Physician Engagement (Transcript)

September 13, 2011

Operator: Excuse me, everyone, we now have our speakers in conference. Please be aware that each of your lines is in a listen-only mode. At the conclusion of the presentation, we will open the floor for questions. At that time, instructions will be given if you would like to ask a question. I would now like to turn the conference over to Peter Pronovost. Sir, you may begin.

Peter Pronovost: Hello, and thank you all for joining us on this really important topic. As you all probably remember, our work in this area has been galvanized by the haunting words from Josie King’s mother, “How do we know she’s any safer?” And, for us, that translates into, “Are our infection rates truly as low or near zero as possible?” And we spent a lot of time talking about the checklist and technical work. What we want to talk about today is some more of that adaptive work, particularly how do you get physicians truly engaged in this work? And what we hope to cover today is for us to have a conversation about what we really mean by physician engagement and what might some strategies be to make sure, both at a management level and a staff level, that we have physician engagement.

Now, we’re learning about this science of physician engagement more and more every day. And for example, we just recently published this summer; if you haven’t seen it, I would encourage you to look at it, a paper in The Milbank Quarterly entitled “Why Michigan Worked.” And it was partnered with us and a couple of anthropologists and sociologists who said, “Okay, there’s a lot of quality improvement programs. Why has this taken off like wildfire when others have seemed to stall?” And there are a lot of reasons for that, but the main reason that they came up to was their conclusion that we framed CLABSI as a social problem that was capable of being solved. And there is some really deep wisdom in that.

You see, when we started this work, when I started this work back at Hopkins, our rates of infections were abysmally high. But the culture was, “Hey, we’re trying hard. Stuff happens. Our patients are sick, not a big deal. Very little accountability and nothing else we can do about it.” The infections were what I would call in the “inevitable” bucket. And the only way this works is if you change that switch in your clinician’s mind to say, “No, no. these infections are preventable, and I’m able to make a difference in them.” And changing that social norm is absolutely key. But it’s a journey, and it’s a journey that really has to begin with this process of getting your clinicians and, particularly today, getting physicians engaged in this work. And I say, why physician, and why not nurse? Not that nurse engagement is any less important at all. In much of the quality improvement work, though, I think the nurses tend to be more engaged, at least they have been living this, and physicians, I think, have, to a large extent, haven’t been at the table. Perhaps they haven’t been invited to the table, but they’re not at the table, and they absolutely need to be there and leading these efforts.

So what do we mean by engagement? Well, quite frankly, it’s active support of the program. We need physicians at the table, participating deeply, and feeling that they’re part of this, that they’re doing this rather than this is being done to them. Now, you can think of this engagement as a journey where we go from frank aversion, “Get of here, I’m not doing this program,” to apathy, “Okay, go ahead and play if you want, but don’t bother me with my work,” to absolutely engaged meaning, “Hey, this is my problem, and I’m going to do something about it.” And the question is how do we move clinicians -- how do we get toward that all of our clinicians are on that far end, where they’re truly engaged. That is, they say, “This is a social problem that I am responsible for.”

Well, there are a number of ways to do this and some mistakes that we certainly made that I would hopefully encourage you to avoid. The first is show them the evidence of your baseline performance, and that might be performance on a process measure, if you audit them. Certainly your performance of your infection rates. You can use the opportunity estimator to put that calculator that converts your baseline rates into number of deaths or dollars spent needlessly, and you can use that to get their attention. Now, if you’re going to do that, having valid data is absolutely key. There is nothing that will turn your clinicians off more than presenting them data that’s noisy, that has a lot of missing data, that doesn’t have standard definitions, so we have to do our homework. But if you present good data, it gets people’s attention. Another way is to show them what others have achieved, really to show that they are no doubt every bit as good as these other hospitals. Indeed, we showed 60 percent of Michigan hospitals within a year, 25 percent went 2 years without an infection. If one in four can go 2 years, certainly your hospital could absolutely do that. And then, finally talk about the idea that they could achieve these results, that these infections are causing needless harm, and it’s within their scope of practice. In terms of phrasing this engagement, so much of this lack of physician engagement is really deeply rooted in our social norms in medicine.

There was a fabulous book by a University of Pennsylvania sociologist called Charles Bosk. He wrote a book called “Forgive and Remember.” And in the book, he describes how he spent six months at an academic medical center following surgical teams and was trying to tease out which mistakes were deemed preventable, forgivable, and which ones were truly unforgivable. And what he found was just remarkable: That bad outcomes were completely forgivable as long as you tried, and you didn’t lie. Like there was this puritanical sense that if you worked hard and you were honest -- in other words, a very individualistic responsibility -- that was all that counted independent of if you have that outcome. So they are typically acts of God or outside of your control, and we didn’t worry about them. And what we’re realizing now, in this quality and patient safety movement, is that we have to start being responsible for our outcomes, and we could absolutely make a difference.

