Engineering New Solutions and the Firehouse Principle
For some time now I have fought to deny an image that keeps popping up in my mind: the simplistic view of American Healthcare as a mechanical device essentially assembled backwards. Lately, however, I've grown unsure that that image is as other-worldly as it originally seemed.
Consider the basic interest we all hold as Americans in having our healthcare system promote the general health and welfare of the population. We have failed utterly to define over the years what that really means. Instead of something proactive, the average American and the average healthcare professional both tend to think of a good healthcare system as one that dynamically works to discover new drugs, procedures, and services, but a system that only works to promote the overall health of the society when those advanced services and medicines are actually used (as measured by the existence of a billing code for each one). In fact, that is the system we have. When called on to put out fires, it works fairly well - even if it costs too much.
But now envision a system with the same advanced science-based capabilities that does not wait for a fire to start before sprinting into action. Envision a system dedicated to dramatically decreasing the public need for those services, and measure it's effectiveness by how many billing codes did not have to be processed for a given segment of the population.
Sounds somewhat bizarre, right? It would be, to coin a phrase, the Firehouse Principle, a frame of reference in which the citizenry willingly pays for the best firemen and the best equipment and the best training, but in which the medical professionals have a concurrently intense and dynamic desire to never see those superlative services actually used.
In fact, it would also be akin to the way we now view seismic safety in earthquake prone regions (which, believe it or not, includes 39 American states). While we need excellent emergency response training, assets, equipment, radios, and coordination to handle what happens when the earth shakes to the ground the things we build upon it, the best use of our resources is to change the way we build so that our buildings don't fall down on their occupants to begin with. This is the difference between "Mitigation" of a hazard, and the "Emergency Response" when a hazard becomes a fact.
It is much the same as mitigating the occurrence or even the effects of uncontrolled coronary artery disease in order to prevent the application of emergency response protocols.
More to the financial point, what would our world look like if we could find a way to pay our physicians and nurses and hospitals and pharmacists more and more as the need for their services grew less and less? How could we - how would we - do that? But can't a nation that reached the moon in less than ten years with no headstart figure out such an upended systemic reformation? And you just know I'm enough of an optimist regarding the American Spirit to yell "Yes!."
But let's take it back to the original, somewhat-strange vision. Doesn't this dilemma more or less have the feel of a mechanical problem? A mis-assembled device? Pull the lever and everything connected to the pulley goes down, but it was designed to go up? (And yes, I know, those last two were not complete sentences. Consider it literary license - which may need renewing soon).
Despite the immense complexity of the human system that is Healthcare, this basic relationship is clearly backwards. When doctors and hospitals do the right thing and work hard to change and embrace proven best practices, they usually succeed in minimizing the occurrence or complications of specific disease processes - which in turn decreases the need for repeat services, and thereby decreases everyone's...income!
But do little or nothing to address known disease processes (such as diabetes, depression, asthma, congestive heart failure, and coronary artery disease, all of which can be controlled with startling predictability), and the health of the nation gets worse, but we're paid increasingly well for the medical response.
We all need to realize that this is not a moral failure, it is a systemic malady in what Kaiser Permenente's CEO George Halvorsen calls a "Non-System" of Healthcare (George is also the one who most recently identified the above 5 chronic diseases as costing us more societally than all the others we combat - his new book "Health Care Reform Now!" should be required reading).
But again, how do we respond? Here is the entirety of American Healthcare bogged down with financial, legal, safety, professional, and insurance problems of planetary size, and yet this one, basic reality won't go away: Our non-system can promote either the use of myriad medical services, or it can improve everyone's overall health, but the present machine cannot do both. It's mis-wired. It's upside down.
Now, after a very long evening of creative debate lubricated with a variety of wine or scotch or even iced tea, most of us will suddenly arrive at the brainstormy point of realizing that the American Engineering community might just have a way of understanding a few things about this upside down machine - even though in the same breath it seems silly that the highly disciplined profession of designing things could even begin to understand the dynamic kaleidoscope of healthcare. That prejudice, of course, flies in the face of the reality that human factors engineers and process engineers have long been involved in melding humans to their systems, inclusive of the American Hospital.
