The dramatic improvement in the airline industry safety record has been attributed in part to changes in practice introduced as a result of investigations of accidents by an independent third party, the National Transportation and Safety Board (NTSB). Would this model work in health care for high severity iatrogenic injuries? This question has frequently been asked over the last few decades, but the idea has never gained traction. The APSF organized a workshop focused on this topic with speakers from multiple disciplines including experts in health care system safety, the NTSB, and the medico-legal profession. Robert K. Stoelting, APSF president, and David M. Gaba, MD, associate dean for Immersive and Simulation-based Learning, professor of Anesthesia at Stanford University School of Medicine, director of the Patient Simulation Center for Innovation at the Veterans Administration Palo Alto Health Care System, and member of the APSF Executive Committee, opened the APSF Board of Directors Workshop at the Anesthesiology 2013 Annual Meeting in San Francisco, CA, by posing the question “Should anesthesia incidents be investigated as they are in other high-risk industries?”
Charles R. Denham, MD, the editor-in-chief, of the Journal of Patient Safety and the chair of the Global Patient Safety Forum, noted that because the database of health care accidents is sparsely populated, we should consider fast-tracking specific patient safety events and generating “Red Cover Reports” so that health care systems can learn from each other’s errors. This process would challenge the current risk management policies of institutions that prevent sharing of information nationally. He believes that we should use methodology similar to the NTSB to eliminate the more than 30 deaths per hour that are estimated to occur in U.S. hospitals. Dr. Denham published an article in 2012 with actor Dennis Quaid and famous airline pilots and authors “Sully” Sullenberger and John Nance in the Journal of Patient Safety making these points.
The Honorable Mark R. Rosekind, PhD, member of the NTSB and one of the world’s foremost human fatigue experts, was unable to attend the meeting in person (because of the temporary government shutdown last October), but provided his slides that Dr. Gaba kindly presented. The 2 major goals of the NTSB are to determine the probable cause of transportation accidents and to make recommendations aimed at preventing their recurrence. The NTSB was created in 1967 and has investigated over 132,000 accidents and has generated more than 13,500 safety recommendations. Although the NTSB is credited with making significant advances in safety in the transportation industry, it does not have the authority to regulate or enforce its recommendations. It oversees transportation accidents in the aviation, marine, highway, railroad, pipeline, and hazardous materials industries. Dr. Rosekind noted in his slides that the major strengths of the NTSB are its rigorous investigations, independence, transparency, use of a formalized structure and process, and the people involved who bring their expertise and passion to the organization.
Richard I. Cook, MD, professor of Healthcare System Safety and chief of the Patient Safety Division at the Royal Institute of Technology in Stockholm, Sweden, who is recognized world-wide for his research in human performance, complex systems failures, and medical accident investigation provided his insights on the possibility of investigating medical or anesthesia events as in other high-risk industries. Cook stated
“Experience in medical settings shows that high quality, independent accident investigation is possible and can provide valuable insights into the genesis and aftermath of accidents. We conclude that there are 3 important criteria that need to be met to achieve high quality results. First, the investigation must be independent of all the stakeholders. These include the practitioners involved and their affiliated organizations; the health care facility, its management, and its owners; the regulators and authorities governmental and non-governmental; and the patient representatives. Stakeholder independence is critical to the conduct and credibility of the investigation. Second, the investigation requires high level technical competence. The investigators and their support staff must be experts in the areas involved and have access to high quality tools and methods for forensic examination of the setting, technologies, and physical data. They must be skilled at interviewing all those involved in the situation at hand. They must be able to critically analyze complex data and to write and present their findings in clear, unambiguous terms. Third, the investigation must begin immediately. In contrast to transportation accidents, medical accidents take place under a wide variety of circumstances, leave little forensic evidence for examination, and rarely halt work in the effected facilities. The investigating team must be able to appear at the site, secure forensic materials, and begin debriefing participants within a day, preferably within hours. There are substantial technical, social, and legal obstacles to high quality investigations. Our research, however, shows that these are not insurmountable.”
John H. Eichhorn, MD, professor of Anesthesiology and Provost’s Distinguished Service professor at the College of Medicine, University of Kentucky Medical Center and a consultant to the APSF Executive Committee believes that a specialized anesthesia accident investigation by an independent party is a good idea, but enormous pitfalls would make it impossible to carry out. He noted that members of the APSF had proposed this idea as early as 1990, but that logistical, personnel, financial, and medico-legal constraints would prohibit its success. Organizing a team of highly qualified experts who would be readily available to travel on a moment’s notice to a anesthesia incident site would require significant financial resources and it would be unclear who would own the findings and recommendations from the investigation.
David C. Epperson, JD, of the law firm Epperson and Owens in Salt Lake City, UT, brought the expertise of a liability law defense lawyer in both health care and aviation. He suggested that the NTSB process has access to all the participants and parties to an accident, something that would be difficult to achieve in health care. He indicated that even to contemplate a health care investigation system would require certain legal protections. The Patient Safety and Quality Improvement Act of 2005 conveys some protection to “patient safety organizations,” but the strength of these protections has not yet been tested in court. Although it is not admissible in court as proof of causation, the final report and analysis of the NTSB is a public document that can breed lawsuits in aviation. Would not the same be even more true in health care? Mr. Epperson noted that in his opinion even the NTSB’s analysts sometimes “get it wrong,” with direct experience on his part with aviation accidents where the defense has proof of a different proximate cause than that indicated by the NTSB.
A spirited discussion and question and answer period ensued and was moderated by Matthew B. Weinger, MD, APSF secretary and Norman Ty Smith chair in Patient Safety and Medical Simulation and professor of Anesthesiology, Biomedical Informatics, and Medical Education, Vanderbilt University School of Medicine, and Dr. Jeffrey Cooper, APSF executive vice president and professor of Anesthesia, Harvard Medical School.
Some suggested that currently the local “root cause analyses” are often not performed very well and that attention should be focused on making the local investigations more solid. Others suggested that a federal agency would not be appropriate but a private organization using the same methods might be applicable. One comment was that investigations should focus on the positive learning from what was done right as well as the critique of negative aspects.
In summary, the panel aired many of the key issues about the desire for better processes to extract the maximum organizational learning of the health care system from the analyses of adverse events. Many panelists, and many in the audience embraced the ideal of an investigating body that combines independence, technical competence, a focus on safety and learning, and the ability to execute rapid-startup of investigations using a similar methodology to that used by the NTSB. However, the barriers and pitfalls of such a system seem daunting, particularly the medico-legal issues, the ability to obtain full participation of all parties in a rapid fashion, and the high cost and complexity. The opportunity cost is also high—would the same effort and investment yield more patient safety if it was devoted to addressing problems that we have already identified by other means but have yet to solve? The goal of independent expert analysis remains enticing, but may remain elusive for the foreseeable future.
Dr. Gaba is Associate Dean for Immersive and Simulation-based Learning, Professor of Anesthesia at Stanford University School of Medicine, Director of the Patient Simulation Center for Innovation at the Veterans Administration Palo Alto Health Care System and member of the APSF Executive Committee.
Dr. Lee is Professor of Anesthesiology and Neurosurgery at Vanderbilt University Medical Center and Co-editor of the APSF Newsletter and member of the APSF Executive Committee.