Don Berwick:
Think of the network as a latent laboratory that’s already working. So there’s already experiments underway. We call that experiment variation, because you do it one way, I do it another way, and if we can reveal that, think of all the time we can save. I don’t have to try everything. Someone else has tried it out too.

Alec Patton:
This is High Tech High Unboxed. I’m Alec Patton, and this week we are welcoming Don Berwick back to the podcast. We first had him on the show in February, interviewed by Stacey Caillier, head of the Center for Research on Equity and Innovation at the High Tech High Graduate School of Education. And Stacey had more questions for Don, so she interviewed him again.

You don’t need to listen to his first episode to understand this one; it’s not a two-part series or anything like that, but if you haven’t heard it you should, because it’s just really good. It’s season 2, episode 13, and it’s called Don Berwick on Improvement is Learning. Don Berwick is a pediatrician. He’s also been president of Institute for Healthcare Improvement.

He served as administrator of the Centers of Medicare and Medicaid during the Obama administration, and he received an honorary knighthood from Queen Elizabeth II for services to healthcare. But in spite of all of this, his ideas about improvement are pretty radical. He once wrote an article called Confessions of an Extremist, and listening to this interview, you can see why, in a good way.

Here’s his interview with Stacey Caillier.

Stacey Caillier:
So Don, I’m so excited to talk with you again.

Don Berwick:
It’s a total pleasure, Stacey. Nice to talk to you again.

Stacey Caillier:
And I wanted to actually start with an appreciation, because last time we talked, you shared this beautiful story about co-design and asking those we serve how we can get better, and you shared a story about driving in the car with your boys and telling them that you had thinking a lot about improvement, and did they have any ideas of how you could be a better parent.

And of course they had super thoughtful feedback, as kids often do, and I just want you to know that I thought about that story for a long time afterword, and also how after they shared some feedback, they asked you how they could be better kids, and just that beautiful reciprocity. And I think especially right now so many of us in education really believe in the power of student voice, but we don’t always take the time to ask that simple question, like “How could I be a better teacher?”

And I realized I certainly don’t do that as a parent. So I wanted you to know that you inspired a trip to the mountains for me with my six year old, Asa, where I asked him that same question, how I could be a better parent, and he had amazing feedback of course. So I just wanted to think you for inspiring that, and being that reminder and inspiration for asking that important question.

Don Berwick:
How cool. I’d love to meet Asa some time. And thanks for taking it so seriously.

Stacey Caillier:
He is full of personality. You would love to meet him. All right, well so since our last conversation, I’ve had so many questions percolating in my brain that I wanted to ask you about, and also just dig a little deeper into some things that I know our team has been grappling with and that I keep hearing from other folks in education that they’re grappling with.

So I know we talked a lot about how do you build a culture where improvement can flourish, and how do leaders need to show up in that culture to support that, and I’m hoping that today we can dig into how do we support really robust learning networks, where we’re learning from variation and learning from each other and dig into that a little more. Sound good?

Don Berwick:
Absolutely.

Stacey Caillier:
Okay. So we’ve been talking a lot about learning from variation lately, and your friend and author of The Improvement Guide, Lloyd Provost, has been quoted saying, “The purpose of networks is to learn from variation.” Can you say a little bit about what that means to you, and what you think learning from variation actually looks like in practice?

Don Berwick:
Sure. Well, if by network we just mean a community of actors, people, organizations, whatever, that are sort of trying to accomplish the same things, we can start off by a very strong assumption that they know more together than they do separately; that the pool of knowledge exceeds that in the hands of any one particular member, so why not put that knowledge to use?

If we’re both trying to make a better chocolate cake and you know some stuff about chocolate cakes and I know something about chocolate cakes, I’d like to know what you know, and you’d probably like to know what I know, because you can try it out locally. We will avoid the mistake, assuming that what you’re doing I can simply plug into where I am. I can’t. I have to adapt it and vice versa, but that doesn’t mean your knowledge isn’t valuable to me. It is.