But could structural engineers help us in any way?
I will not try to report to you here that anyone seriously has a "yes" answer to that question, but it is fascinating and provocative that the Institute of Medicine at the National Academies has already formally started exploring at least part of the question of how the engineering community might give us some greatly needed help in redesigning the structure and the operations of health services. In fact, I have been privileged to participate in just one of several workshops in this area with thought leaders who are actively evolving the scope of this question, and coming up with some promising synergy. I do not have permission to reproduce any of the work-in-progress of the workshops, but I can offer a glimpse into this arena though my own contribution, and what follows is a draft paper expanding on the address I gave in April to the group. I warn you in advance it's a bit lengthy, and it is in fact aimed at providing a view of how aviation dealt with some of the same structural problems, but if it provides some food for thought, it might be worth scanning.
As it is a draft, all errors are mine and apologized for in advance. Your comments and counsel, as always, are welcome.
--------------------------------------------------------------------------------------
IOM Workshop: Engineering a Learning Healthcare System
Edited and Revised Talk - John J. Nance - April 29, 2008
SESSION 3: CASE STUDIES IN TRANSFORMATION THROUGH SYSTEMS ENGINEERING
AIRLINE SAFETY
John J. Nance, J.D.
A Founding Member of the National Patient Safety Foundation at the AMA
While it would be hyperbole to say that the solution to much of what troubles American Healthcare can be found in engineering disciplines, I truly believe that engineering and the engineering community can provide unprecedented expertise and contribute substantially, if not pivotally, to the national task of creating order out of the chaos that is American Healthcare today.
This is not to demean healthcare, but merely to be frank about the reality that a cottage industry based on individual physician autonomy has grown to gargantuan and unmanageable proportions on a thoroughly inadequate organizational base. Hospitals were few and far between a century ago, but the amazing momentum involving science-based advances in medicine and incredible advances in equipment have been essentially stuffed into the same archaic mold, and it is clearly not working - either in terms of reliable and safe delivery of the best care, nor in terms of the best value. Engineering philosophies, approaches, and discipline can't cure everything, of course, but where medicine itself has been unable to conceive of a structural approach to so many different aspects of the problem by using traditional methodologies, the benefit of new thinking from external disciplines may be able to break the logjam.
American Healthcare needs to find a balance between two extremes: On one end of the spectrum is the clearly inadequate 1800's model of the individual doctor and the hospital as a sort of farmer's market which provides the beds, nurses, and lights. On the other end is a rigid, thoroughly mechanized approach to healthcare in which autonomy is limited to small differences in techniques physicians may use within the context of inflexible procedures and full employment directly by healthcare providers. Obviously, neither extreme can take advantage of both the amazing advances in science-based medicine, nor the eternal need for the dexterity, intellect, and analytical abilities of individual physicians (as well as the human caring-based attention of nurses as the bedside eyes and ears of the physicians). One extreme needs no engineering, the other would grossly overuse both systems engineering and the lessons from such fields as airline safety, but in fact a careful balance must be crafted that preserves the humanity and the individual expertise of healthcare practitioners while at long last providing an efficient and workable structure that serves the prime directive: doing the best possible job for the patients and enjoying the profession.
These points of overture are important in any discussion that looks outside medicine for clues and answers, and that is especially true when seeking to find the applicable lessons from airline and aviation safety. That transference - as well as a brief and cogent background to how the airlines in the U.S. have achieved what they have achieved (almost perfect safety) - requires a basic understanding of the strategy, and not just the tactical details of individual training programs and methods.
My hybrid professional background melding aviation and medicine, for one, stems from 18 years of experience in translating to healthcare the very surprising human lessons we were forced to learn in the aviation industry (along with other fields such as nuclear power generation). In a nutshell, by the late seventies aviation had reached the effective end of its ability to significantly improve safety through merely mechanical and procedural means, and only startling and industry-changing lessons from human factors and performance disciplines enabled the airline industry to take the final step towards zero accidents and incidents.
