Editor's note: Mark Graban, MS, MBA, is an internationally recognized expert in Lean Healthcare, which has become one of the most popular paradigms for health care organizations seeking to improve performance and decrease waste. He has authored numerous publications about engaging health care workers in implementing Lean strategies. We spoke with him about applying Lean in hospitals.
Dr. Robert Wachter, Editor, AHRQ WebM&M: You came into health care from a business and industrial background. What are the real differences that make it important to adapt Lean to the specifics of health care?
Mark Graban: Some parts of the hospital can be reminiscent of a factory. It's no accident that some early adopters of Lean in hospitals were clinical laboratories, where there was not much direct patient interaction. There are opportunities to use Lean principles in rearranging the laboratory equipment to reduce turnaround times. And in collecting specimens to improve the flow instead of just focusing on cost. In hospital lab settings, there are huge opportunities to improve the way we manage and to engage employees in improvement, instead of just expecting them to come to work and follow a standard operating procedure. Lean helps us to create workplaces with better flow, better safety, better quality, with lower cost as an end result. When you work together with nurses and other staff to design the workplace to be efficient, that workplace also becomes a safer and a more caring patient care environment.
I've always viewed my role as a Lean person as not coming in and telling anyone how to do their job, but rather teaching them Lean principles combined with the mission and values of hospitals and health care professions. "Let's take a look at what we're trying to accomplish and figure out how to accomplish that better." That's a different approach. I think sometimes people hear Lean and think manufacturing, and they're understandably skeptical. They think, "You're trying to turn us into a factory," which is not what Lean is about. It's trying to take these caring individuals and caring settings and making the hospital the best hospital that it can be by engaging people and managing differently.
RW: During the first 7 or 8 years of the patient safety field after the Institute of Medicine report, people talked about PDSA cycles, safety culture, root cause analysis, learning from mistakes. Lean was not part of that vocabulary, and it seems like it's cropped up in the last 3 to 5 years as efficiency imperatives have grown. Did we make a mistake not embracing Lean as an important model in the early years of the safety field?
MG: I wouldn't fault the patient safety movement for this. The discoveries people were making in the very early 2000s were about the applicability of Lean management concepts. The porting of "Lean manufacturing" into health care was happening in some limited experiments in a few places. ThedaCare and Virginia Mason Medical Center both started in roughly 2003. But all of the things you mention there—PDSA cycles, root cause analysis, problem solving, continuous improvement—those are all core pieces of Lean. We talk about Lean as this different way of engaging people in the workplace, a different way of managing. Those are core themes from Toyota, and there are similar origins in the work of W. Edwards Deming. I love reading Don Berwick, Lucian Leape, and others who directly worked with or learned from Dr. Deming because we see those same things in the Toyota or Lean approach.
RW: UCSF is doing a pretty big Lean push, and it's a relatively new technique to me. One thing that has surprised me is the degree to which it appears to be about changing the relationship between frontline people and senior leaders and managers. Talk a bit about the degree to which Lean influences hierarchies and relationships, and top-down versus bottom-up.
MG: Lean is very, very different than that traditional top-down—what's often called command-and-control—management model. Lean has a much more participatory approach. Toyota talks about one of their core pillars being the idea of "respect for people" and realizing that everybody in an organization—regardless of their level of formal education, the size of their paycheck, and the number of letters after their name—has an important role to play. We can respect that role they play as part of a team. Everybody has ideas to bring forward about improving their own work. Everybody can point out problems. When an organization listens to those people and then helps them fix things, these are the cultural themes that lead to improved safety and improved quality in a factory or in an operating room. John Shook from the Lean Enterprise Institute spent many years at Toyota. He was the first American to work for Toyota in a management role, and he said it's both top-down and bottom-up. He would say it's not laissez-faire management or "do whatever you want as long as you get results, I don't care." It's focused on the process and the results. The top-down role that leaders still have is to help define strategy, set direction, and define what the organization needs to do and why. That's a critically important role.
