Improvement Science Eats Org Charts for Breakfast!

What if quality wasn't just a department but the entire operating strategy of your organization? In this episode, host Rex Wallace sits down with improvement scientist and author Dr. Dave Williams and co-host Ryan Bernstein to explore how healthcare organizations can move beyond project-level process improvement and start thinking in systems.

Dave breaks down the tools and mindset shifts that help leaders understand their organization as a system, why testing before scaling is non-negotiable, and what it really means for leaders to own outcomes rather than delegate them.

Quality as an Organizational Strategy and the QOS Field Guide are available on Amazon. Connect with Dave at davidmwilliamsphd.com or on LinkedIn.

Transcript

SPEAKER_00 0:00

Hey everyone, Rex here. Before we get to the episode where we have a great discussion with the author of this fantastic book, I want to talk quickly about our collaboration with RISE in season one and something that's coming up. So as you probably know, RISE National is having its 20th anniversary this year, March 23rd through the 25th. It's the only Medicare Advantage event that brings every operational discipline together under one roof. Sixteen hundred plus leaders in the room. Direct updates from OIG, DOJ, NCQA, plus an NBA Hall of Famer is keynoting this year. It's at the Orlando World Center Marriott. If Medicare Advantage is part of your job, this is the place to be. Use code BYONDVIP to save 20% on a registration. Sign up at risehealth.org. Now onto the show. Quality improvement in healthcare. It's typically described as a framework for systematically improving care. In reality, it's an endless struggle to align policy, data, and processes with human behavior. Join the RWC team and guests from across the industry as we tackle the toughest challenges facing quality today. From culture to capital, from strategy to technology. Join us as we go beyond quality. Hey everyone, welcome back to Beyond Quality. As you heard in the intro, our objective here is to really have compelling and thought-provoking conversations around the things that drive quality, healthcare quality specifically. So from culture to capital, from strategy to technology, and you know, kind of everything everything around and in between those things. I'm your host, Rex. I lead RWC, a healthcare quality improvement organization. And today we're going to dive into the concept of leveraging quality as the guiding strategy for the entire organization. My co-host today is Ryan Bernstein, senior consultant and Medicare lead at RWC and facilitator of STARS Council, which is a mastermind style community of Medicare Star ratings enthusiasts from all around the country. Ryan earned his drops as a research analyst before diving into STARS Analytics and then STARS Strategy. And our guest today is Dave Williams, an improvement scientist and advisor who co-wrote the book Quality as an Organizational Strategy, which we're going to dive into quite a bit today. So, Ryan and Dave, welcome to Beyond Quality. Thanks for being here, Rex.

SPEAKER_02 2:30

Good to see you.

SPEAKER_00 2:31

Yeah, you too. Um hey, can each of you give a little more of your story so our listeners can learn more about kind of why this topic matters to you? Ryan, can you start?

SPEAKER_01 2:41

Yeah, for sure. So uh Hey everyone, excited to be here today. Um as Rex mentioned, I kind of started my career in the quality space as a data analyst, um supporting hospital research at Boston Medical Center. From there, um after earning uh my master's in biostatistics and epidemiology, moved into quality analytics, which is a space, as we all know, um sucks you into a deep dark hole that you can't get out of. Uh, but it's a hole that we love. So um kind of from there expanded my scope into strategy. As Rex mentioned, worked at a health plan as a Medicare Star Ratings leader for five years before joining the team at RWC in July of 2023. Since then, have just really enjoyed diving into many of your organizations to help promote uh quality strategy and engagement with our members um that we serve in Medicare Advantage.

SPEAKER_00 3:41

Great. Thanks, Ryan. And how about you, Dave?

SPEAKER_02 3:44

Great, thanks, Rex. Well, hi, my name is Dave Williams, and uh I'm an improvement advisor. I'm based in Austin, Texas. Uh, I got my start in quality um 26 years ago uh when I became the chief quality officer for uh major metropolitan ambulance service. Um, and I had studied uh quality improvement in my graduate work. I ended up uh working initially in the ambulance space, and then I uh moved over into healthcare when I started working as an improvement advisor supporting uh programs and projects for the Institute for Healthcare Improvement. Um I spent uh five years on the leadership team there as the senior lead of improvement science and methods and leadership for all the global programming, uh, and now I'm an independent improvement advisor. Um, most of my work today, as uh Rex mentioned, is is centered around kind of moving from uh taking process improvement at the project-based level and starting to think about how do you um apply some of the concepts of the science of improvement and uh a series of activities across the entire organization so that you can leverage the same um ideas and and theories that go into uh process-based improvement to organization-wide improvement at the leadership level.

unknown 5:04

Thanks.

SPEAKER_02 5:04

I love that.

SPEAKER_00 5:05

Yeah, thanks, Dave. And real quick on background, um you know, I I got introduced to Dave by taking uh a Harvard X online course around improvement science and um was yeah, impressed with the theories and the way he the way he practiced and and taught on there with with some other people like Don Berwick. It was a a great um a great a great course. And um yeah, I just got really enamored with the you know the more I guess rigorous thought of improvement science, right? Something that I think a lot of us in quality improvement um maybe don't leverage enough, right? So really excited to dig in today. Awesome.

SPEAKER_01 5:45

Yeah, you know, one of the most intriguing ideas I think in um your book and kind of the overall philosophy, Dave, is this idea of organizations of systems, which we've referred to a a little bit up to this point, and that those systems are responsible for processes. You talk about kind of an organization's approach to process improvement, but what I think we want to talk about today is uh how to shift to system improvement. Within that, you know, I think about organizations of as layers of systems potentially, and that each of those systems are responsible for their individual processes. I think one of the clear concepts that is clear in what in what you talked through is that the the benefit of zooming out uh from those individual processes and systems to seeing the bigger picture is what most uh enterprises are missing. So can you talk about that balance between um organizational level system versus those more granular things that we're more used to talking about in process and quality improvement? Sure.