In many ways this period of almost unnoticed transition was from the principles of mechanical/aeronautical engineering to the acceptance of the principles and benefits of human systems engineering. But clearly, at every step in the process, the discipline that engineering brought in helping the airlines accept the realities of human failure potential, and the ability that gave airline safety leaders in imposing better order and function, guided the sometimes painful transformation from myopic focus on mechanical reliability to focus on overall systemic reliability. In other words, we finally had to stop fine-tuning the machines and the black boxes as the only bulwark against accidents and admit that when even the finest airplanes could be flown into a mountain by a well-trained but confused and distracted aircrew, the failure modes of the carbon-based human being would now have to be addressed and understood. But the important point for this discussion is that the very same elements of transition are needed in American Healthcare, and even more dramatically so, since both the procedural/mechanical side and the human systems engineering sides of healthcare are equally undeveloped and undisciplined. So, to further this discussion, let me focus on aviation's experience.
For perhaps ten years now there has been an accelerated understanding that aviation's experience in transitioning from a high-risk to a low-risk-high-reliability industry has something to say to American Healthcare. The problem has been oversimplification in translating that message. Too many folks on both the aviation and medical sides of the fence have believed that aviation's best technology transfer to healthcare consists of a few specific programs and methods, such as CRM (Crew Resource Management) courses and checklist procedures. The belief, which turned out to be extremely naïve, was that such tactical solutions could be neatly dropped intact into medical arenas in order to gain the same level of dramatic improvement they brought in aviation. (CRM, by the way, is essentially a discipline which recognizes that no one person, leader, captain, physician or otherwise, is capable of perfection, so that the best prophylaxis against human error causing disasters is to continuously utilize all the professional talent and cognitive abilities of all participants in a discipline of collegial communication that can be codified, trained, and required).
The reality is that while the principles of each of those tactical measures can be of great benefit to medicine when properly translated and adjusted for the realities and complexities of medical practice and application, there is a far richer lode of lessons and benefits to be derived from aviation's experience. In fact, aviation - and the airline industry in North America - have much more important lessons to impart within the context of the difficult journey we undertook to deal with the inescapable and permanent inability of humans to be perfect.
Aviation, of course, is not inherently smarter systemically than healthcare about preventing disasters. But the fact that our failures were both very public and very frightening to our future customers, and the fact that our death tolls were counted in large numbers with each major accident, meant that we had to come to grips with the last remaining unsolved cause of airline accidents, human-mistakes decades before healthcare. We simply did not have the luxury, if you could call it luxury, of sitting back and waiting for improvements to evolve. We had to solve some of these problems and we had to solve them quickly. More specifically, we had to figure out why dedicated and intelligent and well-meaning aircrews continued to fly mechanically perfect airliners into the ground or otherwise create horrible accidents from purely human causes - the so-called "pilot error" accidents.
In truth, the safety challenges that the airline industry faced through the 1970s was nothing short of perplexing. We had enjoyed an incredible run of very effective progress in airline safety from the dawn of commercial aviation in the late 1920's all the way through the dawn of the jet age in the sixties and into the seventies. In fact, the "curve" of major accidents plotted against time had been diving downward at an amazing rate as the machines were greatly improved and instrument flying became sophisticated and the new jet engines introduced far greater reliability. That descending curve also represented greatly decreasing passenger fatalities, and while our metrics left something to be desired, we clearly improved many orders of magnitude over time as mechanical failures triggering accidents became increasingly rare. Boeing, McDonnell Douglas, Convair, and later Airbus all learned how to build significant redundancy into their products, and helped pioneer the principles that no single or even dual failure of any component should ever result in the loss of control of an airplane. In fact, in one of the earliest instinctive iterations of human factors engineering based on the expectation of constant human propensity for failure was the decision to keep at least two pilots in each commercial cockpit specifically to provide a human backup system. For the most part the positive safety trends - albeit mostly mechanically based - continued into the 1980s and 1990s. With the exception of a faulty cargo door latching mechanism (United 811, 1989, south of Honolulu), a destroyed engine and flight control system in a United DC-10 ending in Sioux City, Iowa in July of the same year, and the loss of the upper forward fuselage of a highly corroded Boeing 737 belonging to Aloha Airlines south of Maui in 1986 (and the fuel tank explosive loss of TWA 800 years later near Moriches, New York), by and large it had become a rule that when an airliner was destroyed with or without the loss of life, in over 90% of the cases the primary contributing cause was human mistake or human error. Even the term "pilot error" (which bespeaks a professional discretionary mistake, such as making a conscious decision to violate the rules with catastrophic results), came under attack as inadequate to describe the state of being human, which was also the state of being able to make errors that sometimes caused accidents.
