4th International High Reliability Organizing Conference: Making HRO Operational Washington, DC April 21, 2011

Present Daved van Stralen, M.D., Loma Linda University School of Medicine Karl Weick, Ph.D., University of Michigan Mark Chassin, M.D., The Joint Commission Chris Hart, Vice-Chair, National Transportation Safety Bureau Tom Mercer, Rear Admiral, USN (Retired)

Panel Discussion Daved van Stralen: We’re going to move right on to the panel discussion. As you know, Tom Mercer, as the captain of the Carl Vinson, had invited Karlene Roberts to study his ship and how can he improve the performance of his crew. From that experience, she — not so much how he can improve, but she codified what later became a high reliability organization from his command philosophy. Chris Hart, with the FAA, did some of the early work on error reporting systems, where even though there was distrust between the pilots and the FAA, how can you sanitize the information through NASA so that we can have an honest reporting system? Mark Chassin, who you were just introduced to, how do we in health care, whether you call it recalcitrant or resistive or difficult, a problematic situation, how do we change a large organization like that? Karl Weick, who did the early work with collective mindfulness and looking at high reliability through social interactions, and then characterizing it into his five principles, and then leaving with us how to interact on that. Karl had given us an article a while back to review of an organization that had a tragedy. They implemented high reliability and they had the tragedy. The question is, as we come

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 1

up from this morning’s lecture, we want to implement high reliability, but how do we know we’re there if the events are rare? Karl Weick: I’m going to be reporting or describing to you a single study done in the UK that I think sharpens some of the issues that we’re talking about. Yesterday you heard some sample comments. You’ve heard them throughout the conference that are relevant to the particular study I’m going to talk about. Both Naj [phonetic] and Mark Griffin commented about being in the middle of an investigation so that they couldn’t really say much or be clear about what the interventions would be. Peter Davidson talked in the gas explosion about people knowing that a gauge was faulty while they were continuing to work. And during the question and answer period, as an example, there was talk about how the small events tend to be collected in offshore drilling exercises. Organizations have some kinds of momentum and inertia and sequences that both make reform itself difficult. And their operations, as they continue, can sometimes undermine even the best of reform efforts. And these kinds of natural processes are the ones that are at the heart of the article that I’m going to talk about just briefly. The article is titled, and you will see why we picked it out, the title is “Failure to Mobilize in Reliability-Seeking Organizations: Two cases from the UK Railway.” The article and the research was conducted by J. S. Busby. He goes by those initials, so I know no more than that. This shows up in the Journal of Management Studies in 2006, readily available, so you can draw your own kinds of conclusions from it. He basically asked the question, what tends to undermine reliability seeking in organizations? And that’s what I take this panel to be about. So we’re talking about reliability seeking that turns out to be inadequate or it’s not working, and so you have this odd — odd to me — odd specter of an organization that’s actively reliability seeking, seeking reliability in its operation, known problems, and it experiences catastrophic failure while it’s doing these reliability efforts. If you look at the reliability literature, and we’re talking about failed attempts at reliability, there are two different pieces that you see in there. One of them talks about organizations that are, by design, vulnerable, so that we’re talking about, for example, tight coupling among units and excessively complicated interactions, and that increasing the vulnerability of an organization in the sense that if something goes wrong, it ramifies, it explodes. And then there’s also a different kind of literature that talks about degradation over time, a small error accumulates, accumulates, errors pile up, and we finally find an organization is in a vulnerable position, but the focus is on degradation over a period of time. The two catastrophes, railway catastrophes in the UK that were looked at were the Ladbroke Grove disaster in which 35 people were killed. Essentially, what happened was a train went past a signal that it was supposed to stop at, crashed into the rear of another train, 35 people were killed. This is called a signal passed at danger as a category of an accident, and there have been several of these, not collisions, but signals passed at danger that the organization at that time was actively seeking to deal with.

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 2

The other one is the Clapham Junction disaster in — this one was in 1988, and due to a technician wiring a signal incorrectly, when a train passed past the signal, rather than turning red, it stayed green. The train stopped for a station and another train coming around the bend collided with it from the rear, and in that incident there were 35 people killed and 500 injured. Now, both of these organizations at the time were actively trying to develop more reliable operations. The reasons that the catastrophe happened anyway are — I’m going to put them in general terms — these are Busby’s terms — so that you can get a sense of how they might apply to your own organization, and this is stuff you already know. These are reminders, if you will, empirically grounded, but they’re reminders of stuff that you’ve heard and stuff that’s already clear. There are problems with the reform itself. You’re trying to get your organization more into high-reliability-like activities, and one of the problems just with reform or just one of the processes that’s inherent in reform is mobilization or trying to turn knowledge into action. A good example of this is what has come to be called massing, m-a-s-s-i-n-g. This is the fact that the system tends to degrade quite a bit before there’s a single large-scale investment in correcting a whole host of problems. So what you’re doing is you’re replacing obsolete technologies that have become obsolete over a period of time all at once, but this means, notice, that you’ve got deficiencies prevailing uncorrected and they’re simply accumulating over a period of time. So the obvious thing, you’ve got knowledge, you want to turn it into action, “We’ll get that thing corrected.” In the meantime, the deficiency continues. Also, as each new problem comes up, some new taskforce tends to be formed in order to deal with that particular problem, and the more taskforces you get, the harder it is to synthesize the product so that each reform doesn’t undermine or cancel or interrupt the preceding one. The second obvious problem in reform is simply to do with coherent lines of development, or continuity would be a better way to put it. You’ve got a lot of personnel changes that are going along, people move in and out of a position. They’re unable to check and verify that in fact an intended plan was put into place. So you get the ironic situation here that a reform gets undermined by a subsequent kind of a reform. Third, there’s a general problem that, in a sense, pulls together this conference, mainly operationalizing or converting general things into particular situations. A lovely phrase they use in this article is what they call local empiricism, which means that the technician stationed at the Clapham Junction signal box had a particular routine that in fact had been flawed. He implemented that kind of routine, and that caused the wiring problem itself. You’ve also got the problem of many little problems happening. They get aggregated. They get averaged as they move up the hierarchy, and depending on those average numbers, you find a problem to be significant or not. This is normal, natural trouble in an organization, the averaging. The problem is what that doesn’t account for is outliers, and it only takes one outlier to blow up a particular activity.

