In the 1980s the Commercial Aviation Safety Team (“CAST”) determined that eighty five percent of aviation accidents and fifty two percent of all fatal aviation accidents were the result of cognitive error by pilots. CAST pushed for the use of systems, checklists and foolproof communication systems to reduce the human element in aviation. The result was a fifty percent decrease in aviation accidents. The success of the systems method in aviation led healthcare safety advocates to campaign for the application of the systems model to tamp down the impact of human frailty on adverse healthcare outcomes. The systems model is now a fact of life in high risk medical procedure, pharmaceutical administration and other healthcare delivery modalities.
You can see it in the operating room when a nurse paints a sign “the other leg” on your healthy limb or where two nurses must sign off on a prescribed medication. The interest in the systems application to quality improvement continues unabated. See Kahn, et al. , “To Err Is Human: Building A Safer Healthcare System.” National Academy Press. Washington, D.C. (2000); Stripe, et al. “Aviation Model Cognitive Risk Factors applied To Medical Malpractice Cases.” (JABFM.org.);Inglehart & Fleming, “New Patient Safety Uses Aviation System Model.” Health Affairs Blog, April 13, 2009.
Since the 1980s and the onset of the digital age the aviation model morphed into a much more pervasive systems intervention through the surge in computer driven maintenance and operational applications, not unlike parallel developments in healthcare. The crash of Air France Flight 447 A330 Airbus over the Atlantic Ocean on May 31, 2009 thrust to the fore two separate, but related computer systems driving both airplanes and healthcare, albeit to differing degrees. The Airbus utilized the Aircraft Communication Addressing and Reporting System (“ACARS”). The onboard computers sent ten automatic maintenance messages to Paris indicating failures in the plane’s electrical systems and decompression of air in the plane’s fuselage.
The current proliferation of medical devices with direct internet monitoring and reporting capacity provide the potential and fact of real time reporting of human body systemic failure to medical record repositories. The healthcare mimicry of aircraft maintenance information technology is upon us and we will likely embrace the use of “wearable” computers with growing fervor in the future.
Despite the apparent differences between healthcare and aviation, the information exchange paradigm used to insure quality in the aviation industry can be implemented to reduce the impact of the aforementioned system errors in health care delivery system (sic).
Martin, Lance, “Reducing Medical Errors Using an Aircraft Maintenance Information Technology (IT) System Model,” Dartmouth undergraduate Journal of Science. Dujs.dartmouth.edu/2004s/healthcareit.pdf.
Despite Mr. Martin’s optimism and enthusiasm there appears to two principal kinks in both the aviation and healthcare systems that need to be addressed. The first relates to the information systems. There is no question that computer monitoring systems can generate a great deal of useful data. The critical question is how to sort and present the data in a format and application that can realistically serve the needs of pilots and practitioners. There needs to be assurance that the data is correct and not affected by defective “pitot tubes” or other malfunctions and a means to sift and prioritize mountains of data so as to provide real time, meaningful input to those sitting at the controls.
The second concern relates to the ultimate control of the mission. The Airbus A330 contains the latest in fly-by-wire technology that substantially limits the operational control of the aircraft by the pilots. This is an exclusively electric system, without manual or hydraulic features. There is a debate ensuing in the blogosphere as to the degree of control, if any can be wrested from the computers by Airbus pilots in the event of an emergency. (Think of HAL in 2010:A Space Odyssey. That would be next year.) The issue is at what point should human beings be able to reassert control from computer systems, particularly where onboard computers are juggling conflicting information from various sensors?
Did the computers aboard Flight 447 malfunction causing a crash or did a last ditch effort by the pilots to regain control interfere with the failsafe functioning the planes systems? Was it a combination of both or an unremediable "act of God?" We may never know the answer.
Remember, more often than not, an airliner goes down at the end of a long chain of unrelated, seemingly innocuous decisions, malfunctions, mistakes and external factors. Remove any single link (or even change their sequence) and you have an on time arrival. Miles O’Brien Uplinks, June 19, 2009. trueslant.com/milesobrien/2009/06/10the-paradox-of-simplicity/.
In the future computer systems will have an ever increasing role in healthcare decision making and treatment as the incorporation of artificial intelligence applications into electronic medical records provides a whole new platform and paradigm for healthcare delivery. The healthcare industry will eventually face the conundrum of when and under what circumstances a physician will need to regain control over the treatment process. The crash landing of U.S. Airways Flight 1549 in the Hudson River earlier this year underscores some legitimate concerns to keep in mind. That crash also involved an Airbus with a similar air flight control system, which may have unnecessarily shut down the plane’s engines after the bird strike, causing the plane to lose all thrust. There may come a time when we need a doctor with the nerve, experience and professionalism of a Chesley B. “Sully” Sullenberger, III to seize the controls and manage a life threatening emergency.