SPEAKER_02 7:01

Well, I mean, as you already mentioned, you know, one of the terms is process improvement, right? So the an underlying theory of improvement is that uh most of our work can be described as a process, and part of our job is trying to figure out well, how do we create a reliable process that gets the outcome that we want, right? And and in most improvement work, that's that's what we would call project-based improvement. There's a particular um metric or outcome that you're trying to achieve, there's a series of steps or things that come together in order to achieve that outcome. And so a project uh if that um if the process that you have isn't getting the results that you want, project is created to work on that. Um, and that is the heart and soul of improvement, quality improvement. Um, as you mentioned, you know, if you think about any particular process in an area of an organization, uh they they don't operate in isolation, right? So so in any space in you know any department or any um uh unit in a hospital or department in a in a health plan, or um, you know, even if you go to your favorite uh you know uh fast food restaurant, all of them have multiple different processes that are working uh together. And what you're describing um is that there's a recognition that um not only do I have to have each individual process working, but they gotta work in synergy. They gotta uh work with the ones that they're linked to and interacting in. And uh sometimes people think about this uh as what's called microsystems, right? So you start to look at the space around where things get done and then start to ask like what are all the various people and and uh equipment and processes that are involved in there, and how do I work on that? And uh that is like the the you know the the layer up from thinking about just one process, right? So there's the process, then you you go to something that um is that that uh micro system or that that frontline um set of systems of processes, um, and then you can start to pan out from there. And so it's it's almost like uh the best way that uh I can describe it, the way we think about it in the book is it's like if you think about Google Maps, and when you you know when you pinch in or pinch out, right? Um so you could start, and you've probably seen this online where somebody starts with a real macro level view of the world and they can see all the continents, right? And then you can pinch in and you can come to the United States, and then you can pinch in and you can see the 50 states, and you can pinch in and you can look at, you know, where I am in Texas, and then you can pinch in and come to Austin, and then you can pinch in and you're in my neighborhood, right? So systems operate in the same way. You know, the process is like looking at the street level on the map, um, and then each layer up give take pans you out and helps you to see where that particular element is in um relationship to. Uh and so jet, you know, in an organization you pan out for a process to the processes that are in an area. Uh, when you get to an area or a department, you might start looking at how they're connected to other parts of the hospital or other departments within the system, and then eventually you get up to the whole organization.

SPEAKER_01 10:07

I love that uh kind of this concept of zooming in and out as needed. Um, you know, I think about some of the pitfalls when we don't do that sort of thing. A lot a lot of the work that we support with our um base of clients is around this gathering of cross-functional teams, entities to accomplish one big thing, right? And a lot of times those teams work in silos where you know an improvement initiative is happening to um you know uh better an outcome on this side of the house, while there's also this um element that could improve outcomes over here. You know, I think about a concrete example controlling blood pressure on the medical side versus adherence to medications on the pharmacy side, right? Like a lot of times those two things are happening in silo of each other, but if you were to zoom out and see how those processes interlink and ultimately how they synergistically can improve outcomes, um that's how I think about what you're describing. Is that right?

SPEAKER_02 11:18

It is. I mean, one thing to appreciate is that any system, um, not all the parts aren't equal. Um and and there actually may be some some fluctuation about uh like which parts um get more attention or less attention at various points, depending on what you're trying to do, right? Um so one of the problems when we break things down into all their parts and then we try to treat them all as equal and we try to get them all perfect, um, often we do that in isolation. We don't think about how they work together, right? And so I can create a perfect process here, or my colleague in pharmacy can create a perfect process over there. Um, and then you know maybe there's a third person that's processing claims that gets a you know a process over there. They're all related, but they never thought about each other and and how they're connected. And so now you have three processes that are perfect in isolation, but not as a system of things that have to work together. And so sometimes you you have to understand like the full sort of uh uh flow of the various processes and how they're gonna fit together, and then ask, you know, well, how what does one need to do to support the other and how do they hand off to each other? And and that has to, you know, that's more than just you know us communicating with each other in our silos and then saying, you know, here's here's how I think, here's how you think, now we better understand. It's actually defining and saying, well, if you do this, this is what needs to be on the outside so that I can take it and do my part and and build upon it. Um and how do we how do we work um uh together as this this synergy um where things um you know are are um at their proper potential uh for for what's needed at the time. Totally.

SPEAKER_00 12:56

Uh are there certain tools or you know, when you're when you're helping an organization understand those interdependencies, um to help to help an organization because they're used to looking at like we're like Ryan is saying, and this is like you've got some good visuals in the book, full you know, going from the framework of a an org chart with silos to to this system with these interdependencies and in microsystems. Um how do you help organizations appreciate and really understand those interdependencies that you're talking about? Are there other visuals? Are there certain tools that you that you leverage in those cases?