There’s a social side to it also, which is change is difficult. It’s uncertain. When you try to do something better, you can fail, and it’s often helpful to have somebody with you that says, “Oh, I know what that’s like.” I mean just the social support system of being in a network of trying to do things together, it’s more fun and it’s more sustaining. Someone can help you get up when you fall down. You can celebrate together.

You can grieve together. The variation that Lloyd Provost is referring to is the observation that we’ll be varying a lot. Like you’ll do something different from me because we haven’t got a standard together and so automatically … sort of think of the network as a latent laboratory that’s already working. So there’s already experiments underway. We call that experiment variation, because you do it one way, I do it another way, and if we can reveal that, oh my goodness.

Think of all the time we can save. I don’t have to try everything. Someone has tried it out too. We’ll also have variation in results, because if you’re trying to do the same thing that I am, like teach a kid to learn algebra or help a patient not mix up her medications, then one of us is probably getting a better result than the other. It’s unlikely we’re identical, so if I’m getting a better result and it’s non-random, we have an opportunity to actually have a … we can learn from realtime data.

And so that’s another kind of variation. It’s the variation in outcomes or results that we can then work backwards and say, “Oh, I wonder what did that.” That of course requires transparency and the courage that underpins transparency; the courage to discover that someone does something better than I do, and not be embarrassed or shamed, and the courage to think that maybe the way I’ve been doing it might not be the best way that I could do it.

But you can set up a network with those social characteristics. So Lloyd, as usual, is exactly right. All sorts of variation and all of it potentially instructive. Think of improvement as the enterprise of making invisible variation visible for the purpose of learning.

Stacey Caillier:
I love that. I’ve always appreciated that you seem to approach the work with kind of this unconditional positive regard for people, and I think that equates really nicely to how we try to approach improvement too, is like we want to do a lot of bright spotting. Like we use that as a verb; like we want to surface where people are doing amazing work and dig into that, and excavate it so that we can spread that to other folks and help them make sense of what that could look like in their context.

Can you share a little bit about how you … like what are helpful ways of surfacing and understanding bright spots, so that other people can learn from them too?

Don Berwick:
Transparency helps. An episode that I think Lloyd was involved in … I may have mentioned this in our last conversation; it always sticks in my mind … is a collaborative network that I was working with on organ transplant. Does that ring a bell?

Stacey Caillier:
Mm-mm (negative).

Don Berwick:
Well, the secretary of Health and Human Services at the time, Tommy Thompson, wanted to haveHHS support a national endeavor to increase the availability of organs that could be transplanted. A lot of people are willing to be organ donors, should they die, but a remarkably small number of the available organs are actually harvested.

Most of the organs that are harvested in the United States at that time were coming from about 200 hospitals, but they varied. They varied a lot in the percentage of the available kidneys or livers or corneas that were actually harvested. The variations I remember was from about 15% of the ones that could be obtained up to over 80% of the ones that could be obtained.

So we, the Institute for Healthcare Improvement, formed a collaborative led by a wonderful guy named Dennis Wagner, and we invited all of the procurement for all the hospitals that were doing organ procurement together. We said “All right, you’re varying in the percentage of organs you harvest. There are a lot of people on waiting lists for organs. Should we study how we’re doing it differently?”

And they said yes, so the first thing we did was get the data on each place. Like I was getting 15%, you were getting 80%. I think there were a lot. I want to remember well over a hundred organizations, maybe a hundred and fifty or something, and I remember the day we had a meeting and we said, “All right, now we know. We know who’s got the highest organ retrieval rate and, by the way, who has the lowest.”

And we said to them, “Do you want to light up your dots?” Meaning we have a slide, we’ve got anonymous dots. We could put a name by every dot, and everyone here could see where they are and where everybody else is. What do you want to do?” Little bit of squirming in chairs, little bit of rumbling and whispering, a little bit of groaning, and then I remember that moment when everybody said, “Okay. Light up the dots.”

And there were no dissenters and we lit up the dots, and then we knew. We knew who had the highest procurement, who had the lowest procurement, and that began a whole series of meetings and visits and rounds on what are you doing, how do you do that, how do you do this, how do you do that, and the culture of openness and exchange then became a culture of learning, and everyone got better.