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 3

You also have the fact — fourthly, this is all still reform activities — you’re trying to reform something, and these are things that are difficult in reform. One organization oftentimes — think about the regulators that we’ve been talking about trying to encourage reliability seeking in another kind of organization, but a lot of this has to be done by persuasion. The regulator can only sample. The regulator can’t be there watching continuously, and I’m sure this is the case with the air traffic controller right now. Sometimes the regulation and the regulators are hesitant to take an enforcing action at the risk of demotivating the people who are actually doing a pretty good job. It’s not all the controllers that are on DVDs or trying to deal with sleep cycles, and the difficult intervention is simply the fact of not wanting to have a negative effect on those people who are doing it well. Okay, that’s a cluster of things that you’re going to run into and you have run into in reform. There are organizational problems. You’ve got to keep going while you’re also trying to do reform. There are four things there that can make for failure. One of them is the fact that you’re continuing along and you continuously need to deal with emerging events. Those can tend to decouple an organization. Disarticulate it is the term that’s used here, but just think about decoupling or pulling the organizations apart. You need a fundamental kind of structure to continue through time in order to organize, to keep the thing together, and a lot of ad hoc adjustments over a period of time can make any kind of program, reform, or just continuing operation more difficult. You’ve got the obvious issue of lag. You come up with an idea, you want to do better hand hygiene. It takes a while, it takes time for that to be enacted. The problem here is that as that is filtering into organizations and being gradually and incrementally adopted, other problems are coming up, and so now you need to modify the hand hygiene requirement or availability or how it’s handled because of changing circumstances, and then you’ve still got a discrepancy. So the discrepancy between the concept and the action may well be permanent over a period of time. It’s taking time, it’s getting implemented, something else comes up, you’re still behind, you’re still behind, you’re still behind. That’s, again, normal stuff that you deal with in an organization. So a lot of the stuff that we’re saying is really tough because it’s — high reliability is secondary to the fact of just we’re dealing with organizations. Propriety is another thing that goes on in organizations that can, in a sense, sort of keep the organization glued together, and there are proper ways of doing things in an organization, and that’s what keeps it greased. But notice that reports on accidents that are still pending mean that people aren’t willing to make judgments because the implications of a recent disaster are still not spelled out. This leaves you in the really bizarre position of the system in question, because of a previous disaster, is left — left to continue because it’s in a system that’s already proved to be inadequate. So we don’t want to make an intervention yet, because the findings aren’t final, the system remains in the condition that caused that disaster while we’re waiting for those kinds of results. This is in the interest of doing proper procedures, on the other hand, it’s leading to very high vulnerability. And then the last one is just people coming up with rationalizations for what’s going on, and this blinds us to what we would see in other kinds of settings. During the Ladbroke Grove inquiry, for example, the chairman of the investigation board says, “Was the overall

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 4

feeling that if a signal was passed at danger, it’s a driver problem?” And the answer he got back was, “How can it not be?” Well, at the time, there were task forces in this organization that were working on the problem of signal visibility. It wasn’t a driver problem at all. Because of foliage growing, angles of sun hitting all drivers, it was impossible for them to see the signal other than for just a microsecond. A driver problem? That misses something that was actually going on in the organization, namely trying to work better with the signal visibility problem. So I don’t have any kind of answers to this, and neither does the article itself, but the point would be that organizational processes have built into them the very kind of ordering metrics, the ordering processes that in fact make reform difficult and that can also undermine even the best of reforms. So it’s against that kind of background that I would urge you to listen to the other commentaries on failing in reliability seeking, because when we talk high reliability organizations, it’s easy to sample on the deep-ended variable and say, you know, “Okay, here’s a handful of people who really pulled it off.” There are a lot of people trying to pull it off, and what this article does is spell out some of the reasons why that’s a lot tougher than we may sometimes imagine. Thank you. Daved van Stralen: With that framework, Mark, can you comment a little bit about health care?

Mark Chassin: I think in the framework that I’ve outlined, that if leadership is not engaged, then you haven’t really started. So I’m going to bypass that one, because there are all kind of reasons why leadership has not yet in health care embraced this as a vision or as a goal. I think when many health care leaders are exposed to these ideas, and I’d say based on my travels around the country, indeed, around the world, it’s probably in the single digits when I ask audiences of hospital leaders, whether they’re physicians, nurses or managers, how many of you have heard about high reliability? It’s probably 5, 6, 7 percent, something like that. But even the ones that have heard about it, and when you ask, “Well, why haven’t you really committed to this? Isn’t this what everybody wants?” It’s obviously what everybody wants. They believe it’s absolutely unattainable and is so unrealistic that in the real world and in their lifetimes, it’s not worth even starting down the road, so there’s a big set of obstacles around leadership. I think, though, beyond that, the problem that healthcare faces is that there are so many broken processes and so many uncommon adverse events, where do you start? Priority setting is very difficult, a very difficult part of this problem. A second component to that aspect of the problem is that most, if not all health care organizations, have devoted very few resources to improvement, partly because leadership hasn’t made quality and safety a high priority. But the fact is there are very few resources and those resources are stretched across many different activities, only some of

