as the National Patient Safety Foundation, the American Hospital Association, the American Medical Association, the American College of Physicians, the American Nurses Association, the Association of Health Systems Pharmacists have expanded their focus on safety. Training, credentialing, and accrediting organizations, including the Accreditation Council for Graduate Medical Education and the Joint Commission on Accreditation of Healthcare Organizations have given patient safety new visibility by setting new standards and guidelines. Insurers and purchasing groups, such as the Leapfrog Group have made safety a priority in their contracting with providers. The patient safety movement is clearly launched.

 It is worth stepping back a moment to recall what the IOM said that catalyzed this unprecedented response. Clearly, the announcement that up to 100,000 people die annually from preventable medical accidents commanded media attention and motivated a wide constituency to action. But that grim statistic was packaged with an insight borrowed from human factors engineering and other industries that provided both hope and a solution to the safety crisis: most errors are not due to carelessness or lack of trying hard enough, but are caused by faulty systems: processes and conditions that lead people to make mistakes. These systems are largely not under the control of the individual on the front line. The important implication from this discovery is that mistakes can be prevented by redesigning systems to make it difficult for people to make mistakes, something learned long ago by safe industries, such as aviation, chemical manufacturing, and nuclear power. But, identifying systems failures and correcting them will not occur if workers are punished for making errors. Safe industries create a non-punitive environment, and also pay attention to factors that affect performance, such as hours and work loads, work conditions, and team relationships, an agenda that health care must now address.

 The legislation proposed in The Patient Safety Improvement Act can measurably improve patient safety both by enhancing the value of reporting of injuries and errors and by supporting the Center for Quality Improvement and Patient Safety.

 Voluntary reporting of adverse events In considering this legislation, it is important to recognize what it can and cannot do. By itself, reporting does not improve safety. That can only occur when health care organizations use the information generated by a reporting system to change their systems. Whether the information enables them to do that depends, in turn, on the quality of the reports that are received.

 Reporting must occur at several levels. First, and foremost, accidents must be reported to the person who has been injured. Every patient has the right to know if he or she has been injured by treatment, and whether the injury has been caused by an error or other breakdown of the system. Full and prompt disclosure is a moral obligation of both the hospital and the professional caregiver. All major professional medical and health care organizations endorse this obligation.

 The second level of reporting is within the hospital. Reporting at this level, so-called internal reporting, is a key feature of all hospital safety programs. However, incident reporting by itself will not reduce errors. The reason is that the causes of accidents are seldom obvious. The apparent causes (e.g., the nurse misread the label, the doctor forgot the correct dose, etc.) are themselves the results of multiple contributing factors (such as confusing labels, distractions, lack of information, miscommunication, excessive work loads, fatigue or sleeplessness, etc.) Uncovering these underlying systems failures requires expert investigation and analysis. Finally, after multiple report analyses have pointed to common systems' failures, changes must be designed to correct these systems deficiencies and then implemented. In other words, it is not reporting that improves safety, but the response to the reports. Without the response, reporting of accidents will have little or no effect on reducing errors.

 Reporting must also be safe. If personnel are punished for making or reporting errors, they will find ways to avoid reporting. The power of a safe and responsive environment is illustrated in the accompanying Figure (next page) which shows the change in the reporting of medication errors in a single nursing unit of a hospital when the nurse manager instituted a nonpunitive reporting system. The number of reports received monthly increased 20-fold. This experience has been replicated by many other hospitals as they have implemented non- punitive systems. Stimulating the creation of non-punitive, responsive reporting systems within health care organizations has been a major goal of the National Patient Safety Foundation, the Veterans Health Administration, and a number of major health care systems.

 Error Reporting (Image not transmittable)

 The third type of reporting, external reporting, is the subject of the proposed legislation. Its effectiveness clearly depends on internal reporting systems to generate reports. An external reporting system can advance patient safety in several ways. First, if incidents are promptly reported, it can serve as an "early warning system" to alert all providers of new hazards. The medication error reporting program (MERP) run by the US Pharmacopoeia (USP) and the Institute for Safe Medication Practices (ISMP) has performed this function for medication errors for years, as does the FDA MedWatch program for adverse drug reactions. Second, by aggregating large numbers of reports from many institutions, a national reporting system can identify patterns of injury that are not obvious to those at the local level, particularly for rare events or those with unusual causes.