That trend curve for major airline accidents, especially in the United States, by the seventies had flattened and was riding just a few points on average above absolute zero. But it refused to descend to absolute zero. In other words, while airline flying had become amazingly safe and reliable, especially with respect to mechanical accidents, no amount of industry effort, FAA pressure, or pilot training could seem to completely eliminate the human-caused disasters, and no accelerated application of the traditional engineering solutions seemed to improve the rate.
But in the eighties a true revolution, however quiet and unnoticed, began to change the equation. As a direct result, approximately sixteen years later, the airline accident death rate for U.S. airlines finally hit bottom and remained at zero for almost 5 full years - an unprecedented and stunning achievement. Although the 5 year zero-accident record was broken by a crash in 2006 in Lexington, Kentucky, the passenger death rate in U.S. service has remained wonderfully flatlined since. What happened? We had finally learned to deal directly and forcibly with the immutable truth that aviation is a human system, and that humans will never be able to fly or operate without making mistakes. The pathway, in other words, to perfect safety was through the process of building a system that fully expected and was ready to safely absorb human mistakes. The engineering-based disciplines that evolved in the airline business (and aviation in general) from that pivotal recognition are loosely known as human factors engineering, but they include systems engineering as well and borrow heavily from sociology, physiology, and behavioral science.
How did it happen?
Before the industry recognition dawned in the early eighties that we had never really addressed human failure (except to ineffectually order humans not to fail), there was a growing, mostly-embarrassed silence about the prospects of ever fully eliminating passenger deaths and disasters. We began to whisper that a certain number of accidents might be the cost of doing business, that maybe accidents, in fact, were inevitable in a system that lofted as many as 3,000 flights daily over the U.S., and measured it's passengers in many tens of millions annually. Perhaps it was unrealistic to expect that we could ever have a perfectly safe airline system without parking every airplane. And, as the airlines came under tremendous cost pressure during the beginning of the 1980s with the ruinous influence of airline deregulation and cut-rate competition, established airlines began looking in desperation at reducing costs. In that environment, the worry that exponential new investment in maintenance and training and new electronics would be needed to realize only an incremental improvement in safety (given that there were already so few crashes) did little to generate an enthusiasm for expanding safety measures, or investing in new disciplines such as CRM (which was in its infancy at United at the time). The heavy price of small improvement, in other words, furthered the expedient idea that maybe we would just have to accept a continuous low-grade drumbeat of accidents as the cost of having an airline system. And, of course, this was not an illogical argument at that time. In fact, one major airline executive rather infamously replied to the question of why his airline did not spend millions to establish a safety department by saying: "We don't need one. That's why we have insurance."