We say, and I try, “I’ll teach. I’ll learn.” Because even the lowest procurement places were doing something different that was informative to everyone else. That was also the first time we used a live case that I mentioned at a meeting you and I were at, which blew my mind.

Stacey Caillier:
Please tell us about that. That was going to be one of my questions.

Don Berwick:
This was an invention of … I first encountered in the Harvard Business Review paper by a colleague of mine named David Garvin who described the following approach to that kind of variation. Well, I’ll tell you how it played out in the organ transplant world. So the live case approach is this: you discover, in this case, who’s high and who’s low in getting organ transplant, and let’s call them the team. A team from the low-end place, the low team is formed.

And that would … maybe a surgeon and a nurse and a manager and someone else, and then there’s a team at the high place, the best place, and the lowest performer team visits, physically flies to, goes to, the highest performer for a day, and that day is organized as follows: the day begins with the high performer presenting its processes in a room to the lower performer. “Here’s how we approach organ donation procurement.” And they walk through the whole thing.

That’s meeting number one. The second part of the day, the visiting team, the low team, breaks up and they go around and they say well, they talked about the nurses, they talked about doctors and surgeons, they talked about the operating room, they talked about the emergency department. And they visit all of those places and they say, “Well how do you do it here?”

And what they discover is that the actual process is not the same as the hosts had presented. Hosts were presenting honestly, now the visitors were learning more than the hosts knew. The third part of the day is the visitors then present to the host the actual process. They say, “You said you did it this way, but actually not. What’s really going on is this.” And then the hosts respond and think about it, and then everybody goes home.

And what happens is amazing, which is improvements everywhere. What we did in the organ procurement collaborative was we then had the next session. We had the next collaborative meeting. Everybody came together and then we had a session in which the visitor and the host, the high team and the low team, co-presented a session to everybody describing what they discovered.

Now that’s a culture of learning. That’s a culture of improvement. Think of the trust and the transparency, and by the way, the fun and the excitement of what was being discovered. That’s a live case and it’s a manifestation of the kind of culture you’re asking about.

Stacey Caillier:
I’m really struck by that example in particular because I think we talk a lot about peer-to-peer learning and we talk about that in schools too, but what you’re describing is people seeing each other as resources in their learning, and it’s going both ways. It feels very reciprocal, and that just feels powerful.

Don Berwick:
Yes. Each one sees themselves both as a giver and a receiver. Because everybody gives gift. Everybody gives the gift of explaining what they do, and everybody receives the gift of others explaining, it’s amazing.

Stacey Caillier:
Mm-hmm (affirmative). Are there other ways you’ve seen that generate that same kind of peer-to-peer learning that you would recommend?

Don Berwick:
Yes. First of all, COVID notwithstanding, I’ve become a real fan of visiting. I think an actual on-site visit in the workplace beats a conference room or a phone call any time. It definitely beats third-party learning, because a lot of places would have done this for that; they would have noticed that this hospital has the lowest procurement rate and this hospital has the highest procurement rate and then they would have sent a team of researchers or journalists or something and said, “Okay, go visit here and find out what they do, visit here and find out what they do, and then tell us what you found.”

That is a very weak octane way to learn compared to the actual visiting and the visitors become teachers. A couple of other ways I’ve seen this, one is extension agents. This is practitioners who have been successful becoming the teacher. So in that case, maybe you would have set up a system in which the surgeons and the nurses and the managers in the high-performing place would be given time and support to travel around, visit places and teach each other what they’re doing.

That’s very powerful. Another is just observation. There was a wonderful group in the olden days called the Northern New England Cardiovascular Disease Study Group. This was the cardiac surgeons of Northern New England and I think six or seven medical centers. The government had published data on cardiac surgery survival, which showed large variation. They didn’t believe it, so they organized their own private consortium to study their own outcomes, and what they discovered was the variation was even greater than the government had published.

So they began, quite courageously, a round of visiting. So everybody visited everybody. So the surgeons from the University of Massachusetts would visit and operate with the surgeons of University of Vermont, and they began noticing things as they actually traveled and worked together. The equivalent in education would be teachers who then team up and teach in each other’s classrooms.

Nothing beats visible participation; just being together, it really does work.