SPEAKER_02 13:34

Yeah, so you you um described already, you know, when you uh go to an organization and say, Well, you know, is there what what would you use to show me how your organization um is uh created or or works? Very commonly somebody will present you with uh an organizational chart. They may even have it on their website. And an organizational chart is uh is a visual display of the accountability of roles within an organization. It doesn't tell you anything about how work gets done or who does what or how much of what. It's really who's over certain elements of the organization. And those may be service lines, they may be departments, they may be full activities, right? So that's that's not as useful. Um, going back when we were talking about work, you know, all of our work is uh can be uh described as a process. So obviously, one of the elements that's really helpful for us to have a conversation about how things get done is to use something simple like a good old uh you know process flow diagram that helps us to walk through like what's what are the steps of the process, whether that's you know the imagined steps of what we think it is, or actually following claims through a process or or a patient through the pharmacy to see like what it's like in in practice. Um, but that helps us to understand like the way that work is. Now you mentioned um you know uh a bigger picture view, which is to think about like a system map. Um and that's where we we used to call it a linkage of processes. That's where you take all the different processes in a in a space and then you show their interdependence. Like you connect the dots and say, like this one, when it when this process ends, it it moves over to this one, and you could show the relationship and how they're linked together and how they move together. Um and that's one of the tools that we um bring when we're talking to organizations is how do you understand all of the processes that are in a particular space or in the organization as a whole and how they're linked together. And then the second thing that's tied into understanding the system um is uh having a set of measures, a vector. We call them a vector of measures because it's more than one. Uh usually it's a dozen or more measures that um that are um interrelated that tell us about different things. Maybe they tell us about the key outcomes of the of that area or the organization, maybe they tell us about key things that are tied um to your purpose, uh, maybe it's very specific things that have to happen successfully. Um but we you know we uh display those visually in a time series chart like uh a run chart or a shoe heart statistical process control chart. And together the you know the system map is kind of showing um this is how we work, and the data is showing uh it's like the vital signs of the process in the system that says, and this is what happens. Um and so together those really help us to learn because we can now uh with pro with a process map, we can we can talk about how we think work happens or how we've observed work happens. With it when we you know pan out and and look at multiple processes, we can look at the system map and see how those are all linked together. And then we can bring in our vital signs of the data and look at time series charts together so we can see, oh gosh, well, here's the current um sort of outputs or outcomes, I should say, of those processes and how they work together. And when I make a change in the system, here's where I I can see whether it's actually moving in the direction of goodness or not. Um and I can also uh be able to determine like you know where in the system I need to work, like which stuff is um achieving the goals that I want, which is not. Is it a stable process that is built that way and I just don't like it and I want I need to change it to improve it? Or is it an unstable one? It's one that actually is uh is uh chaotic and isn't predictable enough that I can even worry about improving it yet. I just got to stabilize it first and get it to some degree of order before I can worry about whether I like it or not. And so the uh those tools, the flow diagram in in uh and then panning out to the system map and using uh a vector of measures of time series data that's tied to that stuff as a feedback mechanism is really helpful.

SPEAKER_01 17:51

I love the way you talk about those signals, right? Like the things that tell us when things are going wrong. And presumably things could go wrong with the system, right? Maybe upstream kind of structurally, and then things could go wrong with a process. Um when I think about improvement, um sometimes people will try to tackle the system itself, and sometimes people will drill down micro and try to tackle the the individual processes. We talked about that a little bit. Um I think one big faux pas that organizations make when we when they talk about you know the zoom out and and look at the the system as a whole is that they might translate that to the need of large efforts of improvement, right? Like a an entire organizational shift or you know, a huge project that costs a big bucket of money, right? And I think um one of the points that you make is that improvement um we should think about it as a muscle to get trained, um both in identifying the areas that need improvement that we were talking about, either systems or processes, but also in the actual effort of improvement. When I think about building a muscle, I'm not gonna start by bench pressing 500 pounds, right? I'm not gonna start with the organizational overhaul. I'm gonna start with maybe a pair of light dumbbells, right? And so how can you help us think about you know these huge improvement efforts within the context of a system and how to get there by building the small improvement muscle?

SPEAKER_02 19:45

Yeah. Well, one thing to think about in in organizations, leaders um, when they're uncomfortable with um the current performance of something often look to change uh to get a different result. Right? So there's there's two challenges that you can run into, right? One is that you assume that the way things are working are just fine, and uh my attention should go on just sort of pushing harder on people or maybe using incentives, or maybe it's an educational problem. You know, so so there I make an assumption that the system I have actually is good enough, and and there's something else that's wrong. There's some misalignment. And that's that's really Is that ever the case? Um actually a lot of times uh when you the system that you have it, well, it's perfectly designed to get the results that you get, right? So that's an old saying that's been used around healthcare for a long time, uh, popularized by uh Paul Vitalden at Dartmouth, um, that actually came from from uh uh a book. Um, but but it's true. So so um any system that you look at is built to get the the results that it gets. And one of the first questions, you know, like you said uh we were talking about before is um do I have something that is stable and predictable or do I not? And that should be the first question that uh anybody should ask is um, you know, because if I have a system that is predictable, like it'll I don't like the results, but if I don't if I don't do anything, I know it'll be this way next month or next week, um, then that's useful because that helps me know that if I want a different result, I've got to change the way we work in some fashion, right? If it's unpredictable, if the data in every time I look at it is going in different directions or has weird patterns or I don't I can't make a guess as to what's gonna happen next time, then I gotta figure out how to get it uh stable. And I and often that is means we've got to go back to the fundamentals. We've got to figure out like how do we create a shared way in which we we work together on a particular process. And it doesn't have to be perfect, it just has to be similar. Um the way I often describe it is um it's like cruise control on your car. If you drive, I don't know if you use cruise control, but this is a great way to learn about variation is to get on the highway, not necessarily in rush hour, go in a time where traffic's moving. Somewhat and put yourself in the in the fast lane and then set your car for a set speed. Do something like uh so let's say it's you know 65 miles an hour, maybe go 68 and set your cruise control there. And then watch what happens. Um and it's rather fascinating. If you've ever done a road trip like this, you've probably had this experience. Um but all of a sudden you'll see that cars will pass you. Um, and then if you do nothing in a minute or two, um they'll start falling back and you'll pass them. And then without you, I mean you're a constant. You've established yourself as a constant, but all of a sudden you'll watch people pass you, and then you'll pass them, and you know, there'll be all this uh, or somebody will whip by you, and then uh five minutes later you'll pass them. Um and what ends up happening is you by creating yourself as a stable constant, um, you're now able to see all the variation around you. Whereas when you were driving like everybody else, you were constantly adjusting your speed in relation to the other car. So you didn't notice all the variation that existed. But when you stabilized yourself, when you said, we're gonna, you know, you basically create standard work or follow a standard process, that enabled you to see all the noise that wasn't obvious to you before because you became constant. So in in in the case of process improvement, that's one of the first things I always ask is um can we, you know, what's your process? And if you have one, let's try to use it. There's a good chance you're not using it. So it's really actually interesting to try your process as it's written and see what happens. And that actually stabilizes work. It takes away some of the noise, and it you'll almost instantly identify that there are things that need to be changed, right? And then if the system that you have and the process that you have is not is not working where you want it, then that's a great place to start thinking about how do I change the system I have to get to a different result, right? Um and that and so that that's tied to that idea of stable, you know, getting to a different result. Um now the hard part is a lot of times people decide to just do big switches. So they don't know under how the system works that they have, they don't like it, and so they're gonna switch and do a totally different thing because they just hope something better will happen.