Before the Renaissance of human factors and human failure realizations of the 80's, the airlines had successfully used systems engineering principles in many different ways (sometimes without even labeling it correctly) to develop high levels of mechanical and operational reliability. Across the industry, we had developed sophisticated feedback systems for learning rapidly about mechanical problems, and those systems included so-called Airworthiness Directives issued by the FAA (the strongest legal directive that the FAA can issue to effect mechanical changes), as well as lesser-urgency service bulletins transmitted to the entirety of the commercial aviation world within and without the U.S.. In addition, there were a galaxy of methods by which the airlines could communicate with each other and communicate with the FAA and the National Transportation Safety Board, inclusive of a steady procession of task forces and special industry groups working voluntarily with government on problems of special concern (such as the revelations in the late eighties about the susceptibility of aircraft structures to accelerated corrosion and fatigue in high-salt environments following the Aloha accident in 1986). To a certain extent, those systems have all now matured (along with individual reporting systems such as NASA's ASRS) to the point that any significant problem discovered in commercial aviation can be fully discussed and transmitted worldwide to every operator within hours. Aviation, in other words, worked hard to learn serious lessons about maintenance and training only once, the FAA pushed airline safety greatly by working with, instead of against, the industry.
In the same period of the seventies through the nineties, the major airline manufacturers developed, under Part 25 of the Federal Air Regulations, exquisite redundancy in their designs to the degree that the anticipated failure rates of most of their aircraft and components had a dizzying array of zeros to the right of the decimal place before registering the first failure potential. Through backup systems and a worldwide methodology of preventative maintenance (pulling and replacing or overhauling components large and small long before their first anticipated failure range), even the so-called "dispatch reliability" of airliners exceeded the most optimistic expectations. In addition, airlines developed processes for the computerized tracking of maintenance, parts, and all operational elements, including crew scheduling, reservations, ship scheduling, dispatch, and coordination of all functions, squeezing the most out of the rapidly developing capabilities of computers. They developed computer-assisted standardization of just about everything done in the maintenance hangers, in the cockpit, and even in operations and all of these elements were honed continuously because they were the most cost-effective method of doing business. Airlines realized that far beyond the direct costs of an accident, in a heavy competitive environment they could simply not afford the type of reputational catastrophe that any major accident would trigger.
All of the mechanical and computerized systems were largely in place by the end of the 1970s, but, as previously pointed out, crashes still happened, usually from human failure. In 1982, an Air Florida Boeing 737 crashed on takeoff in a snowstorm in Washington D.C., killing all but 5 of those aboard. There wasn't anything wrong with the airplane. In 1985, an Arrow Air flight charted to bring U.S. troops from the Middle East to Kentucky crashed in Gander, Newfoundland, killing all 256 people aboard. While there is still controversy about that crash, the cause was attributed to the crew departing with ice on the wings--again, there wasn't anything mechanically wrong with the airplane. A Northwest Airlines plane crashed in Romulus, Michigan in 1987 due to pilot mistake in failing to extend the flaps and all but one died. A year later, a Delta flight at DFW Airport also tried to take off with the flaps up and crashed, killing 17 people. The flight crew survived, and were astounded at the NTSB's finding that all three of them had missed clear signs that the flaps had not been extended. Three highly trained, highly qualified capable human beings, with no intent to do anything wrong but fly to Salt Lake City with passengers that day, had caused a major accident, but all three had seen, and were willing to swear they had seen, instrument indications that the flaps were in the correct position (15 degree extension). They weren't.
Given events such as these, the airline industry realized by the early-1980s that such tragedies would continue unless the industry adopted radically different practices and, for the first time, addressed not just advertent human failure, but wholly inadvertent mistakes. To do that, the industry had to adopt more than major changes. It had to change its philosophy and, most importantly, massively change the culture of airline piloting.
Many who look at the aviation industry's excellent safety record today erroneously think it was merely the result of engineering success based on the mechanics of the operation, based on systems, based on getting people under control and completing more and more checklists. In fact, even members of the industry itself are abysmally unaware of the silent revolution in culture that took place under their noses and transformed our ability to prevent the human-mistake driven accident. More to the point for this workshop, the changes I referred to as a Renaissance in thinking during the eighties and nineties have helped us create a completely new paradigm that many realize can be transferable to healthcare. And, in fact, I and many others have been doing exactly that with solid success for the last number of years primarily by focusing on training healthcare professionals in the new discipline of how humans fail and what to do to create a living human system that can prevent those failures from ever hurting a patient. That training goes completely against the traditional, autonomous and infallible grain of healthcare, especially in relation to physicians, in that it holds as a principle element that while individual humans - including surgeons - are forever incapable of being perfect, interactive, collegial teams of humans can achieve perfection. Indeed, this primary point is the legacy of the CRM revolution in airline cockpits alone, where we have saved countless lives and aircraft in the past twenty or more years by requiring more than one human mind to weigh in when something appeared amiss, and to use a teamwork approach based on the common goal of flight safety to approach self-correction and safe operational decisions. Gone immediately in such atmosphere is the angry autonomous leader who has to discipline a subordinate for speaking up by berating, belittling, and ignoring. Gone, as well, are the previous drumbeat of accidents in which a subordinate had the key to save everyone, but could not pass it to the leader.
