Testimony of William E. Golden, MD, FACP President The American Health Quality Association Before the Commerce Subcommittees on Health and the Environment and Oversight & Investigations and the Veterans’ Affairs Subcommittee on Health U.S. House of Representatives February 9, 2000 Good morning, Mr. Chairman. As the Principal Clinical Coordinator for a Medicare Peer Review Organization, and as a physician who has treated hundreds of veterans in VA medical centers, I am particularly happy to have this opportunity to participate in a joint hearing of the Commerce Committee and the Veterans’ Affairs Committee on the important problem of medical errors. While I spend most of my professional time working for the Arkansas PRO, I am also a Professor of Medicine and Director of General Internal Medicine at the University of Arkansas Medical School. The Arkansas PRO has extensive experience in performance measurement and conducts quality improvement, HEDIS measurement, and patient satisfaction surveys for the state Medicaid program. We are also a recognized vendor for the Oryx Program of the Joint Commission on Accreditation of Healthcare Organizations (JCAHO). In fact, we created three of JCAHO’s thirty performance measures in the proposed national core program. I am here today as President of the American Health Quality Association (AHQA), a national membership association of organizations and individuals dedicated to health care quality improvement. Our member Quality Improvement Organizations (QIOs) are private, community-based organizations that promote health care quality in all health care settings. QIOs work in all 50 states, the District of Columbia and the U.S. Territories. The QIOs have several lines of business including work with state governments and private health plans. The work that unites them all, however, is their 3-year, competitively awarded contracts from HCFA to evaluate and improve the quality of care delivered to Medicare beneficiaries. For this work, our members are more commonly referred to as Medicare Peer Review Organizations, or PROs. Congress established the PROs in 1983 to look for single case problems. During the 1990s, the PRO system evolved to become a national network of quality improvement experts that systematically evaluate the delivery of health care in a region and institute projects to educate and alter the clinical behavior of institutions, health professionals and patients. QIOs are staffed with clinical experts, communication experts, and data and statistical professionals who work together to analyze and collaborate with the health care system in their communities. Today’s PRO system is uniquely qualified to serve as the core of a new national system for improving patient safety. One of the greatest strengths of the PRO system is its extensive infrastructure of relationships in every region of the country. PROs work individually with hospital staffs and physicians offices. They are also increasingly engaged with home health care systems, nursing homes, academic health centers, and community groups such as heart associations and cancer coalitions. In addition to technical expertise, they have developed public relations and outreach strategies with professional associations, public health authorities and state officials. This is critical for helping hospitals and other facilities implement improvement strategies as well as tailoring messages to the public about improving their health (e.g. public awareness of receiving pneumococcal vaccinations or getting regular eye examinations to reduce the risk of diabetes-related blindness). This is also critically important for the effectiveness of the PROs’ required projects with underserved and disadvantaged populations. These projects often require forms of outreach and communication that are culturally appropriate. The Institute of Medicine (IOM) report released last November targets both medical errors of omission — care not provided that should have been — as well as errors of commission. In addition, the IOM Committee also states that errors occur and should be detected in all phases of medical care: prevention, diagnosis and treatment. The Medicare PRO Program as a Model Error Reduction Program. Medicare’s national PRO system has been identifying, measuring and reducing error rates for several years. The PRO program is now embarking on an expanded three-year mission to identify and eliminate medical errors. The new program is focused largely on errors of omission — such as prescriptions that were not ordered for prevention of heart attack — and on errors in all three categories mentioned by the IOM. For example, in the prevention area, PROs are working to promote immunizations to prevent the most common fatal infection, pneumococcal disease. In the area of missed diagnoses, the PROs will be working to increase mammography screening and diabetic retinopathy testing. An example of PRO work to reduce treatment errors is that PROs will be emphasizing timely administration of antibiotics for newly hospitalized pneumonia patients. I have attached a complete list of the 22 performance indicators in each of six clinical topic areas for which the PROs must reduce error rates. These PRO performance indicators serve as a useful model for a new medical error reduction system for several reasons. These clinical topics were carefully chosen because they affect a large percentage of older Americans and because the scientific basis for the desired therapy or action is well established. A national error reduction program should also