Traditionally in a command-and-control organization, senior leaders would say what to do, when, where, and all the details of the how. In the Lean environment, there's this back and forth where leaders set direction, look at the big picture, and then rely much more on the people doing the work to help define the details of their process improvements, even sometimes what they prioritize in terms of making things better. The key is those back-and-forths, where it's both top-down and bottom-up. We don't just delegate and dump everything on the frontline staff and say, "This is Lean, you're empowered, go do it." The role of the leader changes from the person who has all the answers and makes all the decisions to being more of a coach, facilitator, and teacher. I realize that sounds uncomfortably warm and fuzzy and squishy to people, but it's that key difference of not always jumping in and telling your employees what to do—it's easy to say, but it's a hard habit for people to develop. One key challenge of Lean is that it's easier to say what Lean is than it is to change our behaviors and the culture.
RW: So how do you do that? It's one thing to teach a technique, it's another to take someone who has been successful by virtue of climbing up the totem pole of the organization and being anointed the leader and say, "the style you've been using is not working and you need to change it."
MG: There are some really compelling examples where very senior leaders have started to be open with their employees about things that are broken in the existing culture. Many times, people talk about culture change without having the hard conversations about what exactly is wrong, what needs changing, and how did it get to be that way. When leaders start demonstrating different behaviors, admitting mistakes, looking at the process and systems, and taking a different approach when things go wrong, leaders can set a really powerful example, whether they mean to or not, from the top down. A chief operating officer that had worked his way up from being a frontline nurse was embracing Lean and was trying to change. He still had moments where he would fall back into traditional management behaviors and sometimes he would catch himself sooner than he would other times. When leaders are willing to strip away at some of that cloak of infallibility and start being open and honest with people, they really have to model the behaviors that they want to see through the rest of the organization. Otherwise people will get cynical about Lean. "You say we're trying to do all of this, but you keep blaming people and throwing them under the bus." People will see those inconsistencies when the actions don't match up with what's being preached.
RW: You mentioned blaming people and throwing people under the bus. Early on the safety field had a philosophy that it's mostly about bad systems, not about bad people. It should be a no-blame culture. How does Lean speak to the tension between no blame in systems versus the need for accountability?
MG: If we look back to the work of Dr. Deming, he would say, and the number changed over time, roughly 94% of problems are due to the system. As much as that number might have fluctuated a bit and it's completely unprovable, he never said it was 100%. The shift in mindset is that traditional management first blames a person. In the Lean philosophy, the first thought is to look at the system and processes, unless you can absolutely prove that somebody was being just completely irresponsible. We're not trying to go out of our way to make excuses for people. If somebody is being pressured into saving time and into doing something that they knew wasn't right but they did it anyway, I'm still going to look mainly at the system because there's this very human desire to put a neat little bow around problem solving. When we blame an individual, fire them, and strip them of their license, it's easy for people in society to say, "Somebody got what they deserved and we have some sort of retribution." But when we start looking at which behaviors actually save lives, I would rather err on the side of blaming the system more often than not.
The best approach to quality and safety doesn't come from punishment and retribution, but from understanding what we can about the system to prevent future occurrences. The huge opportunity in health care, maybe even more so than manufacturing, is the need to take advantage of near misses. We know people are afraid to report those because they do get blamed. They're at the sharp end of the stick and they get blamed for it. When people are afraid to come forward, we don't identify risks that would protect patients. We'd end up with an environment where we can only react after somebody's been harmed, and I don't think that's right.
RW: One early technique of trying to connect leaders to frontline workers in safety was something called executive walkrounds. The idea was that at some periodicity the leader would come out of the office and wander around the floor and ask people about their concerns and safety hazards they were seeing. A few recent studies have shown that they don't work very well. What do you think about that concept and how does Lean approach that issue of management by walking around?
MG: There are a couple of aspects of that. One is thinking about what do we do when managers go out into the workplace? A lot of management by walking around that I've seen is based on, "get out there, be visible, slap people on the back, shake hands, say hello," and it's kind of superficial. Where in the Lean approach what people would call a gemba walk—gemba means the actual place or we'd say the workplace—is more focused. We're going out to ask a specific set of questions. Each particular walk has a theme. I don't like the phrase wandering or walking because I think sometimes a gemba walk means staying put in a place instead of just continually walking. There's a difference in culture where you can see a practice of where they call it management rounding or a gemba walk, you can see that behavior taking place in a high-trust workplace. Where there's mutual trust between employees and leaders and from leaders to employees. Those walks could be productive where if we're asking people to speak up, leaders don't react badly, they're constructive and helpful.
