Testimony by THE JOINT COMMISSION ON ACCREDITATION OF HEALTHCARE ORGANIZATIONS FEBRUARY 9, 2000 Before a hearing of the Committee on CommerceSubcommittee on Health and Environment, and Subcommittee on Oversight and Investigations,and the Committee on Veterans’ Affairs Subcommittee on Health
I am Dr. Dennis O’Leary, President of the Joint Commission on Accreditation of Healthcare Organizations. I am pleased to have the opportunity to address each of the three House subcommittees regarding "Medical Errors: Improving Quality of Care and Consumer Information." Medical errors is one of the most pressing quality issues we face in the health care industry as we approach the next millennium. The Joint Commission is the nation’s oldest and largest standard-setting body for health care organizations. We accredit over 18,000 organizations that provide a wide range of services, including hospitalization; long term care; ambulatory care; behavioral health care; laboratory services; managed care; and home care. Based on its broad experience, the Joint Commission has a panoramic view of the strengths and weaknesses inherent to our health care delivery system. We believe that the problem of medical errors is endemic to the way health care is carried out, but that we have the tools and commitment with which to sharply reduce their incidence. My testimony will focus on the Joint Commission’s Sentinel Event Program which was designed to reduce medical errors among all of our accredited organizations. I will discuss briefly its important features, and relate how the program has assuredly saved lives and prevented injury. But I will also stress the fact that without Congressional assistance, the Joint Commission’s error reporting program will continue to fall significantly short of its intended goals. Simply stated, the Joint Commission’s Sentinel Event Program should be looked to by policy makers for "lessons learned" when designing any national, state or local program of medical error risk reduction. There are two messages that I would like you to take from my testimony today. The first is that medical error reduction is an information problem. I will expand on this message by describing the attributes of the Joint Commission’s Sentinel Event Program, which specifically build on this point. We believe that the solution to reducing the number and types of medical errors resides in developing mechanisms for collecting, analyzing, and applying existing information. If we are going to make significant strides in enhancing patient safety, we must think in terms of what information we need to obtain, create, disseminate and apply to the problem. The second message is that we will not be successful in performing these information-driven activities if the Congress does not pass federal protections that will encourage the surfacing, evaluating, and sharing of that information. I will discuss this issue in the concluding portion of the testimony. The Sentinel Event Program Concerned about a spate of serious medical errors that came to its attention during 1995, the Joint Commission initiated its formal Sentinel Event Program in 1996. It is noteworthy that the recently released Institute of Medicine Report, "To Err is Human: Building a Safer Health System." lists many of the same events that spurred us into action four years ago. Many of these errors achieved high media visibility, but it became abundantly clear that these were the tip of the iceberg, and that even the most premier health care institutions were not immune to serious mistakes. It would be easy to attribute what appeared then to be a rise in errors to the increasing complexity of health care combined with escalating financial challenges in the health care industry. Certainly, health care has been experiencing restructuring, resource constraints, rapid technological advances, and an explosion of medical knowledge that makes it more difficult for practitioners to keep up with the latest knowledge and skill sets. As important as these factors may be, they should more aptly be considered exacerbating conditions rather than root causes of error. In fact, the very high dependence on human interventions and interactions characteristic of health care makes it prone to error. An industry so reliant upon human factors such as memory, emotions, communication, skills, and physical well-being must be supported by organizational and technologic systems to reduce the likelihood of mistakes. The Joint Commission saw the need to take a leadership role in helping health care organizations better understand the epidemiology of medical errors as well as the need for a systems approach to effective error reduction strategies. Therefore, the Sentinel Event Program was launched with the primary goal of applying scientific methodology to the problem to bring about a significant reduction in the numbers and types of medical errors. We designed the Sentinel Event Program with four information-driven functions:
Reporting of Sentinel Events It is imperative that any medical error reporting program operate under a pragmatic and carefully crafted definition of what is a reportable event. Standardization of the information to be collected is an important prerequisite for aggregating events in a consistent and meaningful fashion. Further, without a pragmatic definition, a reporting program would be flooded with hundreds of thousands of lesser injuries that would overwhelm the system. With this in mind, we identified a subset of sentinel events* -- including their nomenclature and taxonomy -- that would be reported to the Joint Commission on a voluntary basis. These reportable events affect recipients of care (patients, residents, enrollees) and meet the following criteria:
_______________________________________________________________ * The Joint Commission defines a sentinel event as "any unexpected occurrence involving death or serious physical or psychological injury, or the risk thereof. Serious injuries include a loss of limb or function. The phrase "or the risk thereof" includes any process variation for which a recurrence would carry a significant chance of a serious adverse outcome. the event is one of the following: suicide of a patient in a round-the-clock care setting, infant abduction or discharge to the wrong family, rape, hemolytic transfusion reaction involving administration of blood or blood products having major blood group incompatibilities, or surgery on the wrong patient or wrong body part. The foregoing definition of a reportable event minimizes the external reporting burden to health care organizations while focusing on the most serious occurrences that have a high likelihood of being preventable. The fact that the Sentinel Event program seeks to collect data on the most serious errors, or "crashes," distinguishes the Joint Commission’s reporting program from the voluntary programs encouraged in the IOM report, which would collect information only on the "near misses." An important feature of the Sentinel Event Program is the non-punitive reporting environment it seeks to create. Hoping to foster a positive culture that will promote error reduction efforts, the Joint Commission has designed the Sentinel Event policies not to penalize the accreditation status of an organization that surfaces an error and performs the appropriate due diligence required under the policy. The resulting atmosphere provides incentives that favor the surfacing of information about errors that eventually contributes to error reduction strategies that can be used by other organizations. Despite the incentive to report errors to the Joint Commission, the fear of public hangings and litigation are significant impediments for the majority of health care providers. Therefore, we have experienced only limited reporting to the Joint Commission’s database. Over the years, our Sentinel Event Program has made procedural accommodations to protect sensitive error-related information, such as having our surveyors review reported errors onsite rather than having information sent to the Joint Commission’s central office. But these manipulations are only stop gap measures that we believe must be replaced by federal protections for error-related information. I am going to return to the need to create a positive culture for reporting later in this testimony, because I believe it is the most important contribution that Congress can make to reducing medical errors nationwide. The Joint Commission has been especially pleased by the past support by some members of the Commerce committee for legislation that would promote a non-punitive environment for surfacing and learning from errors. Systems Analyses to Discover Root Causes While reporting is voluntary, the production of a root cause analysis following a sentinel event is a mandatory feature of the Sentinel Event Program. An accredited organization that experiences a sentinel event must produce an intensive analysis that encompasses a no-holds-barred vetting of all of the causes underlying the event. We call these responses root cause analyses -- a term borrowed from the engineering world’s reliance on a systems approach to both solving problems and producing desired outcomes. A root cause analysis focuses primarily on systems and processes, not on individual performance. While an individual is almost always the most proximal cause of a mistake in health care, it is also almost always the case that the fundamental causes of error relate to systems failures distal to the error itself. For example, systems may fail to provide simple checks and balances; or they may be missing critical safeguards; or may have design flaws that actually promote the occurrence of errors. These intensive analyses are rich learning processes that can elucidate multiple factors that ultimately contributed to the error. Many of these are not readily apparent until the root cause analysis is undertaken. Therefore, the analysis must be comprehensive, thorough, and engage the personnel involved in all aspects of the care giving and support processes. These are also time consuming investigations, and their complexity may require external technical assistance to do well. The Joint Commission has developed several comprehensive guides on how to conduct a good root cause analysis, and continues to be the leading source of guidance for health care organizations in this area. Unfortunately, the majority of reporting systems -- both voluntary and mandatory -- fail to require or encourage the performance of these intensive assessments. This was evident during our review of many state reporting programs. A reporting system that ends with the report of the event itself is not a credible program and will not contribute to error prevention. Root cause analyses also offer extraordinary insights into how processes must change to control unwarranted variations, and they tell stories of what systems must be developed to guard against the occurrence of similar human error. Root cause analyses hold the promise of prevention. They are also the necessary substrate from which risk reduction action plans are created. Not surprisingly, organizations are hesitant about sharing these root cause analyses with the Joint Commission or anyone else. Although many organizations have done so, we must recognize that preparing a document that lays bare the weaknesses in a health care provider’s system is akin to writing a plaintiff’s brief for purposes of litigation. Therefore, we cannot expect uniform preparation of these documents without accompanying federal protections against their inappropriate disclosure. Monitoring Action Plans and Safety Standards The Joint Commission monitors the action plans of accredited organizations which have experienced serious medical errors, in a manner similar to the way we monitor any quality of care area in need of improvement. This ensures that there is an independent review of the milestones associated with anticipated systems changes. Monitoring is an important part of the strategy for preventing errors, to ensure that the response to an error does not terminate in only the report itself or a discussion of what went wrong. We want to see an organizational response that results in preventive actions. The Joint Commission developed explicit patient safety standards that became applicable to accredited organizations beginning in January 1999. These new standards were specifically created to establish patient safety as a high priority in provider organizations. The new standards require that the leadership of a health care organization establish processes for identifying and managing sentinel events and put these into practice. The standards also require that the organization monitor performance of particular processes that involve risks or may result in sentinel events, and intensely analyze undesirable patterns or trends in performance. The standards make patient safety a visible responsibility of health care organizations and a requirement for accreditation. Compliance with these new patient safety standards is evaluated through our onsite inspection process. We view the monitoring function as a key element to public accountability. The public must have confidence that there is an external body requiring attention to patient safety within the organization that is delivering their care. We believe that the public views safety as a threshold concern. While citizens probably do not wish