Now, for you to get your physicians engaged in this work, they are going to have to trust you. And I can’t underscore enough how important this concept of trust is in this work. And no doubt trust is a very complex variable, but if you boil it down into really simple concepts, people trust you when you do two things: When you care and when you’re perceived as competent. And absolutely both of those have to happen. So if your physicians see this as you’re putting someone else ahead of you, you’re using this to get a new promotion, anything other than the patient as the “North Star,” you’re dead. This can’t be viewed as a battle of wits or wills. “I’m more powerful than you.” It has to be about this: The patient is the “North Star.” We have to work together to prevent infections. The second piece is you have to be viewed as competent, that the clinicians have to say, “Hey, this program’s wise. It’s based on evidence that you can actually do the things you’re saying you’re going to do. You’re accurately measuring your data.” And so all the early wins in this project are absolutely key to help you get that baseline understanding that my clinicians see me as caring and as competent.

So let’s drill down to this and say, what do we mean by physician engagement at both the management level and the individual physician levels? Now, we wrote a paper on this in JAMA a few months ago entitled, “The Need for Physician Management Infrastructure in Quality.” If you haven’t read it, for you physicians out there it’s a really good summary of the complete lack of physician management infrastructure. And what do we mean by that? Well, the prime difference between providing care for individual patients and focusing on patient safety and quality as a population is that there’s resources and an infrastructure to manage those populations. That means somebody has to try to make protocols and standardized care. Somebody has to measure performance. Someone has to feed it back. Somebody has to try to find barriers of why we’re not complying with the evidence. And that takes time. And health care has a very mature nursing management infrastructure, so every unit has a nurse manager, and they report up to a hospital nurse manager or a department. But the physician management infrastructure is like the Wild West. Most hospitals will have a chief medical officer or chief quality officer who supposedly oversees quality for the whole hospital, but then outside of that, every doc is still left to their own devices. They’re in private practice, for the most part. When they take time to do this means they don’t bill, and we wonder why we haven’t made progress. We absolutely need a physician management infrastructure to help lead these projects because, again, there is overwhelming data amongst professionals that if we don’t have input into these programs and if these programs are perceived as being done to us, rather than something we do, they’re dead on arrival. And if you think about the way the U.S. has approached quality, it’s being dictated to us by regulators. It’s coming down on management. And I’m sitting here as a practicing doc in the ICU now, as a matter of fact. And we push back, because those policies are often viewed as unwise. That they’re not sufficiently nuanced. They’re not evidence based. They don’t have good measures. And when they’re not science, we should push back, but we also should push forward to make sure that we lead these quality and safety efforts.

So what do you need for a physician management infrastructure? Well, I think, ideally, you need a physician champion for this project, that could be the ICU unit director, a chief medical officer, a chief safety officer. If you’re a small hospital, it may not be an ICU manager. It might be a hospitalist. It has to be someone who is trusted and someone who the other physicians really look up to, and they are going to perceive that they’re caring and that they’re competent. Now, there is no way physicians will keep doing this if they don’t get something out of it. So one of the key strategies are either you or that physician go to your hospital and ask for time to support this person’s effort that they’re making. They absolutely need time if they’re going to do it well. How much time? Well, it depends on what role they do. Many hospitals might have 20 percent. Some hospitals, if they have a lot of administrative responsibilities, do more. But that physician absolutely needs to be covered for their time to lead these efforts, to make protocol, to trial things, to feed back data.

And, too often, it doesn’t work. Now, when I speak to hospital administrators, and for good or bad, because I wear many different hats here, as an executive at Hopkins, as a frontline physician, and as a researcher, I hear a lot of the judging. I hear physicians complaining that, “Well, I do all this work, and I’m not supported for my time for it.” And I hear hospital executives say, “Well, we support these clinicians’ time, but they do nothing. We don’t get anything good for our money.” Right? And in some senses, they’re both right, but they’re both completely solvable. How are they solvable? By creating what we call a compact, that is an explicit agreement about what compensation the physicians will get for their time and what they will do for that. And that compact needs to be reviewed at least quarterly, ideally monthly. So what might that compact look like? Well, it might say, “Okay, we will give you ‘X’ amount of money or time, or whatever that financial arrangement is for leading the ICU or this quality project. And in return, you will agree to chair the quality committee or to meet monthly and focus on what our CLABSI rates are or VAP rates or other infections; that we will agree to do CUSP. We will learn from one defect a month. We’ll make sure that we communicate with the medical staff about this project. We’ll make sure that we learn from defects and that we educate other physicians in the science of safety. Whatever those skills are, it needs to be completely explicit and transparent that what the hospital is providing for this physician and what the physician will do in return. Now, the good news is we just recently published in August the cost-benefit analysis of our work in Michigan and, no doubt, these programs save money. So even estimating that, when a doctor and nurse devoted 10 or 20 percent to this effort, what we showed was that the cost per infection prevented was about $3,000, and the average hospital saved about $1 million a year. So there is no doubt that there is a return on this.