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 5

which are related to quality and safety, so the priority setting for them becomes even more difficult. I will give you a couple of examples. Government has been only partly helpful in this regard, so the quality measurement in hospitals, the program that I mentioned to you that has a lot of very good measures in it also has some pretty bad measures in it, because the good measures require very painstaking collection of clinical information from clinical records in order to create valid clinical quality targets to shoot at. Not every heart attack patient should get a beta-blocker; who should, who shouldn’t — that’s part of the measure. So over the years hospitals have complained about all the resources that are being siphoned from other activities being devoted to this quality-measurement activity, so government said — Medicare in this instance said, “Okay. I understand your burden. We will calculate quality measures from the data that you give us on your hospital bills.” And now data on quality that you see on the Medicare website often come from hospital bills, which fail lots of tests of validity in quality measures. So the same folks that are trying to improve on really important clinical processes to prevent infection following surgery, to deliver life-saving medication, et cetera, are also working to understand why their billing data suggests that they have a high rate of pressure ulcers or other problems when they don’t. So we don’t have alignment around all the forces that are aiming in the right direction in safety and quality. And then lastly, I would say we have imperfect penetration of these very effective process improvement tools. So even when an organization is devoted to fixing a real quality problem with the right resources, they very often only get part of the way there. And when that happens, and especially if they’ve gotten a little bit of improvement and then six months later it slides back to where it was before, that is a very discouraging event for the organization and for those who are really struggling in this uphill battle on quality. All of these problems from the lack of leadership commitment to the many different priorities that organizations have to fight against — scarce resources, imperfect process improvement tools, and I’ll add one more in which is one that we’re actually struggling with right now, and that is the way in which we teach organizations, hospitals and other organizations in healthcare to understand what their vulnerabilities are, what their system weaknesses are by doing an adverse-event investigation or an investigation on a particularly problematic close call, and to understand what that investigation teaches you. Even if you do that well, and, as I said, we have probably the world’s leading expert group in helping organizations do that, and that average is not very good, but even when you do it well, what you find is dozens of system weaknesses from looking at one adverse event. Now, part of our problem is that we teach organizations to treat these events as isolated instances and come up with a corrective action plan and, do it, wrap that up in a bow and put it in a closet until you get the next one, when in fact we don’t have good tools in health care — I’d welcome suggestions from elsewhere, I haven’t seen any — that allow you to compile across events what the information is about your system weaknesses and the weaknesses in your defenses, because the critical question is, if I’ve got dozens of system weaknesses, how do I know which one I need to fix first? Which one is likely to be lethal tomorrow?

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 6

Now, I can do FMEA and imagine it, but in health care we don’t have that problem because we have events every hour, and we can compile this information. We don’t have to guess at what the most problematic system weakness is, but how do I do that in a way that actually gives me good data about where to focus these scarce improvement resources? So those are a few of our obstacles. Daved van Stralen: Now, given one extreme of health care struggling with this, I have asked Chris Hart to comment, in working with them over 15 years, how the FAA has gone from error reporting to the whole commercial air industry, but I’m looking for successful stories. Chris Hart: Thank you, Daved. This is a very interesting conference because we’re talking a lot about problems, and I think we’ve heard a lot about that. I think I’m going to try to take a stab at a few solutions, and that’s always the tough part. And as I said yesterday, I’m going to start by saying the solutions that I’ve seen in our industry that have worked don’t necessarily fit all, and I understand that. But they’re nonetheless food for thought hopefully for people to take back home and say, “I wonder if the things that worked for them would work for me? And I wonder if the things that didn’t work would provide a lesson for me to avoid them so I don’t have to do them.” So let’s look at some of the key indicators, some of the key areas that have to be addressed, and you’ve heard of them. Mark just covered many of them very well. First and foremost, if the system that you’re developing to improve safety doesn’t also improve productivity, then fundamentally you’ve got a huge problem. And, you know, safety people tend to have a halo and they — I’ve had, I’ve known safety people that even got insulted when you started to talking to them about costs, because they figure, “Hey, this is safety I’m talking about. Don’t bother me with these earthly things like cost.” Well, obviously in the real world, it ain’t going anywhere unless it helps the bottom line eventually. If it hinders the bottom line, then it’s headed for a problem. When I say that, I can’t say for sure. I don’t have enough experience to know whether it’s easier to start a new program than to restart a program in a place where it started and failed. Daved Van Stralen can tell you an interesting story about that, and that’s going to go to some of the issues that I’m talking about, is sustaining it versus starting it, but one way or the other, it has to be shown to improve productivity. The only solution that I can come up with on that is to gather some success stories, and we heard some amazing success stories from Pantex a couple of days ago. I think maybe an organization like this organization, like Daved’s organization, could start gathering success stories, and the more diverse they are — because we can talk until we’re blue in the face about this, but unless we can show some real successes that will appeal to the CFOs of the world, then we’re going to have a huge problem with trying to push this rock uphill. So that’s improving productivity. Mark mentioned about you need dynamic leadership, and that’s crucial. That leader has to be one that engages in system think that I talked about yesterday, that collaborates, that engages the employees; that doesn’t have a first response of punishment to every error

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 7

but instead looks at the system issues in addition to the employee issues; that doesn’t make decisions unilaterally and foist them on the employees but engages the employees in developing solutions; that doesn’t do remedies from instinct but does them from actual evidence and gathering of information; that implements and then evaluates the implementation; that has placed a safety metric on middle management, as I talked about yesterday, so that middle management isn’t so narrow-focused into a productivity method, metric that middle management is not interesting in hearing from the floor about potential problems in the system, all these things that a dynamic leader needs to do. And part of that is also a concept that Mark just mentioned that’s huge is trust, because if the employees don’t trust management, then, you know, they’re not going to report anything, and we know that the fuel for this improvement process has to be the information from the frontline about what’s happening every day. A corollary to the dynamic leadership is one that I know about personally as an attorney, and that is, the lawyers have to be on board with it, because if the lawyers aren’t on board with it, then it’s not moving either. And we’ve seen some lawyers who are more progressive, some lawyers who are less progressive, just like we’ve see leaders who are more progressive and less progressive. Again, there I would go to like the American Bar Association and sort of draw from their resources to talk to some of their lawyers who have shown how beneficial this can be because, as I mentioned yesterday, in terms of how much participation do we see in this process by the manufacturers of the equipment that people use, as we have seen so successfully in aviation, the way Airbus and Boeing participated and brought in the end users and made better products. And, you know, I’m sure they had to fight with their lawyers too, because the very fact of doing that constitutes an admission that you have a problem that needs to be fixed, but, you know, many of the lawyers, with the big picture will realize in the long run it’s better to fix the problem than it is to try to ignore it and have it cause liabilities later. So a close corollary to dynamic leadership is obviously having the lawyers on board. And then last but not least, as I mentioned yesterday, the source of the information about what’s not happening well in the — ultimately has to come from the people on the frontline, the frontline worker bees. And that’s why I mentioned about seeing huge differences between industries where the frontline workers can be hurt, like aviation, like chemical manufacturing, nuclear power, and industries where the frontline employees aren’t hurt by their mistakes like healthcare, like financial services. And that has a huge difference in the incentive for the frontline workers to participate in these programs, because that just makes a major difference in their willingness to admit that they made a mistake. In some cultures, if you admit that you made a mistake, that’s like admitting that you’re somehow subhuman. Well, you know, that’s a cultural thing. The pilots, it took — even though the pilots, as I said yesterday, are the first to arrive at the scene of the crash, it wasn’t easy for them to take on this role of admitting their mistakes. So even in that case, this was not an easy transition for them to take, to go from hiding them and covering them to admitting their mistakes. So if the employees aren’t on board — and that involves, it’s not just an employee decision. That involves, again, senior leadership making it clear what the protections are going to be for the information they provide. It involves the regulator making it clear what the protection is for the information they provide. And, again, it comes