SPEAKER_01 24:19

Right. Totally. And we see that you know time and time again. I think one of the things that um keeps organizations from making those small changes or from following a process is that they don't have the evaluation structure in place to know whether or not something is working or whether or not that thing is the thing that is is or is not creating the results that they want. So um as we talk about implementing these process or following along a prop process trajectory, how do you recommend people thinking about the evaluation of that process? Is it within the framework of those vectors that you talked about before?

SPEAKER_02 25:07

So vector tends to be a little bit higher level, but um the measures that go into the vector come out of things that matter and it could be a process. And so when you're when you're developing a process, uh all the way along the line, there are um things that you might um decide are important to that process that need to happen every time in order for the process to be successful, right? We we'll call those quality characteristics. So even though there might be multiple steps, there may be key things. Uh data has to be uh captured, uh has to be entered in a certain way. Um maybe you need to get an approval before you um uh you know uh uh refer somebody to do an appointment. You know, there's there's all kinds of things like that that are key in order for the process to be successful. So one challenge is that frequently people create a process without actually testing to see whether that process works and gets the result you want. So part of process improvement is you you're working to design a process that gets the outcomes that you want. And and to do the uh a project, or whether it's designing it or or or redesigning one that you have to get a better result, you need to both have a sense of what the work is and what are the key things that have to happen. And then you're gonna develop specific measures that help you to understand whether those things are happening. So you know, in healthcare, that a lot of times those are process, we call them process measures, right? They're measures of specific things in the process that have to happen in order for you to have the outcome of the process or the outcome measure work. Um and so uh a fully matured process is one that not only is a process that you know achieves the outcome that you want, but also is accompanied by measures that are showing that it reliably does that. Um and so if you look at a process and somebody says, I don't know how it works, or I'm not getting the outcome I want, but I don't know whether it's working or not, they don't have a fully matured process. They may have a protocol that they've written down, they may have something that they trained on or that they, you know, they laminated and put on the wall and they say, I want you to do this, but they may or may not know whether it actually works or not. And so many times they're they're working on trying to assure compliance to it, um, and they're getting frustrated and they're seeing that as a people problem, and they have no idea actually whether it works, right?

SPEAKER_01 27:24

Totally. I think that's a problem we run into time and time again is like we think that this thing is gonna work, right? Like we might even have evidence that the thing is gonna work, but we don't know for sure if it works with our population or if it works in the exact idiosyncratic way that we're accomplishing it, right? So um in that evaluation step of process link to outcome, what steps might you take to ensure every step of the way that we're we're uh getting to that mature process stage?

SPEAKER_02 28:01

So usually I don't install a process permanently until I've tested it.

SPEAKER_01 28:06

Right?

SPEAKER_02 28:07

So I want to know that it works. Um because if I install it and it doesn't work, I'll feel the pain. So I'll give you a good example. This is a non-healthcare example, but it's a beautiful one right now that sort of illustrates this. Um you may have followed, um uh I'm sure you all have flown Southwest Airlines. Uh I believe it was about a month ago, Southwest eliminated its uh its long-standing process for boarding planes where you used to kind of you know run for your seat and no-assign seats, and you know, you'd line up and do all that stuff. They've been doing that for decades, right? Um now Southwest had the fastest turnaround time in the industry um uh early on. That was one of the things that helped them grow and they were amazing at it. But yeah, in at long reasons why, but uh in today's market and in their growth, they learned that they they uh needed to change and uh they you know uh decided that they were gonna get into this boarding process. Now, if you've been following the news, um they are having a lot of trouble with this boarding process. And it's weird because you're like, well, everybody does it this way, so why would it be so hard for Southwest, who are really good at process, to make this switch? And part of it is that they um uh I'm I I know that they must have tested it to some degree, but when they when they uh moved from their small testing that seemed to work to full scale, they're discovering all kinds of things that they didn't pick up on initially. Um things about like where the crew uh keeps their bags. Um how, you know, uh they they uh you know they also turned off free bags, so that means uh fewer people are checking bags and they're actually bringing bags on board, which means there's more bags than they had before. So they changed two significant things at the same time without uh really looking at what that meant together, right? Um they also uh price-tiered their seats. So whereas I would run for the nearest aisle seat, um, and uh, or maybe I'd I'd look for the exit row and then run to the nearest aisle seat, and that was my method, and other people would go to the front or the back, you know, they had all kinds of theories. Now it's price locked. So that means that a good chunk of us who are cheap go to the back of the plane, right? I was my I was on a plane uh the other day and there was nobody in the exit rows. They had to move people into the exit rows because nobody bought them because they were expensive. Whereas they used to be the prime seats that filled first with all the business travelers, right? So so anyway, so this is a good example though, where they made a change. It probably seemed like very common sense, and I know they tested it because that's just Southwest Way. Um, but they are grappling with a whole host of system issues because they they they went to full scale without fully appreciating all the ins and outs of what was going on. And most of these could have been predictable, and they're gonna suffer some consequences of inefficiency and and uh you know customer friction and things like that, and bad media for a while until they figured out I'm sure they will, but it's a good example of where something that seems like it's so common sense, everybody else does it, I should just be able to flip a switch, it shouldn't be that hard, actually creates some significant um issues because you didn't test it enough to figure out um what are the issues to make sure it worked.