And if you were to ask me now nationally, what is the biggest risk of failure for improving quality and safety, it is unit-level infrastructure and particularly physician support to lead these efforts. We’re drowning in bundles. I mean, I blink, and some other organization is asking me to join this program or do that program, and that’s not where the risk is. I don’t need new ideas. What we need is doctors’ and nurses’ time on the units who have the skills and the effort to do these kinds of projects. And I suspect that’s likely your risk of failure, too. Now, I also think that this compact needs to be reviewed regularly, and that physician, if they are being paid for time or getting some financial support, needs to be held accountable that they actually made progress on what they are going to do. That could be progress for reducing infection rates, progress on learning from mistakes, progress on training, progress on all of those. But it needs to be reviewed at a regular time period. Now, every physician, though, can’t be unit level leaders. And yet, many, many more physicians, many private medical staff who aren’t in the hospital all the time, who need to understand about this program, and absolutely, I think, will be supportive if they understand why.

Now, let me tell you a little story about how I learned about understanding why, a story that I call “Avoiding Monsters in the Bathroom.” My 14-year-old son -- it’s amazing to think how he’s growing like a weed and just started his freshman year in high school -- when he was in third grade, he came home from school one day and said, “Hey, Daddy, Daddy, I’m afraid to go into the bathroom at school. There are monsters in there.” And I said, “Yeah, Ethan. Okay, fine.” Next day, he comes home, clearly distressed. “Daddy, I don’t want to go in the bathroom. I’m afraid to go in there at school. There are monsters in there.” So obviously, I was concerned, and I called the teacher and said, “Hey, help me understand what’s going on in the school.” And the teacher said, “Oh, well, we just installed automatic flush toilets, and nobody told the kids. So there are automatic flush toilets, and they’re monsters in the bathroom.” And I heard that, and I thought about how we implement electronic health records, CUSP programs, BSI programs. And if you’re that community physician who hasn’t participated in the discussion, doesn’t know why, may not even know what your infection rates are, there’s no wonder that these things get resisted, right? Because we think they’re monsters in the bathroom. I don’t know why this is being done. I feel it is something being done to me rather than something being done with me. It’s being done over me, rather than through me, and all of a sudden, I get this pushback to say, “Hey, I don’t know about this program.” And we start getting our decoding errors.

And those of you who remember the science of safety, this concept that whenever we communicate, the sender of a message encodes some meaning, some of it nonverbal, some of it vague or ambiguous language, and the receiver has to decode what they intended. So we gave the example of saying, “Hold two feet if the patient doesn’t tolerate them,” and then the nurse has to decode what the heck we mean by not tolerating two feet. Well, these decodings go on amongst medical staff, so the physicians who aren’t part of this hear, “Oh, we’re doing this program where the nurses could stop central line placement.” And the next thing you know, you think it’s “Animal Farm” going on. Indeed, when we started Michigan -- again, an early mistake I made -- we didn’t pay enough attention to managing these messages. When we had a training session, we talked about nurses stopping takeoff and working with physicians, but if the physicians didn’t comply with the checklist, the hospitals needed to empower nurses to question physicians and make sure they do comply. Well, I get more hate mail the next 2 weeks after that about what the heck’s going on in Michigan: The nurses are revolting. They’re overthrowing us. They’re taking away our autonomy. They’re trying to get independent practice. All these decoding errors that had nothing to do with what truth was, but the problem we got into was we hadn’t managed the message. So we had decoding errors.

Now, how do you manage that message? Well, you absolutely need what I call a “containing vessel.” You need some structure to have these conversations in with your medical staff, and that could be a morbidity or mortality conference, it could be grand rounds, it could be a quality improvement meeting, it could be some other medical exec meeting. But your physician leader or your nurse leader needs to have a way to have dialogue with your medical staff to get their input so that we do these wisely and to make sure that they’re actively involved in it because the one thing we know from the change management literature is that participatory forms of improvement, that is forums that involve all your staff, develop wiser solutions. They devise solutions that are more sensitive to your local context, they state your local work environment, and that they’re more likely to be accepted. So if you want solutions that work in your context, taking what’s worked at Johns Hopkins is a recipe for failure. Talking to your clinicians about where we’re going, about how we have to reduce infections, getting their input about what works in your area, what are the barriers for them to this are absolutely key.

Now, if you’re trying to do this, it’s also really, really important that you approach this with the mental model that your clinicians want what’s best for patients. And you know, I’ve been around thousands of hospitals and, fundamentally, I think that’s true. We didn’t go in this business to get rich. Doctors and nurses are in this business because they care, and they care deeply. And nothing offends a physician more than if you start that conversation from the attitude of judging rather than understanding, that “You’re a bad doctor. Infections are high. Go fix it. I’m going to tell you what to do,” which is typically how we’ve had this regulatory approach to health care. Very different approach than if you have a conversation saying, “Well, I know you want what’s best for your patients, and our rates of infections are high, and I’m confident we can do better. Many others have, and we’re just as good as them, but I need your help to help get our rates down. You have wisdom. You know what the barriers are, so could you agree to work with me?” And then, really listen, and find out why it’s hard. Get that ground truth to say, “Well, we’re not using the checklist because we don’t have a cart that stores anything.” Or one hospital told me, “Well, yeah, the hospital bought us a cart, but the wheels are all broken. It doesn’t roll. It’s a pain in the neck. We never bother using it.” Or “We don’t even have the right equipment stocked. I would do it. I want to play, but it’s too hard.”