The resonance between healthcare today and the airline industry yesterday is astounding, in that every physician, nurse, and other healthcare professional is trained, essentially, to be perfect and never make mistakes, just as we were. Worse, the system is built the same way ours was on a solid and inflexible expectation of human perfection with few if any buffers designed in to allow for major human mistakes. In the airline industry, it was sad to think that thousands of man-years of engineering had been poured (with great success) into providing backup systems for even the most arcane failure modes, but when it came to engineering for human failure, it was traditionally a matter of simply and vacuously ordering the human not to fail. And equally appalling in light of what we now know was the utter lack of emphasis on human-to-human relationships as the platform on which true communication, coordination, and self-correction occurs. Similarly in medicine, there is no traditional expectation of human mistake in good doctors and nurses and pharmacists, therefore there seems to be (in the traditional view) no valid reason to need backup and buffer systems to absorb mistakes that the system refuses to accept as possible in the first place. The eternally self-defeating aspect of that disastrous logic loop is obvious to us now, but has been all but invisible in the past. The lesson from the airline industry then is that making that reality visible - the need for buffers against the normalcy of human error - is a prime safety component for any human system. Of equal import is the reality that healthcare, just as previously the case with the airline culture, has a built-in expectation of hierarchical autonomy that is automatically challenged by any subordinate speaking up to report a mistake or concern. Excessive sensitivity on the part of the subordinates to the feelings (and potential irritation) of a senior created major culture-based reluctance to point out concerns, problems, or even impending disasters, lest the leader become angry or irritated at the implicit suggestion that he or she was in error. Leaders, after all, were trained never to make mistakes. But that left only one carbon-based mind operating in an airplane (or an OR), and other qualified professionals sitting in silence, even (in the airlines) if the captain was a very gentle individual who wanted to hear from his crew. Stated as gently as possible, the resulting, traditional hierarchical "culture" did not encourage all knowledgeable parties to contribute fully to maintaining safe operations, nor did it enable staff to challenge decisions when safety was at risk. And it was those very failures that kept us from improving the safety levels and preventing that last tier of human-mistake driven accidents.
Perhaps the most important experience that the airlines industry can share with and impart to healthcare is the fact that the airline industry one day essentially declared that no human could be perfect, and that no team could function as a team without a degree of collegiality and mutual respect. We proceeded to build a system around those assumptions with the intention and purpose of being able to construct buffers and backups that can absorb all the reasonably anticipatable human failures that might otherwise grow into an incident or accident. And history shows that we have succeeded.
There are, we learned, 3 distinct tiers to a safety system. In the first tier are all the training and indoctrination and agreed to or imposed professional methods such as checklist compliance and "time-outs" that are designed to prevent human error in the first instance. Nevertheless, when we understand that some human error will occur despite our best efforts at standardization and training, we than have to construct a Tier 2, in which we specifically build those buffers and barriers that will catch and cancel out the effects of human errors and latent system failures. But there is a Tier 3, and in that one we have learned that even after highly effective work in preventing and then screening out the effects of mistakes, we will still occasionally fail catastrophically if we don't enter every operational sequence with a healthy expectation that there is a 50% chance of failure. In other words, by doing away with the assumption that now it is nearly impossible for something to go wrong, and by constructing collegial teams whose members have no hesitation in communicating with each other for the good of the mission, we construct a systemic approach that assures our leaders are ready and instantly willing to consider even the most tenuous concern as potentially valid, and "stop the line," the operation, or the takeoff until the team and that leader are sure safety is not threatened. Thus, either a junior flight engineer or a new circulating nurse would get an instant and serious audience by saying "I'm not sure, but I think something's wrong," rather than having to overcome a massive group presumption of normalcy. That one change - the Tier 3 approach - can be the last tumbler in the lock we're trying to construct against catastrophic patient injury or death from preventable medical, human mistakes. But to institutionalize such procedures takes a systemic approach that, quite simply, is foreign to American Healthcare experience, which is why asking from help from the engineering community is such a vital and important step.