focus on high priority problems and adopt a science-based approach. In addition, the standardized national set of performance indicators assures national comparability of data within and between all states, which is critical to accurately measure improvement. We believe this is a sound model for a national system of identification and reduction of medical errors. Recommendations. Based on our experience working within a national system to identify quality problems and work collaboratively with providers to bring about improvement, here are our recommendations for a new system for improving patient safety. Expand Monitoring System for Error Prevention. Congress should expand the current system utilized by Medicare to monitor a targeted list of health care processes and patient conditions known to be associated with a disproportionate amount of medical errors. This system will identify many errors and adverse events, which have not yet resulted in dramatic or catastrophic patient outcomes. The published literature identifies some categories of preventable adverse events that are both relatively frequent and frequently preventable, and might be targeted by a national monitoring system. Some examples include adverse drug events, hospital acquired infections, deep venous thrombosis, postoperative hemorrhage. The Agency for Healthcare Research and Quality (AHRQ) and the Health Care Financing Administration (HCFA) should collaborate with representatives of our national network of Quality Improvement Organizations (QIOs), as well as professional and provider groups to define the highest priority areas of scrutiny for error-prone health care processes, and to develop a standardized system for measurement. Congress will be asked to consider the burden of error reporting. The system of monitoring that I have described can be accomplished without imposing significant additional reporting burdens on hospitals or other providers. PROs can accomplish much of the data gathering necessary by expanding their current mechanisms for review of medical records and abstraction of key data for analysis. Quality improvements based on this kind of monitoring will probably continue to be the major method by which patient safety is enhanced. Because the PRO program has already established the relationships with hospitals necessary to perform this function, there is very little new work that hospitals must do to facilitate an expanded program to address errors in patient care planning and execution. Mandatory Error Reporting. We have recommended that Congress devote substantial resources to monitoring and educating providers about the adverse events that have strong potential to harm patients, rather than wait for patient harm to occur. But the smaller number of more dramatic events that result in patient harm must also be addressed by an error reduction system because the results of such errors are so often tragic and irreversible. This subset of adverse events often captures the attention of local health professionals and often results in demands for system changes to eliminate recurrence. Health facilities should report the rare and seemingly random adverse events that result in patient harm to a regional entity to create a database. Monitoring and analysis of such a database can offer insight into better system design for all of our communities. The reporting of such errors allows for hindsight analysis to be available throughout the health system, so that more people can benefit from the analysis than just those in the local environment that witnessed the adverse event. The PROs are well qualified to manage and interpret such a database in each state, and have proven adept at educating providers and practitioners about ways to avoid errors in the future. Ensure Accountability. Congress should hold providers accountable for measurably reducing the incidence of errors. A qualified expert organization, completely independent of hospital providers, should analyze the incidence of errors and judge whether improvements are being made. The PRO program is already performing this function on a more limited scale. For the period 2000-2002, PROs will be accountable under their Federal contracts for measuring and reducing the frequency of missed prescriptions to prevent strokes and heart attacks, or missed lab tests to help control diabetes. If a PRO cannot accomplish sufficient measurable improvement, it may lose its Federal contract. In a new medical error system, Congress can rely on the QIOs to measure error rates and identify providers that have made no progress in eliminating errors. Providers, which are making no progress on errors, could be reported to a federal or state regulatory agency, or to the Joint Commission for Accreditation of Healthcare Organizations (JCAHO). Assure Confidential Treatment of Reported Errors. Reports identifying specific providers and individuals should generally not be disclosed. Part of the reason for this is that "naming names" tends to fix blame, even when this is inappropriate. The IOM report [page 45] noted, "Complex coincidences that cause systems to fail rarely have been foreseen by the people involved." This suggests that it is more important to understand system failures than to attempt to affix blame on one or more individuals involved in a system failure. It is critically important to not to discourage, let alone punish, the active search for errors. Several studies demonstrate that errors are much more numerous than anyone can know without actively digging to find them. The IOM relied on