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 8

back to one of Mark’s basic points, which is trust. They have to have the trust to be able to provide this information, know that it’s going somewhere and that it’s going to do something useful. Now, on that one, I would suggest, aviation — the unions in aviation, especially the pilots union, the Airline Pilots Association, has wonderful success stories on how well this has worked now that they are taking a leadership role in making this happen. We’re trying to do the same thing with the air traffic controllers union. That may take a little while under the circumstances. But the point is the progress of this whole effort could benefit from labor unions talking to other labor unions about the success stories that they have encountered themselves to help, again, make it not so it’s just, you know, a lot of people, a lot of good people like van Stralen and Mercer talking about what a great idea this is, and The Joint Commission and the IHI talking about what a great idea it is, but it’s actually that we’ve got a real, a real experience base to go on to show people that, yes, it really works. Yes, it really does help productivity. Yes, it helps employee morale. Yes, it actually changes the labor/management dynamic from the all too common — something goes wrong. Labor says, “Management is horrible.” Management says, “Labor is lazy.” And this sort of harping at each other is so much like your end of the ship is sinking, and it’s never been clear to me why people don’t realize that, you know, they’re in this together. This is an unmeasurable benefit. I’m sure it’s a benefit, I just haven’t been able to measure it. The labor/management relations, to change them from adversarial to partners in improvement has to have an enormous, an enormous benefit on productivity, but I’ve never been able to measure that. But I think the starting point would be, get some of the labor unions into this discussion who have had amazing successes and have them reveal that so it’s not just good people talking about, “This is a great idea,” but it’s actually the empirical results of having done it. Thank you very much.

Daved van Stralen: I like the point about working this together. I was at a group a few months ago, Chris and Tom Mercer and a few others, and one of them pointed out, the comment is, “Your side of the ship is the one that’s sinking.” But the lawyer part is important. We work with a federal attorney from the Department of Justice, and we’ve talked to it before, even in the DOJ, about high reliability within their system and the problems that occur because they don’t have it. But it was interesting that during trial, as much as I like Randy Gido’s [phonetic] point about error identification and correction, the attorney pointed out you cannot do that in a trial, because once you make an error in your testimony, it sticks. So that was one area that I learned you cannot identify and correct an error. But what I want to do with Tom Mercer now, is that — in reading about high reliability over the years and identifying that that’s what I have been doing for a while, it’s easy to forget that, where did this come from? And in a phone conversation once, Jim Holbrook pointed

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 9

out that really what high reliability is is Tom Mercer’s command philosophy. Now, we don’t like the word “command” because of a number of reasons, and I don’t want to go into them, but we all know, it sounds autocratic. But what Tom pointed out to us once was command was that part that you cannot delegate legally, and when you start looking at it that way, it changes the way you see these things. And the other part, it’s a ship, and I don’t want to give his lecture, is that nobody can say no except the captain. You have to keep pushing it up until the captain is there. But the point being, that high reliability can be a natural philosophy. It came from a natural command philosophy, and Tom Mercer is doing this. It’s his normal way to work, thinking, well, there’s a better way to do it. So I’m going to have Tom now comment about that. Tom Mercer: I guess my experience with medicine is, again, is the mayor of the ship, and really the XO is the mayor that’s working all the heads and beds and administrative things and supply things and food and all that kind of stuff on a daily basis, but the captain or the leader or the manager is still the man that’s accountable and responsible for those, everything that goes on with all the people beneath him working it. So there have been carrier captains that I know that found it necessary even to — or found it satisfying to even go into the operating rooms and scrub down and observe. Hopefully, they stay out of the way enough, and I never found it necessary to see all that blood. Maybe I couldn’t stand it, I don’t know. But again, again as part of your overall responsibilities, even visiting the sick bay almost daily is one of those type of things that helps. The word gets out all around the organization that you are a leader that cares, that makes the time to do some of those things that, that clearly come out of a very busy schedule but are very necessary so that people know that you’re all a team pulling, pulling together. My personal experience with medicine, and it’s just been in the last couple years, I did have a hip replacement about a year ago, and this was in a Navy setting. At the same time, I’m providing the transportation for my mom who is 96, who fortunately doesn’t have any major medical things. But all these patient identification type things, and certainly time and time again, initialing off on what you’re there for. Initialing off on, yes, it is the left hip not the right hip. The marking of the position before you go under, the, you know, identifying all those type of things. The system seemed to be working that would prevent these type of things, yet we continue to hear the stories from Dr. Chassin and others that there are still a considerable number of mistakes made from time to time. And I wonder, again, being sane and cognizant of what’s going on, a lot of the times it just seems to be just a very routine going through the checklist. You’re required to do it, so they do, but I’ve often wondered whether if you said the wrong thing, whether they would catch it or they’d just slip right over it. I’m not going to test them, but — and particularly as people develop some amount of confusion or dementia and that type of thing. Just, again, for the information of the people in the room here, we, I think are well-represented by Navy medicine. And I want you all to recognize that Navy medicine provides all the medical care for the Marine Corps also, so all your corpsmen in advanced Marine operations are all Navy personnel, wearing Marine uniforms very often, and then in