SPEAKER_00 31:22

Yeah. Um so yeah, you gotta you gotta test it. You gotta and I think in in our world, right, so many organizations focus on these things that are measures, but not necessarily measuring and kind of where you're where you're going with some of this, Dave. And one and one more point, one more point I want to make in. I don't want our listeners to to to miss this key point in you and we're we're talking all about it, but um really succinctly from the book you say, or it's said in the book, you you described um certain distinct ways to make improvements, right? So I think there are five in the book, but the two that really jumped out at me are you know, number one, uh to design something new, or number two, to redesign something that exists, right? And there are some other things too, but uh the point here is and you you make this point in the book, all improvement requires a change. So the system has to be disrupted, right? Would you would you say that I I think in you started off the conversation by saying something like a lot of people think that their system is good, they just gotta you know motivate their people more or give some incentives to the patients or to the doctors or or whatever, but they're not really making a significant change to the actual system or the process. But would you say every improvement uh requires a disruption to the system?

SPEAKER_02 32:46

There's there's some um level of disruption. So when you hear the word disruption, sometimes people think it's very um significant or chaotic. That may not be the case, but um but I I like what you described, which is that um in order to get a different result, we have to do something different. And there's often two things in there. There there are there's more than these two, but these are kind of the two common ones. And one is we don't have a method or a way of doing what we need to do in order to get to the outcome, so we've got to create something, we've got to design it. And that's that's starting from scratch. That's trying to say um uh I'm gonna do a whole new process. So a lot of people are doing that now, for example, with trying to figure out where AI fits into their world, right? It's new, they've got to figure out how to design it or pull it in. Um the other is the most common, which is more redesign. So I'm looking at uh um a process and I have something, it's supposed to work, um, but when I uh look at it, either it's inconsistent or there's certain populations that it's inconsistent in. And so I need to figure that out and uh and then uh figure out what do I need to do to redesign that process so that it works um in a better way.

SPEAKER_01 34:05

I think um what we're talking a little about is innovation versus new ideas, right? And um I think um within that context as well, like the types of changes that we make really matter. Um I wanted to talk a little bit about another point in your book around um the concept of planning when we make change or disruption, we have to plan it first, right? I've got a five-year-old, um and he um likes to build things, but he doesn't go in with a plan. So he just starts building and uh sees what he gets at the end and then tries to say that it was his vision all along. We talked about setting a vision before we start. Um we tried building a treehouse this weekend, didn't it didn't go great. Um so I want to talk about this planning concept, and there's two types of planning that you talk about in your book. One is planning to operate and one is planning to improve. Can you can you talk about how vision works into that? What steps are required for each of those types of planning and what type of results you can expect um after both of those uh types of planning are implemented?