two large studies of the prevalence of medical errors. PROs, in fact, did the medical record abstraction for the second study, based in Utah and Colorado. Both studies found a large number of preventable adverse events through careful review of the medical record. But these researchers also noted that many other errors could not be found in the medical record alone. When researchers at the LDS Hospital in Salt Lake City wanted to find out the true incidence of adverse drug events in their institution, they started by counting the incident reports filed by doctors, nurses, and pharmacists. They came up with about 20 reports a year. But after extensive mining of lab data, prescription records, and interviews with hospital personnel, they found the true incidence of adverse drug events was over 580 events a year. The hospital then tracked down the causes of these problems and reduced their true error rate below the original apparent rate. The LDS project puts the idea of public reporting in context. If hospital personnel know that any error they find involving patient harm will be subject to public reporting, few will undertake the costly and difficult investigations that are necessary to discover errors. If public disclosure and punishment await those who dig effectively to find the true extent of errors, few errors will be found, and fewer still will be eliminated. Congress has repeatedly recognized the importance of maintaining confidentiality for sensitive internal hospital quality improvement activities. For example, Federal law ensures that confidential data reported to PROs shall not be disclosed. Congress can ensure confidential treatment of this information by requiring that error reports be sent to the PRO in each state. The current PRO statute protects such information from unauthorized disclosure. Public reporting of errors should be reserved for those institutions identified by the PRO that cannot or will not improve error rates. At the state level, aggregate information without identifiers for individuals or institutions could be released to the general public. Data reported at the national level would first be encrypted for aggregate public reporting and would then be considered a publicly accessible dataset. Establish a Mechanism to Find Unreported Errors. Experience with other mandatory reporting systems for errors and health quality problems reveals that no mandatory reporting system will receive all appropriate reports. A separate mechanism to identify unreported errors is needed. One such system is already in place nationwide. Individual PROs periodically request records and analyze them for indicators of errors such as delayed administration of antibiotics in newly hospitalized pneumonia patients, and missed opportunities to prescribe medications to heart attack and heart failure patients. In addition, the national PRO program also utilizes clinical data abstraction centers (CDACs) to accomplish this task. These centers also observe strict confidentiality in managing the records, and have achieved a high degree of reliability in finding and reporting errors to PROs, which then work with the hospitals to prevent their recurrence. This system can be utilized to find many more types of errors. Institutions should be required to provide information in response to a PRO request to actively identify or pursue information that may not be readily identifiable in standardized reports. This mechanism will help to ensure the integrity of the mandatory reporting system, as it may uncover reports that should have been filed with the PRO but were not. Promote Best Practices. Once errors are found, their causes must be understood, and solutions must be implemented. This is now accomplished through the national Medicare PRO program by collecting from each PRO their successful interventions to improve care, and then sharing it with all the rest. In this way, every PRO can approach local institutions with the benefit of the best knowledge of all the PROs and providers that have previously tried to solve a problem. By assisting hospital personnel in finding best practices, the PROs go far beyond merely holding hospitals accountable for their failures. Separate Malpractice Reform from the Error Reduction Program. Tort reform and facilitation or limitation of litigation is a matter for a separate set of public policy deliberations. All information should be reported to the PROs for the purpose of assuring that measurable quality improvement is accomplished. Neither regulatory remedies nor liability law need be affected by reports to the PRO or by the confidentiality protections afforded such reports. AHQA believes these are the basic elements necessary for creating a systematic approach to reducing medical errors that will assure both medical professionals and patients that the problem is being addressed fairly and effectively. The key to a successful solution to this problem will be giving the medical community the opportunity to fully identify the possible extent of their errors and do the work necessary to systematically and measurably improve. Without this measurable improvement, the problem will continue to be discussed but never solved and consumers will never be assured that the quality of their medical care will become any better. The nation’s QIOs can provide the accountability and results that the system will require. Thank you again for the opportunity to share this information with Congress. I look forward to continued discussion as you work to improve the safety of patients across America.