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 10

many cases, and we’re getting into another high-reliability-type study where we’re getting involved a little bit in the training of Army medics and Navy corpsmen, particularly in an emergency situation and how much simulation might be advantageous and applicable in those type of settings. But, you know, some of these people may be serving only one tour with the Marines, others spend their entire career with the Marines. Another just connection with medicine and sort of a relationship with the type of — type of things the leader or the high reliability manager needs to look out to, as part of the evacuation from Subic Bay during the eruption of Mount Pinatubo, we had 20,000 people, 13,000 Air Force, about 7,000 Navy, and we attempted to get all the late-term pregnancies on the same ship, not the carrier. The Marine LHA or the Navy LHA carrying Marines where they have seven operating rooms was, of course, a much more capable medical ship. So all the people from Jungle General, which is what they called themselves, the people at the Naval hospital there at Subic, were all supposed to be on the LHA proceeding the day and a half down to Subic Bay, which was the evacuation base before they were picked up by air. The carrier, of course, Lincoln, the Abraham Lincoln came also and did two different pumps of people, and in spite of the fact of all the screening and really attempting to move people around, there was one baby born on the Abraham Lincoln, now known as Abraham, I guess, for the rest of his life. I’d like to see a show of hands on the people that have had an opportunity to read the article that we referred to. Has anybody here read Dr. Chassin’s recent article that he referred to? Good. Well, we’ll be looking to you particularly to ask any questions and lead any discussion on that. We are well-represented here in the audience by leaders in Navy medicine, so we will expect some questions from them also. With that, I think we’re ready to turn it over to either another comment to kick it off from the panel or any questions there might be. Thank you. Daved van Stralen: One comment I want to put out, Subic Bay was being turned over to the Philippine government at the time, and when Mount Pinatubo erupted — and we don’t hear much about that, but we also don’t realize that a hurricane came in the day it erupted. And part of it — we talk about the black swan, the emergency, the crisis, what happens when you have a hurricane during a volcanic eruption and you have to change your evacuation plan? And the reason we don’t hear so much about it, again, is the command philosophy. Question: I’m from Baltimore. I’ve been in my present job for about nine months. I’m a health care CEO of a small system. I’m appreciating Dr. Chassin’s three prerequisites. I have drunk the Kool-Aid. I am working with my board so that my board understands it. We’re working on our fully imbedded safety culture, and I believe our biggest single problem is what you’re calling RPI, robust process improvement. We are about a $480 million organization. I was trying to calculate in my mind, I think we’re presently spending about a third of a percent on improvement support. And I was appreciative of Dr. Weick’s points on the article by Busby, which I have not read, but I’ll go and read it. And the point about

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 11

trying to change while the present systems are still operational and the notion of the emerging events and trying to decouple improvement and keep the business going, I just wanted to reflect that I think that that is our single biggest problem in health care, that if the leadership is committed, as Dr. Chassin said, where do you start? None of our processes were originally designed. They all sort of fell into place. So when you ask a manager, you educate the manager and you get her going on high reliability organizing, she is frequently overwhelmed on the frontline, and the only hope for health care, an organization that’s absolutely committed, is to have enough support to do the robust process improvement. And the quickest way for a CEO in healthcare to be fired is to get in the red. And one of our problems, we’re eating huge amounts of resource. I’m not saying we need more resource, but we need a little bit more time, and I think we could use some help. I’m sorry, I am asking for help. We could use some help from outside organizations on this notion of help us get — refocus our resources. Chris has said that it isn’t a dichotomy of productivity versus safety. I get that, but the incremental changes are not robust enough to really feed the need for resources on the improvement side. Mark Chassin: Thanks for that question. I was looking for an opportunity to reinforce what Chris said about safety and productivity. In health care, the good news is that the business processes are just as broken as the safety processes. And I’ve often encountered the attitude on the part of leaders when we talk about robust process improvement and these strange things like Lean and Six Sigma and change management, which I do want to comment on briefly. The comments are, “Oh, okay. That’s the flavor of the month. It was CQI and TQM and Quality Circles and Toyota. Now it’s this, and if I wait long enough, that will go away and then you’ll come with something else,” and that’s one attitude. The other attitude, it’s a little more realistic, is, “I don’t know anything about this stuff. Neither does any of my staff. It will cost me a huge amount to bring this new technology, if you will, into my organization. My bottom line is tenuous. I can’t afford the diversion of my staff to this new activity or the cost, consultants and everything else.” Well, the answer to that question, which I think is almost the one you’re asking is that because of our fortuitous problem with all of our processes, these tools which, where they have been applied, have most often been applied on the business operational side of healthcare are incredibly powerful in generating revenue. And we are also fortunate. We complain a lot about the financial incentives in health care not being aligned with quality incentives, and that’s very true, but there’s opportunity there, because there are places where you can improve quality and save money at the same time. So start with those. There are business processes that I guarantee you are not getting you all the revenue that you earn, and there are costs that you’re incurring because of wasted steps in processes that you can save money on and have a beneficial effect at the same time. So for example, routine value stream mapping and other kinds of Lean tools show that about 25 percent of nurses’ time — time is completely wasted in efforts like getting all the supplies to do a dressing change or administering medication, because the processes that they access are so poorly designed.

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 12

So the flip side of that is, as a CEO, if you could increase the number of nurse hours at the bedside by 25 percent and not hire another single other nurse or pay another dollar of overtime, would you do it? Of course you would. So it is possible — it is very feasible — not possible, it’s very feasible to design a program of adopting RPI tools that will give immediate return on that investment within 12 to 18 months by focusing on business processes that aren’t working and then by learning how to do these quality improvements that we are all going to be accountable for sooner or later, and I think that’s the answer to the question of, “This is too costly, it’s too time consuming.” On the priority-setting side, much more difficult to know exactly where to focus your efforts. And that, to some extent, is going to inevitably be a trial and error process. Part of the learning of applying these tools is that they have to be applied within a strategic framework, and you’ve got to understand what your business is, exactly what is driving it and what are the critical quality processes that support those drivers. And that’s how you learn to apply the tools of RPI. Find the biggest opportunities, capitalize on them right away, and that begins to build trust in the application of the tools and enthusiasm for them. There’s a lot more to be said about that, but from a strategic standpoint, you can do both in health care and have the productivity and safety improvements at the same time.