SPEAKER_02 35:37

Okay. Yeah, so so one thing to think about, a key um component or key activity for leaders when they're thinking about quality as a strategy is like what what is your purpose and and tied into that as the vision and the mission and uh the tenants or the practical values that you use as the way that you work, right? So so a vision is thinking far out into the future and saying, where do we aspire to be? What's the uh where do we um want um our organization to finally achieve? And and a purpose or a mission is typically a little bit more specific and clear where it's it it explains like what's the need in the world that you um uh fulfill with your services and and how do we uh how do we build a system, right? Going back to talking about systems and system maps and vectors and measures, how do we build a system that gets that, that um serves that purpose, right? Meets the need of people that have it. So, you know, it could be, for example, um uh I I'm an insurance provider and my I provide insurance to um uh uh uh individuals um and they have a need to be able to seamlessly get health care, whether that's preventatively or it's in an emergent uh space. And so everything that I do has to be built in order to enable that. And then I also may have some some risk elements that I'm paying attention to, right? So so um that is you know helps us kind of to be clear about the direction we're going in, and then that we build the system to get there. Now, one of the things that uh uh happens there is that we um there's a good chance we're not gonna build the perfect system right out of the gate, right? Even if we started our organization from scratch, um we're gonna start with our best guess and our best theory and maybe you know build on the shoulders of um uh others and best practice and kind of have it laid there. But but if we want to uh truly kind of be an organization that's better at getting better, we've got to figure out each year how do we get better and move our organization closer and closer towards uh making our services uh uh uh better matches for that need that uh our customers have or our clients have. Right? So that's where planning comes in. And so um one of the elements of uh quality uh as an organizational strategy is is to on a regular basis, and for most organizations that might be every year, um, there's some kind of uh a formal process in which you take in a bunch of information about um your organization and your research your market research and um what else is going on in the world, and you pull that together in an organized way and analyze that information and bring it to all the stakeholders that are going to be involved in a planning process, and you use that as a way to look at what are the strategic priorities or the strategic objectives that we need to try to approach this year. Almost every organization has a list of strategic priorities that they've created through some kind of process, right? Um but in planning, um what I want to do is one, I want to take this information in and ask, are these the right things, right? Are the are these the things that we truly, you know, all our insights and data and information tell us this is what we should be working on? Um and then the second thing is I need to then go back and ask, well, if I want to um keep making progress on these, if I want to keep chipping away and making a difference here, where do we need to work? Like where do we need to uh uh either change things in the organization or do more of what we're doing today in the organization, right? And that's where uh improvement in operating comes in. So so part uh we we actually have a pretty detailed um uh process that we share with people about how to how to do planning. And one of them is to take those strategic uh priorities or objectives and actually go back to your system and say, well, where would I have to um uh make improvement in order to uh uh you know uh make uh move in the direction of these priorities? And if I made improvement on this process or this area or these processes together, what measures are gonna move, what dots are gonna move in the direction of goodness on my vector of measures as a result of that? Right? And so I'm map, I'm actually uh relating the things that I want to do to where I need to do it and what outcomes that are gonna move if I do them, right? So that that's part of the planning to operate piece, is identifying what needs to get done. And then though if if those are things that are, you know, going back to what Rex was saying, either something that I need to create because I don't have it, I need a new process or a new service, then I got to design that. If it and that's that's a style of improvement. If it's something that I have but it's not performing to the level that I want, and I've got to change that system to get a different result, I've got to redesign it, that's a you know, that's an improvement project, right? So those are things that fall in the in the improvement side. The other side is a part that often gets forgotten. It's part of typical business processes, but sometimes you identify that you you need to do stuff, um, but it's not a change as much as you just need to resource something, right? So so maybe you you evaluated um a uh a program where you um use text messaging uh to remind people to go to their employment. And you discovered this is awesome, it works really well, um, but we need uh we've only been using it in diabetic patients, and now we also want to use it in cardiac patients and uh patients with kidney disease, and so we need to expand it. So so we're not we're not creating something new and we're not changing the way that we work, but we're saying that we need to we need to invest more in something that we've decided we work. That's a plan to operate, right? That's a budget decision. It's not really a change of the system, it's just saying we want to fund more. Or let's say that you decide that uh you know, stat you educated staff on something and it was really powerful, it worked really well, you you saw a good impact on it, and now you need to um either refresh on that training or you need to expand that training to other people. Again, it's a that's a budgeting resourcing type of question, that's an operating question. So in the operating side, you're basically looking at what you're doing and saying, well, how do I budget and finance, you know, uh create uh the resources for the stuff that worked? And if there's stuff that we need to uh infuse more um into in order to continue to make it work or to expand its potential, um then that's a plan to operate. It goes into other business processes, but it's different than I actually have to create something new or I have to change the way that we work. And so those they come out of the same process, the same planning process will help us go improvement stuff, operating stuff. We got to do something about this. Um but but they're they're different mindsets in terms of of how we have to act activate those as leaders and managers.

SPEAKER_00 42:38

And uh related to that planning, um you know, when you're planning a new whatever intervention, there's you know, look, we we subscribe to the belief, I definitely do, that uh quality improvement is data driven, but it's human dependent, right? Um largely. So when we're building something new, it's dependent upon the humans. So something you bring out in the book is you say you say um you know one common mistake is to think of change as only a technical issue. There is no such thing as a technical change without a social effect. And I think Durand said this, right? So can you unpack that and and help listeners understand what social effect means and how they plan for that?

SPEAKER_02 43:23

Yeah. So that that quote actually comes from my uh Marvin Weisboard, who wrote a great book called Productive Workplaces. Um and uh so so the idea um there's an old model um called sociotechnical uh systems, right? And and uh uh Marvin was uh big in um uh organization development and human behavior. And so um sometimes, especially if you think of like manufacturing or if you think about software, we create technical processor systems that operate in a certain way, right? Or that should operate that way. Um so you know, Southwest, I'm sure, modeled that. Process and did it somewhere in Dallas. They probably have some facility where they you know got people together to volunteer to pretend to get on planes and things like that. You know, there's a technical element, and probably on paper it's fantastic. And then there's a human element of which, as you mentioned, Rex, I like to say, you know, especially in human systems, like uh improvement would be easy if it weren't for all the people involved. Right. Because it's easy to write steps and say this is the way something should work and create measures. Um, but it's a lot harder when you have different people that come from different backgrounds, have different experiences, um, have different motivations, may have uh uh different limitations and so um uh or or ways in which they they operate, different uh preferences and all kinds of good differences that exist in humans, right? So so that that's an important element is that any time that you're creating something, you have to think about um how do I learn about the different people that are gonna be involved in it? And how do I make sure that I build this so that it I often say it works for most people most of the time. You can't build something that's perfect for everybody, and it'd probably be not helpful for most people if you if you got it so that it it was 100% error-proof. Um, but you know, there are ways in which you can look at the population that's gonna be involved in your process and try to figure out how do I make it work most of the time for most most of the people, or or you might decide that 80% is effective, uh, you know, because going to 95% uh effective would be painful, right? So 80% is good enough and keeps it loose enough that it works. Um and I think that that's a key component is is appreciating um how something works, who's the you know, who's gonna be involved in it, and really bringing in that human element to appreciate that you want people to be able to work in this process and they're not all the same.

SPEAKER_01 45:57

I want to go back to something we talked about earlier in the discussion about um testing. Right? You you said you're never gonna install a process unless you know that it works. I think part of that what you're describing is learning, right? Learning whether or not something meets the need that we've identified within the organization. In your book, you talk about um learning in rapid cycles. So iterative learning, iterative process improvement, building that muscle upon each other, right? Um and I wanna kind of get really practical there and just understand from your perspective, right? Like what how should we think about rapidity, right? We've all been on projects that feel like they take a hundred years um to try to get something done, and whether that's the scope or you know something else that we need to address regardless. I think um in order to make a significant change, it's all about that additive and incremental effect. So how can you how would you suggest we approach rapid learning in the context of a specific timeline? Like how how long should these cycles take?