So seek first to understand, and in your doing this, as I often say, “Value the resistors.” One of the things I found is that many times, physicians express resistance that’s completely legitimate. Now, some of the resistance may not be expressed in the most emotionally intelligent way. You get screaming and yelling, and it’s difficult. But if you could kind of put up with that and see and hear the music beneath those words, what you’ll find is that there is often legitimate concern about barriers that you need to solve as this project’s leader to make it easier to comply with the checklist.

And I will tell you my own journey. When our rates of infections were high and we didn’t have all the equipment that we needed, I would make a conscious, perhaps sometimes unconscious decision, where I would literally go through the economics in my head and say, “Okay, yeah, I don’t have this gown or this mask, but it’s going to take me 10 minutes to go run down the hall and get it. I don’t even know if I’m going to be able to find it, and I don’t have 10 minutes. And if I do that, that’s 10 minutes I’m not spending caring for another patient. These infections are kind of invisible, and they happen in the future, so I’m just going to go without it because the more immediate risk is those 10 minutes I’m not going to spend with another patient.”

It’s leadership failure, really. I shouldn’t ever be put in that position as a physician or as a nurse, but I am daily. And you, as the leaders of this project, have to make sure that you understand your local barriers so your docs don’t need to make these changes. Now, when you go to present to your physicians, how do you talk about this and how do you put up with some resistance that you might have? Well, a couple of ideas for you. The first, and as I mentioned before, assume that they want what’s right for their patient. Don’t go into this judging it. Go into it understanding and listen to them. Ask them why this seems to be difficult for them and what they’re putting up with. Second, absolutely make sure that you tune into WIFM, or what’s in it for me. We’re all human, and we all have needs. And physicians are under time constraints, and whatever they spend on this isn’t time that they’re spending generating their livelihood, so try to understand what they get out of this. And it might be that you can get some recognition from your hospital executives. They can present to the board. If time is a barrier, and I can guarantee you for every physician, time is always a big thing of what’s in it for me, how can you make it easier for them? How can you do something that makes it take less time rather than more time?

You also need, as you get some resistance, to try to understand what the loss is. Now, some of you may have heard me use this discussion before, but we all hear, “Oh, doctors resist this change. Doctors hate change. They don’t want to do it.” And I suspect most of you, if I were to ask you if your doctors resist change, your hands would go up, and you’d say, “Absolutely.” But let’s think about that a little bit deeper. Let’s unpack that concept. If I were to give most of your clinicians the winning lottery for the Maryland State Lottery, and let’s say it’s a $350 million lottery ticket, for most of us that would change our lives pretty dramatically. It would certainly change my life dramatically. And I doubt many of your physicians would say, “I don’t know, Peter. I hate change. You take that lottery ticket. I don’t want your $350 million. Change is bad. Don’t give me that money.” That would never happen. Why? Because it’s not change that people fear. It’s loss. And that loss has two components. A real component, we may ask you to use a checklist. You may work a little bit differently. But often a much, much larger perceived component, and that perceived component grows like a cancer, if you don’t manage the message. You‘ll get into the mess that I did in Michigan where people say, “Hey, the nurses are revolting. This is like Animal Farm.” And you don’t want to be in that position. You want to be in the position where you’re managing the message. You’re trying to understand. You’re showing your physicians that the real loss is very small, and you’re ensuring that you have high communication so that you completely minimize this perceived loss.

The next concept is that you absolutely need to what I call “manage the message.” What’s that idea? Well, that idea is whenever you walk out of a meeting for your improvement team -- so say that your CUSP Team has regular meetings -- ask explicitly at the end of the meeting, “Okay, what’s the message that we’re going to send out of this, if there’s any messages? Who needs to know, and who is going to deliver that message?” So often, we’ll make decisions out of a meeting. “Okay, we’re going to start a new protocol.” Or “This is the checklist we’re going to do.” Or “We’re going to try the daily goal.” And we just assume somehow, magically, that that message is going to get out of your head, out of the heads of those people in those rooms, and everyone else in the organization is going to know about it. All your docs are going to know about it. And they’re just, like fairy dust, going to say, “Oh, this is the greatest thing in the world. I’m so glad you did this.” And they may be on board, but if you don’t manage that message, you can guarantee there will be resistance. So when you leave there, if you don’t have all the stakeholders represented, take a minute to say, “Okay, well, the message is, we just decided to pilot test a new protocol for putting catheters in or for maintaining catheters or we’re going to audit our catheters. And we have to make sure all of our quality people know, our nurses know, the docs know. And who is going to take that message to each of those groups to make sure that they understand not just what we’re doing but, much more importantly, why we’re doing it?”