Chris Hart: Following on what you just heard, the ultimate sales pitch to the workforce about why are we doing this, if it’s pitched to them as this is something to improve safety, that’s going to be kind of flat. If it’s pitched to them as a way to make you more capable of doing your work better, more efficiently, more effectively, and it’s going to improve safety at the same time — that’s why it’s so important to engage them in the development of the process so it’s not, “Management says this is the best process for you,” but instead, “You have helped develop the process, you’re engaged in it.” Their engagement in it will guarantee that what comes out of it is going to be more efficient, more effective for them. So that’s — the sales pitch to them is not, “This is a safety program.” The sales pitch to them is, “This is a program to help you do your job that you’re trying to do right better, more efficiently, more effectively.” Kevin Bowman: Just to build on the impact of productivity on doing these improvement efforts, one of the things that we’ve done within Dow, we’ve done a survey of roughly 1,500 of our employees and just asked them to codify their day, how much time did they spend on unplanned events, whether at home or on the job, and after doing 1,500 employees, we’ve uncovered that roughly, on average, a Dow employee spends 42 days of their year working on unplanned events. And so if we can eliminate those unplanned events, that’s 42 days out of every person’s year that they could start focusing on, you know, things they planned to do when they show up to work rather than reacting to unplanned events. And those kind of numbers are really ringing true with our leadership, because that translates immediately to, “Okay, this is the request I got to add somebody to my staff. I can pick that person up pretty quickly if I start focusing on eliminating unplanned events.”

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 13

We actually introduced it through the maintenance function, and it was tied to an IT project, recognizing that if we didn’t change the culture on how we viewed executing maintenance and reliability, then we were just spinning our wheels replacing the IT tools. And now it’s grown into a partnership with our process safety folks. It’s grown into becoming a manufacturing initiative, and now it’s in the process of becoming an enterprise-wide initiative. Daved van Stralen: And we look at the melodrama of the ICU and the ED and you run into that trap of World War II of, “Don’t you know there’s a war on? Don’t question this.” And all of a sudden, your efficiency drops way below what it should. So looking at it through dietetics, we started seeing windows of where they could become efficient, and it was similar to the maintenance portion. Question: I’m a physician and vice president of quality for a large system in the southeast, a health care system, and so we’re involved in implementing Lean and Six Sigma within the organization, working extensively on the cultural aspects of it. My challenge, and you’ve already kind of addressed this, but my challenge is how do I engage the leadership, both administrative and the academic chairs? And really, my question is what — in the organizations that are successful in this, the health care organizations and others, what were the factors that lead the senior leadership to take this on and how do I engage them? Mark Chassin: That’s a really good question. Trying to do this, this organizational transformation from the middle in is very difficult, and I think you’re right to focus on the engagement of leaders. An organization has to have at least one senior leader that is an indefatigable champion for this transformation, because for my money, if an organization doesn’t approach this as an opportunity to transform itself, you will get part of the way there, but you will never get to high reliability. So it has to be a complete emersion in the devotion to getting to the goal of high reliability, which has the features that we’ve talked about, cultural transformation, as well as measurement and effective process improvement. I think there are a couple of strategies. As Chris has said and our other questioner, John Chusar’s question brought out, you’ve got to have the CFO on board, because in an organization that hasn’t quite awoken to how critically important strategically safety and quality is, they have to know that it’s going to help the bottom line. And choosing careful targets that will demonstrate that these tools that are just as good at improving safety will also affect the bottom line, gets the CFO’s attention. At some point, somebody has got to get the board involved and engaged. If the board, for example, doesn’t get regular reports on adverse events, that’s a big problem. If they don’t understand the need to fix all the problems that the adverse event investigation uncovers, and if they don’t see how those problems are addressed and the measures that are put in place and the monitors that are put in place to make sure that those problems stay fixed, that’s a problem, because that’s a source of pressure on the management, both physician and other managers.

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 14

But there really isn’t much of a substitute for at least one senior leader, whether it’s an academic chair or the chief medical officer or the CEO, starting to frame the question as a burning platform. From a change management standpoint, if there’s no shared need to change, it’s almost impossible to get the rest of the way, and building that case of the shared need shouldn’t be that difficult today in health care. Once that’s accomplished, the next step is what does the vision look like? What’s the future state we’re trying to establish? And getting everybody engaged in — all stakeholders engaged in defining the need and the vision are the first two stages of managing that change, and they all need to be managed at the same time. And we can — that’s what I meant by the third component of RPI is a formal approach to change management. My own favorite is the one that GE developed, the change acceleration process, which came out of work that Jack Welch initiated, which really was the University of Michigan Business School, that asked the question: Why does innovation fail? And what they discovered was very simple, that if you looked at efforts at innovation and change that succeeded — this is business — compared to ones that failed, in 90 percent of both cases, dozens of cases, in 90 percent of the successes and the failures, there was a really good solution to the problem that they were trying to solve. What explained the difference is the failure to manage the organization’s acceptance and implementation of the change. So a formal approach to change management that is just as systematic, measurable, teachable, as the problem-solving part of RPI, is critical to getting this work done in health care and I’m assuming in other organizations.