SPEAKER_02 47:26

Well, so uh it depends. That's one of my favorite improvement uh terms um from the science, is is it it's really relative to what you're doing. Right. So so most improvement projects, so let me separate um an improvement project, um, I try, and this varies again, depending on the situation, but I try to scale and plan an improvement project to take somewhere in the vicinity of four to six months. Um and and uh I uh sometimes that is a a constraint in which people uh feel is unreasonable. Um, but typically uh the my argument is um that if I if I do it in four to six months, you've got to then commit resources to it and spend time on it and and actually carve out the ability for people to do this as part of their work. If if if I don't put a deadline or if I allow it to be endless, or I say we want to get this done by a year, then often what ends up happening is it becomes a side project outside of people's daily work, of which it's always in competition. Um so so the uh one of the primary reasons to pick something that requires uh like a certain degree of daily or weekly effort, um, and then um uh planning a project and scaling it to a defined time frame is then as a leader you've got to resource people to say, I want you to work, you know, uh X percent of your time on this and get it done. I want to get results in a period of time, right? So that that's uh scaling a project. Now, in terms of changes, um within that project, there's gonna be a lot of different things that we try. Sometimes that's gathering information, sometimes it's changing a way that we do something. And in doing that, I teach people how to use the um the plan to study act cycle, which has been around for like a hundred years. You know, Walter Schuhart brought it to us um he he uh worked at uh um in uh ATT or one of the iterations of ATT when it was Bell. And he basically was trying to you know boil down the scientific method to something that anybody could do at the the point of where a process is and do iterative um cycles. Um and and you see this cycle thing come up all over the place. So sometimes people will call them learning cycles, and there's actually something called a cold learning cycle, which is the how the cycle of learning happens, you know, and that's you know, and there's different cycles that that exist. But um in the way that we teach improvement, um we we teach uh trying to figure out what's the smallest scale that you can take a change in order for you to try it and learn quickly. So I I I tend to shy away from saying um rapid cycles because people think they all have to be fast. It really does, you know, again, it depends. Sometimes I can do something um on every um every claim. And so I could do 10 cycles in a day because I might be able to get through 10 claims in a day and try out something different on each one, and each one is building on the other. Or there might be something that's gonna take me two weeks, because it takes me two weeks to get enough um of whatever I'm trying to change uh to see whether it has an effect or not. So the scale of those cycles change. Um, but what's important, a key component of a PDSA cycle is at the front end, the plan part, is that you make a plan. You say, I think it's gonna take X amount of time. Here are the learning questions I'm trying to figure out, and here are the predictions that I have. Um and then when that time is up, or you know, or when you've completed the cycle, you come back and you study and you compare the difference, right? And you go back and you say, well, here's what I thought the questions I was trying to answer, and here are the answers. Here are the predictions I made, and here's what actually happened. And that reconciliation is where almost all the learning happens, right? Um, and so one of the failures that you know uh that results in the experience that you've had, and many people have, is that um there one, people aren't testing enough, right? So you you often find that you say, Well, tell me what you've tried. And people are like, Oh, well, we met last week and then we decided we'd meet again, and we were gonna talk about this, and we needed more data. And I was like, So you haven't tried anything, right? You haven't tried to change anything. You haven't you haven't tried to like go out and touch stuff and and do it. And so therefore, no learning has occurred. You've you've uh you've been you know pontificating or you know having some some interesting thought exercises, but nothing has you know, and the best way to learn is to try to change something. I mean, just try that in your life, as you said, trying to get your son to plant out the the treehouse, right? So like getting them to do something requires effort, and then you learn things, right? So so uh looking at people's uh PDSA cycles and how many they've done is one thing. The second thing I look at is what have they tried? And typically what I'll find is that people have been trying the same change over and over again in almost exactly the same way with slight modifications. Um and sometimes they'll call it different things. They'll say, you know, uh, I gave feedback to somebody, I did a training class, I did uh I uh uh sent them a PDF uh describing an article about this problem. All three of these are the theory under them is that somebody needed information. There were just three different ways to deliver information, of which ne none of the three resulted in an outcome change. Um, uh uh, you know, if that's one of the things that's supposed to happen at periodical uh periodic points is to go, well, we've tried we we had a theory that people needed information. We tried three different ways to give them information and the outcome has not changed. So maybe information's not the problem, right? And maybe we should move on or do something else, right? So so that often that the pausing and building in routines to learn from your data and learn from your PDSAs and doing that at regular intervals really helps you to generate learning, make sure you're making progress, um, and help you to stop things that aren't working and add things maybe that are new, that are different, that do work. And many times that's when a breakthrough happens, is when somebody's been working on the same ideas, and then they say, Oh, but we haven't tried any of these other ideas, um, or we haven't worked on this other part of the process, or we haven't done this other thing. And then they do that and they're like, oh, wait a second. Um, and and some new insight occurs, or there's a breakthrough.

SPEAKER_00 53:54

Well um a couple of final questions as we prepare to wrap here. Uh first Dave, curious, what's what's one belief about quality strategy that most people disagree with you on?