Stacey Caillier:
That’s awesome. Thank you for sharing those examples. I want to go back a little bit just to highlight what you said about third-party things, because that’s what I see happening in education a lot with improvement; not necessarily putting teams in touch with each other or observing each other, but whoever the hub is … and we’re guilty of this too … does a lot of work around figuring out what’s happening in the places where great things are happening, and then packaging it in a way so that it can be taken on by everybody, but we’re kind of serving as this intermediary.

And we’ve been thinking a lot about like how do we just get out of the way so that we can have people learning directly from each other? So I really appreciate that push.

Don Berwick:
I think it doesn’t have to be an either/or. There probably is a role for researchers, observers, ethnographers and so on to say well we’ve visited these eight places and they seem to be doing the same thing, but I think my experience is perhaps not different from others, which is that third=party document tends to end up on a shelf, you’re lucky if anyone reads it, and then if people read it, it’s still a long way from implementation.

The beauty of the visit, of the actual peer-to-peer encounter, is that the dynamics of learning can be so robust because I can say, “Wait a minute Stacey. You just did that. I don’t understand it. That’s not what you told me. Why did you do it that way?” Or, “Could you explain a little more about this?” Or, “Wait a minute. I’m left-handed. You’re right-handed. How do you think that would…” We actually can have a dialogue, which you can never have in the third-party report.

And so I find it a much more motivating, agile instructive way to learn. It’s viscous. I mean you have to visit. That’s time. It takes money and energy, but I think the pay-off is well, well worth it. And if you remember the way I described my visit, a very important part of that was the live case we that did on organ transplant. When those two teams stood up the next plenary meeting and said to everybody, “We visited them. Look at what we found,” that’s a far more instructive event for everybody than the glossy tape with a beautiful diagram on the shelf.

I do think that there’s some importance in extracting maybe basic principles or something, so I wouldn’t ever want not to have the more research-y or journalistic approach. Maybe we’ll get some principles out of that, but if you want change …

Stacey Caillier:
Yeah. This reminds me a little bit of just this moment that you’re talking about with the organ transplant group of are we going to light up the dots and are we going to actually put our names on these things? It reminds me of this … In the last time we talked, you talked about a story, I think it was around cystic fibrosis, and how there was a consortium then all working and they hadn’t unblinded their data yet, and you described this beautiful moment where there was a parent of a child who had cystic fibrosis, and you turned to her and asked what she thought, and I think she said something to the effect of like, “I don’t know, but I can tell you that for my daughter, the clock is ticking.”

Don Berwick:
Clock is ticking.

Stacey Caillier:
Yeah, where they decided like okay, we need to make this data public so we can learn from each other and usher in this new era of collective learning. And can you just say a little bit about what can education learn from this story? Because I think we have a such a long history of having data used for accountability and not for learning that I see this scenario happening all the time, and so I’m just wondering like how do you think we get to a place where we can just share data openly and accelerate our collective learning?

Don Berwick:
Yeah, well the person you’re remembering, her name was Honor Paige, and her daughter, I think, was Laura, and yes, she said that the clock was timing the time when they were very worried about whether to reveal the data or not, but it did change things. I mean thing first about what she meant by that. She meant not one thing, but two things. I think she meant that well, you’re worried that if I discover that the University of Minnesota is better than the University of Wisconsin that I’ll take my kid to Minnesota, you bet I will.

Why as a parent would I not take her to the best place? So who are you to think that you can hide that information from me? You don’t have the right to do that. Now Honor didn’t say that, and maybe she didn’t mean that, but she could have meant that, and there is a power issue here which has to do with the customer. So if in the school world if you discover that one classroom is performing far better than another, or one school performing better than another, you bet I’m going to fight for my kid to be in the better one, and who are we to say, “Sorry, you can’t know that because we don’t want to give you that power.”

So that’s part of the power shift. Of course, the rest of it is all about learning, and I think because of that fear of the power shift, people run away from the learning opportunity, which is the other part of this, which is if my kid’s at the University of Minnesota and it’s better than Wisconsin, can’t Minnesota help Wisconsin? Can’t you help those kids too? And the answer is yes, you can.