Question: Good morning. As one of the Navy medicine representatives, I’m the commanding officer of Naval Hospital, Jacksonville. I think Navy medicine has some real wins that we can put out there. One is what the admiral talked about, we go forward. We are deployed. My command is deployed — 10 to 25 percent of my command deployed forward at any one given time. When Haiti went down, we had guys on the Comfort within six hours of the disaster, and we had guys in country within 12 hours of the disaster. So medicine and the military are quite alike with, we are rapid, flexible, agile. We move to where the threat is, and that’s where I see a lot of where this stuff comes from that we’re talking about today, as a threat. And when you talk about the fixes, the different things that we talk about, hand washing, you know, ventilator-assisted or -acquired pneumonia, things like that, to me when I look at that as a CO, it looks to me like we’re putting Band-Aids on things. And the way we measure our success, HEDIS measures. You know, that’s six or seven items out of the thousands of things in medicine that we have to detail and that we get graded on, but yet it’s a Band-Aid here and a Band-Aid here and a Band-Aid here. And as long as our organization focuses on Band-Aids, we’ll never get to that 100 percent reliability and safety which I firmly believe we can. So in our organization, we have changed, and we have moved to a process thing. Our system is so complex, it’s got to be a systems approach to fix it. So I’ve actually got a

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 15

special advisor for process transformation now that I appointed, whose whole purpose in life is to take all of the different things we do, coordinate them, make them into a systems approach so that they’re aligned with our strategic plan, they’re aligned with safety, patient-centered, safety, quality care. And when I get to safety and the quality fix, the cost will fall in line with that. I’m 100 percent convinced of that. The question that I have to you gentlemen is the same question I posed to the panel yesterday. I’ve got all that stuff working and moving forward, the problem I have — and we just passed our Joint Commission inspection, by the way, smoked it, as a matter of fact, so I’m very proud of that sustainability. I’ve got all of these things going on, but we all know that without the human element of eternal vigilance, of constantly relooking at programs that we’ve already thought we’ve institutionalized, if I don’t put the effort into maintaining those, the human effort into maintaining those, they will fall off. And all that human effort I put into maintaining those, now I’m taking away from my ability to grow and improve in other areas, because I’m maintaining those programs. So the question is how do you maintain and how do you grow? How do you institutionalize those so that you have sustainability? I consider this so important, I’ve appointed my executive officer as my chief sustainability officer to try and raise it to the level that it’s that important that we do it. Tom Mercer: One of the questions I had for Dr. Chassin was, in educating about the close calls in a hospital or in a medical setting, who really does that? Is your appointed patient safety officer, appointed patient safety director, is it truly a well-respected physician that does that? I know I described that I’ve been involved in both Navy medicine and civilian medicine in the last year here, and they are — even the placards on the wall, “There are our patient safety goals.” It’s for everybody to read and everybody initials off on it and that type of thing, but I’m not sure who in those offices are actually managing those programs and really have the ability to stay on top of the numerous different initiatives. Mark Chassin: On the question of sustainability, I think that is one of the biggest challenges once an organization gets rolling toward this goal. We have enormous difficulty in healthcare with sustaining improvement, and I think there are a couple of reasons for that. One is that we have tended to approach serious quality problems as special projects. We put our best people on the project. We throw the kitchen sink at it. We get a little bit better, and then the team has to move on to the next little brush fire, and you can’t sustain kitchen-sink approaches over time. The idea that we have pursued with our Center projects and that these RPI tools require you to do is to really find the most important root causes of process failure. So it’s not — every organization will not have to work on all 25 root causes of hand hygiene failure. Most of then, most organizations will have to work on three or four, and that’s a lot easier to sustain than trying to measure and keep track of 25. Now, how you actually do that is really important, and these tools are really very good at planning from the very beginning of a project that’s attempting to tackle a big problem, how do you at the end of the day turn the improved process over to the individuals who are in

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 16

charge of it in the ordinary course of the organization’s work? The project improvement team that helps find the right solution can’t keep operating the process forever, so that in RPI or in Lean Six Sigma terms is the control plan. You know, how do you turn it back over to the process owner who is going to have to run this new improved process? Well, you give that individual a new instruction manual and a new gauge that says, “Here are the three measures that you will be confronted with every month to know that this process is continuing to work at the high level that it works at now. You participated in this process, so you believe it from the standpoint of change management, and here’s the plan. If it starts to deviate in this direction, you do this. If it starts to deviate in that direction, you do that.” So that’s the underlying philosophy. The additional dimension to this improvement effort is that we have to think about technology in a different way, particularly information technology. We have, and here, again, government has been helpful and giveth with one hand and taketh away with the other hand. We’ve had enormous incentives with the Recovery Act, financial incentives to put electronic health records into place in hospitals and doctors’ offices, but we didn’t, at the same time or before we did that, fix the processes that we were automating. And the best aphorism that I ever heard was from an old grizzled chief information officer who said to me, “You know, computers don’t make us less stupid. They make us stupid faster.” The worst thing in the world to do is to automate a broken process in health care. You get dumb things to patients really quickly. So we have to think about technology, and particularly information technology, from the standpoint of after we fix that process and we’ve got it working really well, which parts of it do I really need automation for in order to help me keeping it working at that high level of performance? That’s not typically the way we’ve thought about information technology and other kinds of technology. One of the most important parts of the hand hygiene story to automate is the measurement. It’s very difficult to measure hand hygiene compliance. In fact, some organizations have actually taken to measuring it by measuring the amount of alcohol rub they consume as an organization. I’m serious. So this, our teams recognize that their method of measuring, which is very accurate and very reliable but relies on trained observers is not sustainable or scalable. So we asked a bunch of companies to come up with solutions to that, and now several of the organizations that plateaued at 80 percent, a real 80 percent, are using technology that combines very simple RFID tags on identity badges with proximity sensors on alcohol-rub dispensers tied into a databases that produces the real-time data on performance that units can use to see whether they’re deviating or not from a high level of performance. So technology, intelligently applied, is an incredibly powerful sustaining, help to sustaining. On the question that Tom asked, I think, unfortunately, the answer is there is no systematic or consensus way in which organizations keep track. The title chief patient safety officer, chief quality officer, chief medical officer, vice president for medical affairs, once you’ve seen one, you’ve seen one, and there is no uniform way in which organizations track these things, and that’s a part of the problem. But it all, again, I think goes back to leadership in

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 17

the organization. If you see a disorganized approach to managing quality, then you know you have a leadership that’s not committed to high reliability.