There are 3 basic ways that we fail as human beings--by making mistakes in perception, assumption, and communication.
Perception failures include, for example, a flight crew's failure to recognize that their aircraft's wing flaps are not properly extended for takeoff. Mistaken assumptions include such mistakes as the assumption of two KLM pilots in 1977 that their Boeing 747 was cleared for takeoff when, in fact, they were not. Another 747 had missed a turn and was sitting sideways on the runway ahead unseen in the fog (The decision to start the takeoff was a human mistake fed and nurtured by a bad cockpit culture and it resulted that day in the loss of 583 lives). The third is botched communication, a human propensity which healthcare and aviation have in common. 12.5% of the time in human verbal communication people who otherwise understand each other, do not. The old phrase "I know you think you understood what you thought I said, but I am not sure you realize that what you heard wasn't what I meant?" is a clue to that universality of misunderstanding. It is a universal malady affecting us all, but even then, we've learned that reading back a clearance or a medical order can reduce the potential for mistake to below a half percent.
Aviation had to learn these basic failure modes and instead of fighting to deny them or ordering them to not occur. We had to learn to inculcate the expectation of such failures in everything we did. So must Healthcare. But in order to accept these radical realities operationally and culturally, we need a structured, engineered approach that - taking fully into account the largely autonomous and maverick history of healthcare culture - can find ways of providing an effective framework within which such sub-disciplines as minimization of variables, collegial team communication, and the three tiers previously discussed can be deployed as a standard operating methodology. Equally important (and not just to avoid the charge of creeping cookbook medicine), the structure that results must support and depressurize the vital need to nurture physicians in using their cognitive, analog diagnostic and surgical skills to do the things checklists, machines, and procedures alone can never accomplish. That, in fact, is the proper balance between a structure in a human system that better enables through technology and enlightened methodologies the ability of the humans within that structure to practice what humans do best: apply judgment, skill, and reason.
We cannot wire in an expectation of perfection in a human system and not create and nurture disasters. We cannot fail to accommodate human attitudes, or feelings, or physiological limitations without perpetuating a societally unacceptable level of patient injuries and service quality. What Healthcare needs from the applied and unique expertise that engineering can provide is a structure that legitimizes and inculcates known best practices, eliminates the need or latitude to reinvent each procedure, and provides the best possible operational buffers against eternal human fallibility while providing the supported latitude within which healthcare professionals can practice with caring and engaged attention.
Let me conclude this way: This exploration of bringing the engineering community in to assist healthcare probably heralds the most important advances in changing the way we've always thought of the problems of patient safety, service quality, and healthcare delivery since we first began to recognize that we had national problems with what George Halvorson, CEO of Kaiser Permanente, calls our non-system. In order to bring order out of chaos, we need help outside the traditional methods we've attempted to apply in the past. And we certainly will have to construct the needed changes from inside, since neither Congress nor any of the candidates in this election seem to be able to address much more than healthcare insurance. While aviation and the airline industry can confirm that we've been down the same road, we can also say that both mechanical and systems engineering provided the keys - both to building reliant airplanes, and to staffing them with imperfect humans who, working together and as colleagues able to communicate without barriers, could accomplish what a single commander could not. If we keep that in mind and borrow liberally from other disciplines, we can at long last engineer an omnibus system that works, that works safely, and that can be financially sustainable.