And then, lastly, and it goes to this changing social norms, it’s really important that you remind your staff that they’re part of something much bigger than any one of our hospitals. We are now a decade since “To Err is Human” has been published. That highlighted the significant number of people who suffer preventable harm. And we’re doing a lot, and I think we’re learning a lot, but the empiric evidence that patient harm has been reduced is virtually nonexistent. It’s virtually nonexistent, except in this area of ICU catheter-related infections. And each of you absolutely has it within your power to reduce these infections. There is no doubt you could all go a year or 2 years without infections. We’ve seen it in hundreds of hospitals. It’s hard work. It takes your leadership. It requires that you get your physicians on board, but you absolutely could do this with your hard work and so reminding them of this concept of ohana, that Hawaiian word for family. And family in Hawaiian means nobody gets left behind. Everyone of you, from the small rural hospital who has your own unique challenges, and no doubt this is hard for some of you, to the large, academic medical center, who has a lot of egos floating around in their organizations, that have their own challenges, that also have resource constraints, different kinds of constraints than the rural hospitals, but constraints nonetheless, you all can make significant progress towards this.

So what might be an action plan for you to do? Well, if you don’t have a physician unit-level leader, you absolutely need to try to cultivate one. You have to find a physician who’s trusted, that is viewed as caring about patients and is competent, that might be willing to put the time in, and hopefully, you can create a compact for this role. That is, you can go to your hospital administrators or they can go and say, “Here’s what I’m willing to do, and here’s what I need in return for this.” This idea of volunteering time to manage populations of patients isn’t going to be productive. This takes real work. It needs to be supported, and a piece of a physician leader’s time, like their nurse leader colleagues, absolutely needs to be devoted to this. You need to make sure you have a containing vessel so that you avoid those monsters in the bathroom. And that might be one for the nurses. It might be one for administration. It might be one for the physicians. But you absolutely need it especially for your primary care physicians or your private practice physicians, a way to communicate with them why this process is important, what you’re asking them to do differently, and to give them feedback on progress. I’d make sure you develop a communication plan. Get into the discipline of it. Whenever you have a meeting, walk out of there saying, “Okay, what did we just decide, and who needs to know? Who is delivering that message?” And most importantly, listen to your physicians who are resistors. At their heart, all clinicians want what is best for their patients. And if they’re resisting, there almost certainly is a barrier for them trying to practice evidence-based practice, trying to improve through CUSP and reduce infections. And your job is to get some ground truth and minimize those barriers so you make it easy for your clinicians to do the right thing, rather than just assuming if they’re resisting this that they’re just not on board and they need to go get steamrolled, or you have to go and get some senior executive to empower them. Believe that they want to do the right thing, and your job is to understand those barriers and mitigate them because the only way we’re going to be able answer Sorrel King’s really, really tough question, “Is Josie less likely to die?” is if we all work together on these efforts.

Donabeian, who was the father of quality improvement, and he spent his life studying quality improvement and working to improve it, was interviewed on his deathbed by one of his students, who asked him, “Now that you’re dying, and you’ve been a patient for the last couple of months, and you’ve devoted your life to quality, what do you think the secret of quality improvement is?” And his words, I think, are so profound. He said, “What I realized is that the secret of quality is love. If you have love, you can work backwards to change the system.”

And I think that’s the same answer that we found in our question of why did Michigan work? When we changed the social norms, we saw this as a solvable problem. We committed to work together. The physicians were on board, and they were leading this. It all boils down to the secret of quality of love, and I hope you take that and go work to make sure that all of you are one of those many and growing hospitals who are a year or 2 years without a CLABSI. I thank you, and operator, I am happy to take any questions.

Operator: Okay, at this time, if you would like to ask a question, please press the star key, followed by the one key on your touchtone phone. Again, that’s star, one on your telephone keypad if you would like to ask a question. We are currently holding for questions, so again, if you would like to ask a question, please press star, one on your telephone keypad. Our first question will come from George Sample with Washington Hospital Center.

George Sample: Peter?

Peter Pronovost: Hey, George, how are you? Great to hear you.

George Sample: Thank you very much. Peter, this is all great information, as usual, and I can understand how enthusiasm, etc., gets the ball rolling. But how do you sustain this for 6 months, 12 months, 18 months? We have these valleys, which are very difficult to climb back out of.