Daved van Stralen: I want to make one comment quickly. I had used high reliability in one of the sections of the hospital. We had a critical mass of physicians who wanted central authority, the doctors in charge, and everybody listens to what they say and does it, and the two of us who ran it in that model left. About five years later, I was giving a lecture to the medical students, and one of them had been a respiratory therapist working night shifts to get her way through medical school, and she came up to me after my lecture and said, “Are you the one?” And I said, “No, I did not father your child.” I said, “I’m the one what?” And she said, “Everything you said in the lecture is what I learned at night.” And apparently at night shift, the respiratory therapists taught everything that we had done. And they said, “Now, the doctors don’t like us doing it this way, but this is how we do it. Just don’t tell them.” And she was indoctrinated, in effect, overnight on how to work in a critical care unit, how to manage the ventilators, how to evaluate a patient, how to think, make decisions, all of this stuff, and it had never been put with the words. Somehow we lost some of the words to it, and when she heard the words I gave, the concepts I gave, it all came together. She realized there was a structure to it, there was a form to it. So it was sustainable, but it was all underground. And that’s really similar to what the Catalans did in Spain. They used the Girl Scout and Boy Scouts camps in summer to teach their language and culture under Franco, and then when Franco left, the language and culture was just as vibrant as before, but they took it underground. So in some of the areas that I’ve seen it, the higher reliability actually goes underground and is still there. Question: This is a comment, and I don’t mean this to sound disrespectful to the panel, if it does. On Karl’s continuum between theory and operations, I’m on the operations side, okay? But I was putting the shoes on of somebody new to HRO listening to this very good panel, and I would almost think that many people in this audience could be bewildered, frightened, and, “I may never poke my head out the door into the HRO arena,” just because there are so many things to think about, leadership, trend analysis, graphs, and all of that. And my comment is a number of years ago I was with Karl in Phoenix and Kathleen Sutcliffe, and I was freaking out like this. “Oh, my God, I can’t figure it out. I don’t understand the book. What the hell is he talking about, requisite variety?” Dah dah dah dah, you know. And I asked Karl, and I don’t think I’m misquoting you, Karl, at the symposium in Phoenix, I said, “Is there any way I can get it wrong?” And Karl’s answer was, “No, I don’t think so.” And I want to tell you, it was absolutely liberating to hear that. That didn’t mean I wouldn’t try to go to the theory side where, of course, Karl comes from, to try to figure it out, but to be told by the theory guy that I could just go do it Tom Peter style, you know what I mean, and I might not screw up was just tremendously liberating.

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 18

[2:01:56] And then the last plug I want to put, he then taught me something, Karl, about small wins. These tiny little things you just started to do, and if you fail at them, you know, you didn’t get kicked out of Dodge. And I have found that to be enormously successful. Just try something very small, a small win. If it works, great. If it doesn’t, dust yourself off, and as Ivan Popoletti said yesterday, the system just starts to magically self-organize. Karl Weick: Just to make a quick comment on what he is talking about. The paper that I just summarized, in a way you can think of it as obstacles and limitations, but I would also suggest that you could say to people, “Look, in the face of these kinds of organizational realities, you’re really doing better than you think.” All of those four things — or eight things that I mentioned are normal stuff you run into in organizations, and so just as Dave suggested, doing something means that in fact you’re able to get past or live with or adapt or use these kinds of barriers that are there. I’m really mindful of Mark Chassin’s concern of sliding back and then people being discouraged and then saying, “Forget it.” I’m concerned about the person at the frontline who John said is overloaded, because what I’m worried about for her is whether she’s dealing with that by queuing up issues, or whether she’s just omitting some of the issues, or whether she’s lowering the threshold of success on all of them. Those are very different ways of coping with that kind of overload. It’s a normal organizational kind of overload, but some of those are going to undermine her reliability seeking more than others. So I appreciate the reminder of earlier times.

Question: I’m from Pantex. In case you guys don’t know what Pantex is, we make the biggest firecrackers in the world. But I guess I want to just put out food for thought here on three topic areas. I heard the word “leadership” discussed. I heard a topic there about systems today or the last couple days, and also this concern about process improvement. What I would say is, on the leadership part, we had a challenge at our place. You would think at a nuclear weapons facility that people would know what’s important, and so we went around and asked all of our senior managers, and many, many senior managers — and people are wondering why your people are confused for. Well, go ask the senior managers what’s the most important thing, and when you get 10,000 different answers back, you probably realize pretty quickly why the people on the shop floor can’t figure out which end is up. So we went though a pretty extensive effort asking our senior managers what’s the most important thing, or maybe what’s the most important thing you want to avoid in terms of high reliability. It took us about a year and a half, two years to come up with a consensus, these are the things we want you to focus on. So a point in the matter would be leadership is important, so leadership needs to focus the organization to help do that. The other thing was I heard a discussion on systems. Everybody here probably here takes a systems approach to their problems, whatever it is. I defy all of you guys, get on a chalkboard and write down what your system looks like. It’s physically impossible. And I

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 19

guess the point is, if you can’t tell me what your system is, how do you know it’s working, how do you know where inefficiencies are, and the whole nine yards? So I guess we’re going through what Karl calls this complex confusion state right now, trying to understand what our systems are, because if you don’t understand what your systems are, you sure don’t have a systems approach. And the last thing we’re starting to get insight about, I guess I’ll call it that, is this issue about how you prioritize your improvements or how do you optimize your improvements. Comment: I’ll present this gift, but the best work I’ve seen on sustainability comes out of the NHS, and it’s called the Sustainability Model and Guide. You can get a lot of it from the Web. What you really want is this, because it includes all of the tools and all the resources, but it’s not perfect, but it allows you to begin to navigate the complexity of the landscape and ask the questions that you didn’t know to ask as you’re thinking about sustainability. So I recommended NHS Sustainability Model and Guide, and I’ll present that as a token of my appreciation for your work.

Dr. Mark R. Chassin The Joint Commission Presentation at 4th International HRO Conference: Making HRO Operational, 04/21/11 Page 20