If we try to plop that same rounding activity into a really dysfunctional culture—and I've seen at times in health care where department directors start walking around in the workplace and everyone freaks out because they think somebody must have died, or what went wrong—why is the boss out here? It's going to take time to build up even just a basic level of comfort let alone building up trust to where you can have those open discussions. Look at what's going on in General Motors in response to ignition switch problems that have led to deaths. The new CEO has announced this huge campaign of "Speak Up for Safety." I can tell you, having worked at General Motors in the mid-1990s, the problem was not that people wouldn't speak up. The problem was that managers didn't listen. I was discussing this with someone online who said they really need to be promoting a program called "Listen Up for Safety" because it sounds like the CEO's blaming the workers for not speaking up. This is true in health care; people will speak up if the system and the culture allow it.
RW: Let's talk about information technology. Health care has gone in the last several years very quickly from largely analog to a largely digital industry. When you walk around hospitals and clinics that have gone digital you hear a lot of complaining about workflows, people not looking each other in the eye, and not having time to think anymore. You don't hear Lean talked about much in the context of implementing IT systems. Is there a connection there?
MG: There are a couple of connections. Some people have done work with Lean and EHR or EMR systems where the application of Lean was to take a look at workflows before they even selected a technology. The Toyota approach is certainly not to view all technology as bad, but it also doesn't view technology as a silver bullet. The best technology supports your people, processes, and workflows. It fits the way your people do their work. We can improve the process and workflow before we end up selecting technology. Technology too often is dictating the workflow change and that's not ideal. I think a lot of organizations didn't take that step of fixing the workflows first. They talked to other organizations. They choose technology for a number of different reasons, and then they kind of force it on people.
I cringe when organizations complain that doctors aren't accepting CPOE. Of course they're not accepting it—when you ask, "did you involve the physicians in designing how this was going to work," the answer is usually no. "Did you improve the workflows and make sure the technology fit the way they want it to work best?" And the answer is often no. Even, "did you give them proper training to know how to use the system?" In a lot of organizations they've really underinvested in that training. Instead of blaming the doctors, we have a systemic issue where the technology either wasn't selected or implemented properly. That's a pretty common problem.
There's a related issue where compared to industry, new technology is supposed to make you more productive, not less productive. It happens way too often in health care where new technology or automation actually hurts productivity. Unless there was a huge quality and safety gain from that automation, that's not the way it's supposed to work. Ideally automation would help reduce the risk of errors and improve productivity. But you're right, everyone complains about their computer systems. That's pretty universal.
RW: When organizations that have a Lean philosophy and are pretty well schooled in it implement IT, do you find that they do it differently and better?
MG: It's a matter of longer upfront investment in that workflow improvement—the upfront investment in getting physicians involved or paying them to be involved in that process. It's going to take longer and maybe be more expensive upfront, but there's a great Lean lesson from industry of "you go slow to go fast." If you want to think about it as a PDSA cycle, putting more time in to the "plan" allows us to do faster, and we're less likely to have to adjust as much when we study the results. Many organizations rush through the plan, get in to the do, and then either end up with something that requires a lot of iterations to fix or that's broken and nasty but you're stuck with it. There are similar parallels to building physical space, when hospitals are using a Lean approach to design and construction. Organizations spend more time on the upfront process, then actually build, and move in faster and less expensively with fewer change orders and less rework.
RW: There's a long history of industrial improvement methods that have been used religiously for 3 or 5 or 7 years, and then people get disheartened, say that doesn't work, and then move on to the next thing. Do you worry that Lean will be a flavor of the month or do you think it has legs?