Peter Pronovost: Yeah, George, so that’s a great, great question. I’ll give you kind of the two sides of the coin. One is, as you may have seen, we published in New England Journal that those hospitals in Michigan sustain these low rates for now over 3 years. So it absolutely is sustainable. I wish I could tell you I really knew why they got sustained. I can’t. We didn’t formally study it, but we have interviewed hospitals, we’ve visited teams. And we have seen a couple of things that it takes to sustain it, and I’ll share some of those practices with you, George, and there’s not one practice, but there are many. They all boil down, though, to the view that these infections, a true belief that they’re preventable, rather than inevitable. So the first thing we saw, George, is that the hospital CEO has to commit to zero. And they have to know their infection rates. If they say, “Our goal is to be low. We’re doing okay. I don’t know what they are, but I haven’t gotten any complaints about it, so we’re probably okay.” It’s guaranteed that they won’t sustain it. Second, the ICU director, not the infection preventionist, needs to be the one accountable for it. And what we’ve seen at those hospitals who stay zero is every infection that they have, the ICU leader has to investigate and present to some senior management forum about what they found. Now sometimes they say, “Hey, we scratched our head. We don’t know why we have this infection. It may be patient disease.” But that degree of accountability is absolutely key to stay at zero. That doesn’t mean that they don’t partner with the infection preventionist, they absolutely do need -- the infection preventionist brings technical expertise and measurement expertise -- but fundamentally, the clinicians putting in and using these catheters have to own the problem. Second is they have to make sure that it’s always easy to comply with the checklist. And this might be where we see the most defects, George, because many hospitals -- I’ll give you an example, including my own. We had a couple flurries of infections in one of our ICUs, and as I just mentioned, we investigate every one that we have. And by investigate, I mean we look to see whether evidence-based practices were followed. But more importantly, we go get ground truth. We literally go watch a catheter being placed to see what changed. We talk to staff. And what we found is that somehow, God knows how, our full barrier drape got substituted out. A small, little window drape got put into our kits rather than a full barrier drape. Now, how did that happen? Who knows? We’re a big organization. Supplies come in through many different ways. I would have thought that the clinician should know enough to say, “No, this isn’t good enough,” but predictably, they’re busy. It was in the kit. It’s what they used. They made do with what they had, and that was the cause of our infection. So we swapped it out. But if I just assumed it was either docs not using the checklist, we have to go with them without getting ground truth, I would have never found that. So every infection, go get some ground truth and make sure it’s easy. I think keeping tabs on this cultural issue, that if not you, George, and I know you have a great relationship and you work really hard for a culture of safety with your nurses. When I go into these hospitals, I’ll often pull a junior nurse aside and say, “Hey, just confidentially, if you were to see a senior doctor not use the checklist in this hospital, would you feel comfortable speaking up, and how do you think that would go well?” And sad to say, George, in the vast majority of hospitals still in this country, the nurse looks at me and says, “Are you nuts? There’s no way I would speak up. Do you want me to lose my job?” And I think we have some culture work to do. The ones who have sustained it, George, have focused on maintenance. I think what we’ve seen is that the insertion checklist and practices are much easier than maintenance, I think, and most of us have licked those. And that gets us a long way, but now the infections that we’re seeing are often in people left catheters for a long time, and our maintenance practices aren’t good. I think technology likely has a role in preventing some of those infections, and people who have catheters for a long time and that needs to be explored. And then, lastly, George, the staff routinely gets feedback on typically both their infection rates and some weeks or months without infection so that they are continuously understanding where they stand. Now that might sound like a lot, but basically what it means is that they have a system of accountability or what I might call this chain of accountability from the CEO committing to zero, the infection preventionist supporting the work, the ICU director and physician and nurse managers being accountable, and making sure that supplies are there, investigating all of these infections, creating a culture of collaboration, focusing on maintenance and getting data back. So it’s not an easy task, but as I say, if there are 25 percent of hospitals in Michigan who can go 2 years, no doubt all of you are at least that good.

George Sample: Very good, Peter. Thank you.

Operator: Thank you. Our next question will come from Sherry Leuthold with Stanford.

Wendell Hoffman: This is Wendell Hoffman from Stanford. Thank you for this great presentation, Dr. Pronovost. I came in just a little bit late but tried to catch up with your content, which is always very meaningful. First is a comment and then a question. First of all, the comment is we’ve had terrific interest amongst our physicians around the CUSP initiative. At our last monthly meeting, I asked the question, “What is it that brings you back here? Why do you want to continue this?” because we have now put together a standardized policy for placement and maintenance of lines in our organization. We’re going forward with our educational initiative and more broadly with our accountability initiative, which is probably the more difficult question. But when I asked that question, the answer largely was they want to be there, surrounding, interestingly, the accountability measure. They were very interested in basically how professionalism is played out with regard to adherence to best practice guidelines, leadership in a unit, and so forth and so on. So, I came in with some of your last statements about what is at the heart of quality, and I couldn’t agree more, and I think we’ve experienced that here.

My question has to do with, and I think I asked this before, maybe a number of months ago. Whenever you start these things, you always follow the data and, of course, one of the places that the data with regards to line-related infections has occurred is in our dialysis population. And so we started meeting with that group, sort of as a CUSP model as well, and so we see some significant potential there. But my question is, have you seen this used in other subcategories of central-line infections, like in dialysis or, say, in the oncology group where you have portacaths and different kinds of central lines that are being accessed, and we think probably impact those infections?