to have detailed data about safety prevention in each health care organization, they should reasonably expect that responsible oversight bodies are acting conscientiously and effectively on their behalf. This includes aggressive and timely follow-up to the occurrence of a serious medical error and holding the organization accountable for making necessary systems improvements. That assurance must be provided to the American public. At the same time, it is error-related data and information that undergird and drive this system of accountability and oversight. Therefore, we believe that any national response to the IOM report must ensure appropriate data sharing among all of the responsible oversight bodies which perform any of the functions discussed in this testimony. The health care quality oversight system has a variety of private sector and public sector players today. Efforts should at least be made to better utilize existing structures through improved data sharing and encourage the broad dissemination of what has been learned from medical mistakes. Dissemination of Lessons Learned To have a positive national effect on patient safety, information gleaned from errors must be aggregated, analyzed and disseminated to the health care community at large. The Joint Commission began its series of Sentinel Event Alerts to share the most important lessons learned -- known risky behaviors as well as best practices -- from its database of error-related information. To date we have issued Alerts in a number of areas, including medication errors; wrong site surgery; restraint-related deaths; blood transfusion errors; inpatient suicides; infant abductions; and post-operative complications. We are confident that these Alerts have saved lives. Unfortunately, we cannot calculate real decreases in error rates with scientific certainty, because the full scope and frequency of serious adverse events is simply not known. However, we have some data which illustrates the effects of our Sentinel Event program in selected areas. For example, we have seen a notable significant effect from our first Alert dealing with the importance of appropriate storage and handling of potassium chloride (KCl) -- a substance that is deadly when given in concentrated form and is easily mistaken for less benign substances. In analyzing the causes of KCl-related deaths in 1997, it became evident that accidental injection of KCl stored on hospital floors was an important cause of unanticipated deaths. The Joint Commission issued its Alert on the subject in February 1998. The number of reported deaths has dropped from about 12 the year before to only one in 1998 and one in 1999. We believe that significance should be attached to how information is disseminated and by whom. The risks associated with potassium chloride have long been known to practitioners. But when the principal accreditor of provider organizations issued a major alert, it caught the attention of organization leaders and health care practitioners. Moreover, it was clear to the recipients of the information that the Joint Commission would be paying attention to this particular issue and following up during onsite evaluations of the organization’s performance. This program of Alerts is an example of the type of vehicle necessary to achieve behavior change in health care organizations. Need for Congressional Action The Joint Commission is pleased that the IOM report has galvanized the professional and policy making communities around this critical set of quality issues. Such synergy of purpose among stakeholders is a prerequisite for solving complex, multifactorial problems that depend upon information sharing among the parties. Dramatically reducing the numbers and types of errors will take a concerted effort by all who play a role in the health care system. However, there is always the danger that in rushing to address a serious public policy issue, all of the elements necessary to success are not considered. The Joint Commission’s Sentinel Event program contains those elements, but it demonstrates very clearly that no reporting system for serious errors can fulfill its objectives without Congressional help. We urge, therefore, that Congress create statutory protections from disclosure and discoverability of the in-depth, causal information which must be gathered in any mandatory or voluntary reporting program for serious adverse events. The Joint Commission took this position publicly several years ago in seeking federal confidentiality protections for the root cause analysis information produced in response to a serious medical error. Soon after we began our Sentinel Event Program, many organizations expressed grave concerns that existing peer review statutes would not adequately protect the production and sharing of the intensive analyses. In fact, the Joint Commission’s subsequent review of state laws verified that they were inconsistent and often unclear about the extent to which health care organizations can share with accreditors or other third party external review organizations any assessments of cause and still maintain peer review protections. Therefore, the Joint Commission began seeking federal legislative protection which would make clear that information developed in response to a sentinel event -- and shared with an accreditor -- would be provided clear protections from disclosure and discovery. We are convinced that without such clear federal protection from disclosure of root cause analysis information, no reporting system will achieve its goals for error reduction. We believe this to be true for both mandatory and voluntary programs, for serious errors or programs for near misses. Fear of reprisals, public hangings, and loss of business will continue to impede both reporting and the production of in-depth, intensive investigations of the root causes behind medical errors. Rather than surfacing reports of errors, our blame-and-punishment-oriented culture drives them underground. Congress can make an extremely critical contribution to solving the information problem by passing legislation to address these legitimate fears. We also encourage you to consider all of the elements contained in the Joint Commission’s Sentinel Event program as components necessary to the successful address of the problem of medical errors, irrespective of whether solutions are considered at national, state or local levels. To actually accomplish the tasks presented in this testimony, many stakeholders must play roles. This will take significant data sharing between the public and private sectors to ensure that all of these functions can be effectively carried out. Thank you for the opportunity to present our views Back to Witness List |