You can go over and help them learn and teach, and I think that that’s … The problem with transparency is that both are going to operate. Don’t be naïve. If you show data, people will use the data. They’re not stupid. But you also then have the opportunity to learn. My experience with this in healthcare is it’s the fear is way overblown compared to the fact, so that when you actually do get over the hurdle and try to light up your dot, very little bad happens.

You can imagine some bad stuff; competition and shaming and blaming or whatever, but that’s gone fast, and the learning can stay. And so I urge people to get over it. In education I think it’s the same thing. I mean if there’s a teacher in the school who’s getting something done really well under the same conditions that others are, can’t we find him or her and say please tell us more? And if it embarrasses on or insults another, let’s get over that.

I think in education the clock is ticking, because you can think you have the leisure to think about it and next month we’ll do this, and next year we’ll do that, but these are kids right now in the classrooms right now who could be better off if we acted, and so I sense some of the same issues of tempo that we ought to be facing in healthcare probably belong in the education world too.

Stacey Caillier:
Yup. Yup. Have you seen anything in healthcare that’s helped people move toward that kind of transparency? Like move past this like, “Well, I … ” defensiveness, or feeling of being threatened by somebody who’s doing something better? What’s helped that shift?

Don Berwick:
Two things, at least. One is the presence of the patient in the room, like Mrs. Paige at the meeting with the cystic fibrosis foundation board. I’m convinced the same would be in education. If you get kids and parents in a room, they’ll know what’s important, and they’ll say, “Excuse me, this is what’s important. Can you please work on this? Because you’re wasting my time and yours on this.”

And I think it happens in healthcare, and I think it could happen in education, especially if the invitation’s authentic. People are going to want to be on good behavior, and it is an art form to set up the circumstances in which they really do feel comfortable speaking up, but it’s self-reinforcing. So that’s one. The other is a little more … It’s that learning is fun compared to accountability, and so when you actually get into the cycle and people are … the dots are lit up and you’re actually saying, “Excuse me, Stacey, how did you do that?”

Smiles come. I mean you start to really … People care about their work. You said earlier there’s a kind of optimistic view of the workforce. I have an optimistic view of the workforce. I think every workforce has people in it who are not nice or troubled in some way that keeps them from being playing well with others, but I think it’s a good bit that … what, 85% of the people doing the work, the doctors, the nurses, the teachers, the aides, 85% like to do well, and they kind of would enjoy doing better, and so set up a context in which that’s possible.

Stacey Caillier:
Yeah. One question that I wanted to ask you about, too, because we’ve talked a lot about bright spots and supporting peer-to-peer learning, and I’m curious … In improvement, we also talk about that the idea is to reduce negative variation also, and within our own networks, we have some schools who haven’t made progress for a couple years, and we’re trying to kind of figure out like how do we best support them.

I feel like some of it is the peer-to-peer learning stuff that you’ve already talked about, but I’m curious, how do you think about supporting teams or sites that aren’t making progress, also knowing that that’s our responsibility? It’s not all on them, too.

Don Berwick:
I think that the answer’s probably much more contextual than makes it safe for me to be sure or comment, but I say … Well, first, stop the blame. I mean blame has no role in improvement, and any intimation that these are worse people than some others, or that they’re not trying or whatever probably is wrong. Creating a safe climate for inquiry, it matters a ton.

I think visiting is probably way underestimated as a tool, and I really do think it’s a fantastic thing to get people into each other’s shoes and spaces and say work together. And one of the other things has to do with Everett Rogers’ diffusion curve and the concept that there … in any setting, even the school that’s in trouble, there’ll be some early adopters.

There’ll be some people there who they really would like to try something, and if you find them and help them, they’ll become your partners and agents of change. And so that’s part of it. It’s a stratified view of the workforce. I guess one other thing I have to mention is Deming’s idea of constancy of purpose. Anybody, certainly places that are in trouble, the ground is constantly shifting under … Today it’s this headline. Tomorrow it’s that problem.