Peter Pronovost: Great comment and question. Let me just make a reply to the comment, because I think this is a really key issue. The feeling I get with national quality is, “Oh, these bad doctors and nurses, and they just need to be spanked, and they’ll do it.” And I will tell you from my own mistakes, when we started this and our rates of CLABSI at Johns Hopkins were 15 or 18, unacceptably high, I was one of those docs causing those infections. And I know I certainly didn’t want patients to be infected. I just didn’t know any better, but my mental model wasn’t that they’re preventable. But, certainly, my intent was to do what was right for them, and I think it’s important for us always to remember that. The other part of your comment that why the docs are coming back. Because again, there’s overwhelming data in the sociological literature that when you invite people to participate, that you don’t dictate to them and say, “Here’s the checklist you will use.” That will fail miserably. There is no one right checklist. The right checklist is the one you develop in your organization because it fits your context, and having these forums for participatory leadership and involvement are absolutely keys to success.

The second question is, yes, this approach, not only does it work for outside of the ICU for CLABSI, but it works for other infections. So a couple of backgrounds: We published a few months ago how we use this same approach for VAP. The evidence-based practices are obviously different. CUSP is the same no matter what you’re working on. It’s basically getting local unit ownership for a problem. We have just started a project, and we’re happy to share with you, to target dialysis patients for CLABSI. The reason is, while the CDC reports that ICU CLABSI are down like 60 percent nationally -- it’s really been a remarkable success, virtually every other infection is flat -- CLABSI in dialysis patients seem to be going up and an increasing cause of their mortality. And so, we’re trying to understand this same approach, and let me just summarize some of that. The CUSP team is the same. For putting evidence into practice, we’re trying to see what are those evidence-based practices that work. Once we have those, and believe me, that evidence is less mature than it is for ICU CLABSI, what are the barriers to doing that. So we talk to clinicians. We observe clinicians. We go walk the process to see, “Okay, when you show up for dialysis, what does it take for me to actually hook you up, to sterilize the line, to hook you up to the machine, where are the potential breaches, what’s hard about it?”And so we’re early on, but we’re quite hopeful that this method absolutely will apply to those efforts. And there’s an effort in pediatrics, my wife is actually leading it, to do the same approach for chemotherapeutic patients who have long-term indwelling catheters. Again, the technical details of what the evidence practices are and the barriers vary, but the approach is exactly the same. You need a unit-level ownership of the problem. They need a structured approach of finding what are the evidence-based practices, what are the barriers to those, what are the measures of performance, and then, how are we going to change the system to make sure every patient, all the time, gets the evidence. So I really think this is great, and maybe we can communicate offline. If you’re learning about dialysis patients, we’d love to hear about it. And we can certainly share with you some of the stuff we’re starting to do for dialysis patients.

Wendell Hoffman: Thank you very much. I would be interested in dialoguing with you a little bit offline in part related to literature pertaining to research in the area of safety and quality, so if I could figure out how I can accomplish that, that would be wonderful.

Peter Pronovost: I think you can go through Stop BSI. There’s an email JHMIstopBSI@jhmi.edu, and our team looks through those and then funnels them tonwhoever the appropriate person is for a question.

Wendell Hoffman: Very good. Thank you very much, very helpful.

Operator: Thank you for your question. Again, if you would like to ask a question, please press star, one on your telephone keypad at this time.

Kristina Week: Hey, Peter, while we’re waiting on other questions; this is Kristina, also from Hopkin. I was talking to one of our teams, I don’t recall which State, but they were a smaller hospital, and they only had one or two admitting physicians and they seemed to be struggling with engaging one physician on providing all of the best practices. And it didn’t sound like he had his patient as his North Star, and I thought maybe help folks brainstorm, if you’re in a situation where you only have one or two docs to work with and they don’t want to play right.