MG: I think it has legs. We're well past the point where people can say Lean does not work in health care. But that doesn't mean that Lean is guaranteed to work and it doesn't mean that Lean is easy. I worry when I talk to an organization that is now embracing Lean, and their thought process is often like well, TQM didn't work, reengineering didn't work, Six Sigma didn't work, benchmarking didn't work. If all these other programs and approaches and mindsets failed, what's different this time? Because these are different methodologies, it's unfair to say such and such failed or this didn't work. Unless they're going to take a different approach with Lean, I don't know why they would expect more success that time around. If people really read about Lean and study, they'll see that Lean is not just a program that some experts learn about like with previous quality programs. Lean is much more broadly involving of all employees, and it's not just about doing these projects but getting everyone involved in continuous improvement every day and trying to change the culture and the way we manage.
RW: If a hospital really was committed to doing it the right way, hired the right people, created a broad enough approach, and they said to you we're going to give this x number of years to see whether it succeeds. If it doesn't, we'll pull the plug. What is x?
MG: Well, it probably depends on how large the organization is. I think it was no accident that some of these great success stories, like ThedaCare and Virginia Mason, tend to be relatively small organizations. They're not the nationwide, huge system and collection of hospitals. I think it's multiple years. There's a progression where an organization should be able to find the right problems to solve early in their Lean journey, whether that's pressing safety and quality issues, whether that's trying to improve their cash management cycle. They should be able to find something where they can make a relatively limited but significant impact fairly quickly, within the first 6 months. That's at least a local demonstration of how Lean can help us be a better hospital.
Starting to train managers and leaders and to change the culture, that probably takes 3 to 5 years where it starts feeling like this is our new culture. We're not perfect at it yet. We still fall back on old habits and old behaviors, but it's starting to take a life where it would be really hard to undo this. That transition might be 6 to 12 months within one department. It's probably multiple years for a hospital where it really starts to feel like this is less of a program and more of just our new way of doing things. The people at ThedaCare and Virginia Mason are incredibly humble and say, "we've been at this for 10, 11, 12 years, and sometimes it feels like we're just now understanding what the problems are." They're not perfect hospitals. They're not perfect leaders. It's like we're trying to adopt a new healthy lifestyle where we eat better, and exercise all the time instead of relying on a new pill or a new surgical procedure to magically make us healthy. These are tough changes for people and organizations to really make happen.
RW: Anything else you wanted to talk about particularly in the realm of patient safety?
MG: If we look at the systemic root causes for errors and patient harm, many of these things are fixable. It's thankfully not a lack of medical training or lack of good people. It's both encouraging and frustrating sometimes that what tends to go wrong in health care is very preventable, and health care as a profession or an industry, frustratingly, tends to repeat a lot of the same mistakes because organizations don't learn from each other. They hear something in the news and they think, wow, we wouldn't make that mistake here. Trying to better understand human factors of when people start to get physically or mentally fatigued and the impact that that has on safety and quality is an important struggle. The other thing that we're still really trying to get across is don't blame people.
For example, if people aren't washing their hands often enough, there are usually systemic problems getting in the way of practicing proper hand hygiene. It's frustrating that a lot of hospitals and leaders still think the solution is telling people to be more careful and putting up clever posters and engaging signs and slogans. Again, this comes back to Dr. Deming's teachings that if slogans and posters worked, we would have already solved these problems in health care.
One great advantage we have here in San Antonio is a Toyota truck assembly plant. I like to volunteer and organize tours, and the Toyota people are very open and gracious about letting people come in. So we'll have medical students, MHA students, nursing professionals come to the Toyota plant. I always try to teach them in advance, there are a lot of cool robots in the factory, but Lean is not about robots. Lean is about asking the questions about what happens when an employee needs help. They reach up and there's no shame in pulling a cord and, within seconds, a team leader is there not to yell at them for slowing down production but asking, "hi, what's the problem, what can I do to help?"
That's one of the most powerful things when the health care folks see that and wonder how much better our workplace would be if we could actually flag a problem and have someone there to help. You see those types of lessons visiting a Toyota plant. It's not because they have great posters with slogans telling everyone to do their work more safely. They've really tried to address these systemic issues. They try to embrace a culture where everyone participates in continuous quality improvement, continuous safety improvement. Whether you go visit a Toyota plant or another good Lean manufacturer in your community, I encourage people to try to build those partnerships and try to see how the culture and the management system is different in a very tangible, practical way. It's very powerful.