The day after, it’s this crisis. And that drives people crazy, so some sense of constancy, some sense of “We’re going to stick with this,” I think matters. And there’s a wonderful … I’ve come to know a wonderful criminal justice reform group in Lowell Massachusetts called UTEC; United Teen Equality Center. U-T-E-C. And UTEC works with kids who are justice-involved, mostly young people who’ve been in prison, and their aim is to get them back into a sense of self-worth and efficacy and productive roles in society and some sense of accomplishment.

And to reduce recidivism rates. Recidivism rates for that population are 70% other 80% will be back in prison in five years. UTEC’s got it down to 15%. One of the things that I’ve noticed there is their fundamental value is … they call it love. Mad love, they call it; M-A-D love, and it means we’re going to be with you, and no matter what you do, we’re going to stay with you. And it’s this sense of foundational security that you can’t get rid of us.

Don Berwick:
No matter what you do, we’re still going to be with you. I think that’s fundamental to the effectiveness of that program, and the same applies to schools in trouble, which when you come in, instead of saying “You’ll be out of here in six months if you don’t fix this,” you say, “I’m going to be with you no matter what.” Think of what that does.

Stacey Caillier:
Mm-hmm (affirmative). I love that. The tagline for our CARPE College Access Network this year was “Tenacious Love.” Like we’re going to get you through. No matter what, we are not going to give up. So this kind of … Something you said just a minute ago around kind of moving past this doing harder, I think that’s-

Don Berwick:
Trying harder.

Stacey Caillier:
Yeah, trying harder. I think that’s something we encounter a lot in improvement and that takes some shifting out of like how do you move folks from like “I’m just going to keep trying harder” to really rethinking their systems. Do you have any thoughts about that?

Don Berwick:
Yeah. It’s W. Edwards Deming. He wrote “Trying harder is the worst plan.” There’s a logic here, let me be a wonk about this. Okay, so what’s the fundamental science of improvement? In my world, it’s systems science, and it’s the idea that every system’s perfectly designed to achieve the results it gets. So if you have a certain death rate in your cardiac surgery unit or math readiness achievement in your school, that performance, it’s the property of the system; the system’s perfectly designed to get you that number.

Those numbers vary, so they’ll be common cause variation, random variation. So it’s not the same number, but it’s in a band. There’d be variation. In order to be better, fundamentally, you have to change the system, because you need a system with a different characteristic. That could be any number of changes. Hundreds, thousands of changes might make a difference, but pushing the current system?

No. It’s going to get you what you’ve got. So what Deming … When you try harder, you get a little blood from the stone. That day, that effort, you will get a little, often, at the cost of demoralization and pressure and pain, but you’ll get a little more. When Deming said trying harder is the worst plan, it’s defective first because it isn’t change. It’s the opposite of changing a system, which should make things easier, by the way, and second, it temporarily reinforces your belief that you could squeeze more out of the system.

It’s really toxic. It’s a very bad thing. And it’s everywhere. It’s just a matter of trying. The other thing is trying is a personal thing. So most of the performance we’re seeing in healthcare, and I suspect schools, is coming from interdependency. I’m sure teachers are important, I’m sure doctors are important, but overall, performance is not about this teacher doing that, I don’t think.

It’s about interactions and relationships and so when you say try harder, usually you mean you try, or you, individual, try harder. You be a hero. You overcome this obstacle, when actually that’s the wrong answer. It’s got to be done together.

Stacey Caillier:
Yeah. How do you shift people from this trying harder to revisioning the system and rethinking the system?

Don Berwick:
I wish I knew. I do believe it’s highly leadership dependent. So in a hospital, we look to the chief of medicine or the chief of surgery, to the head nurse, to the CEO, to the CMO, CFO, because when the leadership changes, and that has to include the board, then you could start to create the possibility of a different approach to variation and the conditions for learning.

In a school, I’m not sure yet. I suspect principals, superintendents, district leaders, boards of education, I think they probably matter. So can we get their attention and help them start to rethink the conditions that they like to create in schools to permit learning to occur? More optimistic moments, I say look, even if the CEO’s not on board, even if they just use the try harder mode, you use … doctor, you nurse, you teacher … you have some span of control.