 But the greatest benefit of a national voluntary reporting system and database would be to enable all health care organizations to benefit from the lessons learned in each. For this to occur, the institution must thoroughly investigate the incidents and report the findings in a standardized way. The instruction to AHRQ to develop a common format for reporting is, therefore, an important aspect of this proposed legislation.

 No voluntary reporting system will be successful, however, unless, like internal systems, reporting is safe. While some have suggested that physicians and hospitals should be required to publicly report all errors or accidents, such a proposal perpetuates the climate of blame and punishment that has been the major barrier to making progress in safety over the years. The evidence is clear that doctors and hospitals can readily avoid reporting incidents if it exposes them to risks of censure or litigation. The proposed legislation wisely provides legal protection for information submitted voluntarily to patient safety improvement systems. This feature is crucial for the success of all voluntary reporting systems, public or private.

 The potential for reporting of medical errors and incidents to contribute to patient safety is substantial. An oft-cited model is the "close-call" Aviation Safety Reporting System (ASRS) run by NASA, which has been successful because reporters are protected from disciplinary action and reporting leads to expert analysis and feedback of information to the FAA which makes procedural and other changes as needed. This voluntary system receives over 36,000 reports annually and has contributed significantly to aviation safety over the years. Properly managed, a medical incident reporting system could be even more successful.

 Mandatory reporting. It is important to distinguish the methods and purposes of voluntary reporting of medical errors and accidents for improvement of patient safety from state-based mandatory reporting programs designed to monitor accountability of hospitals and other health care organizations. These programs focus on deaths and serious injuries that are obvious and are generally understood to be preventable. Their occurrence suggests that internal "safety net" procedures may not be working properly. The IOM recommended that the responsibility for these programs remain at the state level, but that de-identified information from these reports could be used to improve safety if it were aggregated and analyzed by AHRQ. To facilitate this, it recommended establishment of a standardized list of serious reportable events that could be used by states that wished to participate in such a program. The National Quality Forum is currently developing such a list.

 The Center for Quality Improvement and Patient Safety

 While the provisions for protection of information and the establishment of a Patient Safety Database are important for encouraging reporting and disseminating information about errors and injuries, it is the proposal to establish a permanent Center for Quality Improvement and Patient Safety at AHRQ that holds the greatest promise for advancing the science and the practice of safe health care. Congress should continue to support it fully.

 The science of safety in health care is in its infancy. Many hospitals and other health care organizations are committed to reducing errors, but are not sure what they should do. They need advice and information on a variety of safety issues, ranging from best practices to improve safety and how to implement them to policies related to staffing and hours. Before we can expect them to implement safe practices, we need to better define what those practices are. The AHRQ is uniquely qualified to provide that kind of help.

 Under the outstanding leadership of John Eisenberg, AHRQ has moved rapidly this past year to bring together diverse stakeholders to define areas in safety needing further research, disseminate information on known safe practices (such as medication safety), provided help for consumers, and requested and funded research proposals addressing a broad range of safety questions, including such diverse and important topics as the design and evaluation of reporting systems, improvement of medication systems, enhancing collaboration and teamwork, and the effect of working conditions on safe performance. It has commissioned the National Quality Forum to convene panels to evaluate and recommend best practices and to develop a standard list of serious reportable events.

 In a very short time, the Agency has already become what the IOM called for, a central focus of activity in safety, the demonstration of a national commitment to safety that facilitates the work that must be done at the local level, in hospitals and health care organizations. This is government at its best: not telling doctors and hospitals how to practice, but providing the research, resources and information needed for them to do what they need and want to do, make health care safe for all citizens. AHRQ is now the major force for improving patient safety, and with continuing support it can become even more useful.

 The initial funding of $50 million provided by Congress in fiscal year 2001 for the safety effort at AHRQ has had another very important effect: it has attracted a number of talented researchers and practitioners to work in patient safety. This is particularly important at this time because this young field of health care safety has only a handful of experts. Developing that expertise, which is analogous to specialty training in medicine, will take time, but a good beginning has been made. It is, however, just a start. The number needed is huge: providing just one qualified person in each hospital, for example requires nearly 5000. If the momentum and commitment are to be sustained, funding for the Center for Quality Improvement and Patient Safety must be increased substantially each year.

 In conclusion, I strongly support all of the legislation proposed in the Patient Safety Improvement Act. Both the protection for voluntary reporting and the establishment of a permanent Center for Quality Improvement and Patient Safety at AHRQ are high leverage measures that can have profound effects on advancing safety in health care. I urge the passage of this Act.