Peter Pronovost: Yes, so that happens all the time. I think there are some unique challenges at these small, rural hospitals. It’s also a challenge at our bigger ones. What I’ve seen work for those kinds of challenges is a couple of things. The first, and it’s a really important lesson, is to remember that whenever there’s a conversation going on, and this comes from a book, “Difficult Conversations,” that you could think if there are conversations going on at three levels. There is conversation about what are the facts. “You’re asking me to use the checklist.” There is conversation at a deeper level, often internal, about “What do I feel about that? Well, are you telling me that I’m killing all of my patients with infections? Are you saying I’m a bad doc?” And then the third, deeper level is, “What are you saying about me or my profession, about me as a person?” We see this all the time when new interns start, and a nurse questions them about something. And the facts are benign and trivial, but what that doc, that new doc, is internalizing is, “I just spent 4 years in medical school and all this money and I’m in debt. Are you telling me I’m incompetent because I don’t something and you’re challenging me?” It has nothing to do with the facts. This is kind of inner conversation about what it means about me and who I am and my role as a physician. And so oftentimes those deeper conversations cloud or they make it difficult for us to focus on the patient center areas. So how do I do those? Well, one, is I have a conversation with that person. And I always do it alone and in a safe space, always alone and in a safe space. This isn’t something you do in public because it’s just going to create that loss. You’re going to create an embarrassing situation, and they’ll explode. I make it safe to open up, so I always say that I validate their commitment to your patients. “Hey, Dr. Jones, I know you’re a really good physician, and you care deeply about what’s best for your patients,” so right away disarm them, acknowledge that you’re not judging them, that they’re a good clinician. “So I was wondering if you could help me understand why you seem to be resisting this project, or you don’t want to use the checklist because I know you care deeply, but our infection rates are high, and there’s no way it’s going to work without you. We need you to do this project.” And again, I would be very direct. I would say that you need their leadership. “This isn’t going to work if it’s only nurse run. It isn’t going to work if you don’t’ have one of the few physicians who are doing this. So there must be something that I’m not seeing. Can you help me understand what you’re resisting?” And then just be quiet and listen. And you might have misinterpreted the resistance. “I’m not resisting. What do you mean? Nobody asked me about this. I didn’t know about it. Of course, I’m happy to do it.” It might be, “Well, I thought this was an infectious disease thing. I’m an ICU doc, and I didn’t think I was involved in this,” or “I’m a primary care doc. What do I know about this?” Or, “I’m really busy. I heard that this thing is going to take a lot of time.” But once you kind of surface those issues, then you could really start to persuade and understand how the perceived loss is much greater than the real loss because these projects don’t take that much time. If anything, they often save the time of clinicians, and you could have that conversation to say, “Okay, well, how could we make this a win-win for both of us so that you could get what you need out of it, and I can tune into what’s in it for you, and we could make sure that the organization and our patients get what’s best out of it.” What I see doesn’t work, Kristina, is if we judge. So, in other words, we sit there, and that doc doesn’t seem to be in it, and I walk into the nursing break room, and I put my hands on my hips and say, “That Dr. Smith just doesn’t get it, right? He’s not into this project. He doesn’t want to wash his hands. He doesn’t listen to us. This culture’s never going to change.” That’s not a very constructive thing to do. Now, it’s scary as hell, at times, to go have these conversations with people because you don’t know which way they’re going to go. But I can tell you the vast majority of the time, just like any problem we face in our life, personal or professional, if we have the courage to address them, and address them in a respectful way, you’ll always end up at a deeper, more meaningful place. Now, I’ll share with you, Kristina, I had this amazing conversation with this nurse, a feisty little thing, who probably weighed 80 pounds. And she had this doc who she thought totally wasn’t on board, and their infection rates are high, and they didn’t have many catheters, but the few they had had really high infections. And she said, “Peter, I’ve been here for 20 years, but I never really spoke up, and I feel like I really need to advocate, so I’m going to go say something to him.” And she was scared to death. She was really petrified. She walked up to him, and in these community hospitals the small number of physicians have an amazing amount of power. They can get nurses fired on the spot. And she read the script. She said, “I know you want what’s best for your patients, and so I was wondering if you could help me understand why we’re not doing these things.” And she was all ready to be fired. Indeed, she said, “I just called my husband and said, ‘Morally, I feel like I need to do this for my patients, but as you know, I’m going to be gone.’” And much to her surprise, the doc said, “Well, I didn’t think you guys wanted me to play. I thought it was a nursing thing or a hospital administration thing. I think I have something that I can contribute to this. If you want me to play, I’d love to be able to contribute and be part of the team.” And they went on to having this great relationship where the guys are a safety meter now. So I think putting some structure to how you do this and kind of avoid the whining and the judging are the initial first steps. Other questions, Operator?

Operator: At this time, there are no further questions in the queue, but again, if you would like to ask a question, it’s star, one on your telephone keypad.

Peter Pronovost: So as we’re waiting for questions, let me just remind you that -- and we will close up here – with this program, you’re all part of a national community, perhaps international, because several other countries are doing this. And we’re trying to learn and share together. Within your State, you’re a community, and there is a support network. But perhaps the most powerful community is within your own hospital, and so I would make sure that you have a structure to house that community, that you nurture it, that you get broad input, that you seek consensus, that you don’t dictate because your local community is where this thing is going to survive or die. And frankly, it’s where your patients are going to either be safe from these infections or suffer from infections. And it’s only going to be with your leadership that you create and house that local community. So with that, Operator, we’ll sign off. It’s getting near the hour. For our colleagues at HRET, at MHA, and our Johns Hopkins colleagues, we thank you for all of your work on this. I think we’re somewhere near 160 lines on this call. It’s just an amazing amount of support. I would encourage you to keep trying and remember that question Sorrell keeps asking you, “Is Josie less likely to die in your hospital?” I thank you, and have a great day.

Operator: This concludes our teleconference. You may now disconnect your lines.