There is something you control. In that box, you could behave this way. You can take learning to that scale, and create an environment where it’s safe to celebrate defects and study variation and light up the dots and all of that just locally in your own little place. So that’s another answer.

Stacey Caillier:
Mm-hmm (affirmative). Yeah, I love that. Just kind of like don’t lose sight of your own locus of control.

Don Berwick:
Yeah. Try to do that, yeah.

Stacey Caillier:
Yeah.

Don Berwick:
The other piece of advice there always is … We say in IHI, “Never worry alone.” Find two other teachers that want to do it, because a group a three is more than three times as sustaining as just one. It’s geometric, not arithmetic. So even a small group can do more than it thinks if it gets together.

Stacey Caillier:
Yeah. Great. Are there other things that, when you think about where or why improvement efforts stall, are there other reasons that come up and kind of … if you want to offer some antidotes or things you’ve seen that have helped with typical stalls?

Don Berwick:
Yeah. Again, I know much more about healthcare than education. One is a crisis can stall improvement. The headline, suddenly there’s a scandal and everybody rushes to that place, and it sort of proves that the constancy is not there. Antidote to that one to me is the leader says “Yeah, we’re going to deal with that, but everybody, let’s remind ourselves: we’re here to do this. This is what we’re working on, and we’ll solve this problem, but that did not become the key thing.”

Or even better, “I’ll take care of it. I’m going to open umbrella. You keep your work going. That’s rain. I’ll take care of the rain.” But don’t get seduced by crises. Another is change of leadership. This is leadership-dependent, so when a chief executive leaves or a principal leaves and the new one doesn’t understand this, doesn’t have this theory, then you can rapidly run aground.

I’ve seen that happen … entire hospitals where improvement has run aground because the new CEO isn’t interested or doesn’t know it. The prevalence of outmoded and destructive habits, like for example, pay-for-performance, is a very bad idea for improvement. It creates winners and losers, and when someone comes in and introduces a system that is antithetical to the improvement culture, you can get into trouble really, really fast.

It’s also slow. I mean improvement takes a while. When a kid doesn’t learn to tie their shoelaces on the first go, and so you have to have a kind of patience with this, and I’ve seen a lot of places become impatient too soon, not counting on … that the investment now in getting it right all the time will pay off later in liability.

Stacey Caillier:
The issue of leader turnover, I think … I mean, well everything you’re saying, is very relevant to education as well, and that one around leadership turnover is really present with superintendents, principals, and I think a lot of schools end up feeling stalled because they don’t know if they have the support of their leadership, or the leadership just kind of doesn’t know about the effort or the improvement happening.

I’m curious if you have recommendations or thoughts about how do you bring new leaders into the fold?

Don Berwick:
Well, the governance matters, that’s why boards matter; somebody’s hiring leaders, and the hospitals that … I mean hospitals where the board is aware of investment and improvement as a strategy and look for succession that will maintain that, they more likely to succeed than places where the board’s to lunch or not too into this. I guess any good leader has a dose of curiosity, and I suppose if I were in a context where the leadership is changing, I would try to spend some time with the incoming person in a trusting way, and explain we really having great success with this approach, and can we spend some time telling you what this is, and maybe you have questions about it or come and visit.

But it’s very hard. Dr. Deming, W. Edwards Deming, wrote about the deadly diseases. These were the five causes of failure for improvement, and number one, I think, was turnover of leadership.

Stacey Caillier:
So I have some grab-bag questions, if we … One of them is just there’s a lot of interest in investment right now, and improvement networks in education, and I’m wondering just what should we be attending to now to ensure that the work continues in schools when the grant money goes away? Because many of us are working with schools on three or five-year grants, and we really want to build sustainability for this work to continue when we’re gone.

Don Berwick:
Could you have your school simulate the end of the grant a year early? Okay, pretend every time you have to go to the grant, stop and say “Okay, no grant. How are we going to do this?” So you’ve got a chance to think through or crack this resource building and process building that you could then turn to and rewind even before you need to. Why wait for the shoe to drop?

I think also peer-to-peer support. So if you have your schools working with each other, that’s very productive. Can you find a way that that would be maintained in the absence of the grant?