SPEAKER_02 54:06

Um Well, I think the biggest thing that that uh well there's a couple things. One is that um that you have to make stuff easy in order for anybody to do it. Um I you know there's a there's a natural sort of tendency in and maybe this isn't the greatest answer, but uh in our world there's there's a desire for the easy button, right? I think it was Office Max or Office Depot who came up with the easy button. Um and and we have TED Talks, and uh, you know, people love TED Talks and like the short compactness. Everybody now is asking for Chat GPT to summarize stuff for them so they don't have to read, right? So um there's an assumption that um the only way things uh will actually happen is if we make them easy and digestible and simplified. Um and there's some truth to that in terms of the fact that I think we need to figure out ways in which we can reduce some of the friction and help people to do the right work in the right way. That said, changing organizations, leading and uh changing organizations to get results is not easy. Um it's and and uh actually very few people know how to do it really, really well. Um and what I what I always tell people is that when you start really getting people into the mode of being able to deeply learn about the work that they have and figure out how do I change it, and I can see the impact, whether that's it makes my life easier or it gets better results for the goals that I'm trying to achieve, or I see that the people I serve, my clients and my patients, are having a better experience. That like is like a game changer. That is super motivating, engagement is super high, you feel proud about the work that you're doing, and you want to do more of it, and you're willing to do really, really hard work. And I think right now, a good example, you know, we've got the Winter Olympics. You know, none of these people that are you know uh uh uh skiing for medals or skating had an easy button. All of them had to work in very disciplined ways, and I think very few of them would tell you that they would ever do anything differently because of the fact that they uh through their work and their dedication and their focus, they are deep experts in their field. Um, they see the results of their their um uh progress, and when they when they really are successful, it excites them and motivates them. Um and and actually it it probably less to do with the competition of the others, it's more of the competition with themselves and getting better at getting better.

SPEAKER_00 56:39

Right. Yeah, such a great example uh with the with the Winter Olympics right now. Um if you can leave healthcare leaders with one mindset shift about quality that would immediately improve outcomes, what would it be?

SPEAKER_02 56:52

Well, the biggest thing is to go back to the system and ask Do I do I have a process and a system of processes that that is built to get the results that I want? Um and most people move to some some other theory before they've confirmed that. Um and so I think that that's that's an important piece. And one of the first steps I often do in any any process is say, well, let's look at the data. Let's just look at your measures and see over time, do you have a stable process? Is it built to get the result that you want? And then from there you can make decisions about what to do. Um but uh but I as we talked about before, Rex, I think it I think you want to be careful of immediately choosing uh a different direction without understanding your system at all. I think you want to be um thoughtful about grabbing um the the you know what what's the latest thing that you saw scrolling down your screen on on LinkedIn or the bestseller that came out, or or you know, uh uh I'm cautiously optimistic about AI. I think there's a lot of potential there, but I but uh everything I read also says if we survive it, uh, that it it won't be something that it is immediately changing tomorrow. Like there's there's some a lot of things that have to happen, and you have to bring a lot of expertise to it. So so I think I think that is is an important piece. Is that um and the other thing, and this is a hard one, is that leaders own the organization. They are the ones responsible for setting the vision, um, uh uh you know, clarifying the mission. They're the ones that decide what gets uh what the system um will pursue and what gets funded and not funded. And so they they own the major uh levers that can uh change the system in the direction it it goes. If you're not getting the results that you want, it's a you know you should really look in the mirror and say, well, how do I enable the people that work with me and for me to be able to do their best? Um and it and so if I'm if I'm pointing away from me and I'm doing you know, pointing to other people and saying, I wish my people would be in compliance, or I wish my people would work harder, or I wish my people would, you know, not work from home and come to the office because that's the only way I know how to manage, right? Like uh, there's a lot on leaders that it's really on us as organizational leaders to recognize that we are the ones that that own and can make uh uh decisions about how to change the system, of which our people will be more than happy to dive in and try to figure out how to make their work work really well and be better and get results, and they'll work together. Um, they just need to have a lot of that friction that we've we enable or allow or don't change uh to be reduced.

SPEAKER_00 59:38

Such great insights, 100%. And uh okay, last question how can people find you and follow you?

SPEAKER_02 59:45

Well, the easiest way is uh my website is uh David M is in Michael Williams PhD.com. Um and that is also the same uh way that you find me on something like LinkedIn. Um that's probably the most uh social place uh that I exist, and I always encourage people to reach out and connect. Um know that I might actually ask you a question or something if you try to connect with me, but I I welcome all people uh trying to uh connect in that way. And uh I post a good bit there in terms of things that I write on my uh my blog or things about the book or upcoming uh you know programs that might be offered for the public and things uh there. So that's a that's a good way to find me as well. And then also at the bottom of my website, there's a place if people want to sign up uh for my my once-a-month uh newsletter where I share um uh articles and things that I've found and things related to improvement science and quality as an organizational strategy. And then, of course, you already showed the the book. Uh if you're if you're interested in anything, we're talking about the the quality as organizational strategy and the its accompanying uh field guide, the QS field guide, are both available on Amazon and other places and uh great way to learn about more deeply about these methods.

SPEAKER_00 1:01:00

It's it's a really good book. I I read a lot, and uh you know, most things I read are maybe shorter and a little quicker read, and you know, you come out with a couple of a couple of nuggets and a little inspiration. This is a long read because it's it's so much good information. I find myself like underlining um you know several things on every page. So it's really for people in our world, in our space, um, I think for most of all of our listeners, the book will resonate, right? So definitely, definitely consider it. And I follow Dave on LinkedIn, and Dave, I get yeah, your recent posts about chief quality officers and sort of the history of that, like very interesting, and I think would resonate with uh a lot of our listeners as well. So we'll we'll tag you when we make these posts with the episode and so people can can find you on LinkedIn. Thanks so much for being on Ryan. Thanks for co-hosting. This was a lot of fun. This is a great conversation, Dave. Lots of good information, good insights, and really appreciate you taking some time to be on with us.

SPEAKER_02 1:01:56

Well, Rex, Ryan, thank you so much. I appreciate you inviting me to be part of this conversation, and I wish you the best of your uh with your work uh with your clients and in such an important industry.

SPEAKER_00 1:02:07

You too. Thanks so much. Thanks, Ryan. Okay, thanks everyone. We'll talk next time.