Stacey Caillier:
Love all of them.

Don Berwick:
There’s one other thing to do that is … I mentioned this in NSI context a bit, but it’s waste reduction. It’s lean thinking. Use some of the skills that you’re building now to actually free up time and resources by cutting waste. The form of that that I mentioned several times is stupid rules. I think that NSI should have a stupid rules week right away, and try that in your network.

Have a search for a hundred stupid rules, document them, and stop them, and then you will receive applause from the workforce that will then say, “Ah, I have a little bit of breathing room now. Let’s use it for sustaining effective improvement.”

Stacey Caillier:
I would love to host that.

Don Berwick:
Certainly even at High Tech High there might be some stupid rules [inaudible 00:41:47].

Stacey Caillier:
Oh yeah. There are stupid rules everywhere, and it would be so fun getting rid of them together.

Don Berwick:
Yeah. Remember, we had four kinds of stupid rules. They probably apply, but I’ll tell you what they were in healthcare. One were real rules, laws, regulations that just made our lives miserable. These are rules that stood in the way of proper patient care. Second were administrative rules. That is, within school. Within hospital rules, where there was no regulatory agency. The CEO or the COO could say, “Nope. We’re not going to do that anymore.”

Third were habits. It’s not a rule at all. It’s just “This is the way we do things around here.” But then you’d say, “Why?” And then the fourth were myths, which is “There is a law that says … ” but then you actually look for it and there is no law. There is no regulation. So that was it. Myths, habits, administrative prerogatives, and real rules.

Don Berwick:
Well over I think 85% of those stupid rules were not laws or regulations. They were totally within the span of control of the organization or the workforce itself.

Stacey Caillier:
Mm-hmm (affirmative). Okay. Well, when we host our first one, I hope you’ll come. That would be so –

Don Berwick:
I will. That would be so cool. Yeah.

Stacey Caillier:
Okay, I have two last questions that I know we can get to by 11. So one of the things is I’d love to hear just what have you learned that’s most important to consider when thinking about measures or data for improvement? Because data continues to be such a hangup for us in education in many ways, and I’m just curious what you’ve learned about how to make it most useful and meaningful.

Don Berwick:
That’s really a long list. Okay, so immediacy matters. That is, the greener the data, the closer to my action the data are, the more useful it’ll be to me. Don’t tell me my performance last year. I don’t care. Last year’s gone. Tell me our performance the last hour. The last minute. So proximity to the current time, focusing on what matters so that if it’s a stupid measurement, it’s a stupid rule.

And so people should look at it and say, “Oh yeah, that’s what I meant. That’s what I want to do.” Another is measures with meaning, but here’s the lesson: a measurement is never the truth. Measures are shadows. They’re like images of what I really care about, so never honor the measure. It’s not what we care about. No measurement has ever been what we really care about.

It’s just an attempt to represent what’s really important, and if the measure doesn’t seem important because it isn’t really what you care about, trust that instinct. I think a last point is the voice of the patient, and in your case the voice of the student and family, is the best metric. If I ever had to choose only one thing, it would be turning to a patient and saying, “How’d it go?”

That beats everything else, because they know. In the end, that’s what we’re all about.

Stacey Caillier:
Cool. Okay, last question. I know you’re off for a summer of writing, and I’m curious just to hear if there’s … what’s one idea you’re hoping to get out into the world right now?

Don Berwick:
Well, it’s a particular thing I’m writing. I’m writing a book about a patient of mine who had a serious disease and we cured it, but whose health and wellbeing depended on a lot more than the medicines we used. I mean this eye-opening understanding of what actually generates full lives and health, robustness and wellbeing, and it’s how to widen the zoom lens. So I’m trying to help people. It’s in the context of their whole lives, not just the pills we happen to have in a bottle.

Stacey Caillier:
Is there anything else you want to say?

Don Berwick:
Thank you for the chance to talk with you. It’s always a pleasure.

Stacey Caillier:
Thank you so much, Don.

I really appreciate it. It’s so much fun.

Don Berwick:
Say hi to my friends there. I look forward to the next time.

Stacey Caillier:
I will for sure.

Don Berwick:
Okay.