Healthcare organizations are always looking for ways to increase efficiency and profitability, while still delivering high-quality care and a great patient experience. In this informative webinar, Frank Cohen, Director of Analytics and Business Intelligence for Doctors Management, shares how Lean Six Sigma methodologies can help identify opportunities to reduce waste, boost productivity and drive improvement in your practice.
We also hear from Gaynor Rosenstein, Chief Clinical Operations Officer at Crystal Run Healthcare, a large multi-specialty group practice based in Middletown, N.Y., who shares how the organization has used Lean and Six Sigma principles to achieve its goals.
What is Lean Six Sigma in Healthcare?
Lean and Six Sigma are two improvement methodologies used to reduce waste and eliminate defects. Lean Six Sigma combines elements of both, including analytics and statistical rigor, a focus on process improvement, and optimization of resources. Healthcare organizations use Lean Six Sigma to improve quality, service and efficiency.
Attendees will learn:
- Why Lean Six Sigma principles are a good fit for healthcare organizations
- How to get started implementing Lean Six Sigma in your own organization
- How to identify “low-hanging fruit” for initial projects
- How technology systems can help improve efficiency and reduce waste
- Lean is scalable. It works in any size project or organization.
- Think small. Break projects down into their smallest components—small wins are powerful. For instance, if you’re thinking of doing a throughput analysis, start with just wait times.
- Focus on antidotes, not anecdotes. Always include the use of evidence-based decision modeling, which requires data.
- Be sure to define the end of the project. Project creep can destroy your efforts.
- Think outside the box. Don’t be constrained to the same solutions you’ve always used. Get creative.
Lean Six Sigma Webinar Transcript
Maureen M.: 00:00 Hello, everyone. Thank you for attending today’s webinar, Lean Six Sigma: Driving Improvement In Your Healthcare Organization. I’m Maureen McKinney, the content director at Phreesia, and I’ll be your moderator for this webinar.
Maureen M.: 00:12 We chose today’s topic because we know so many of you are looking for ways to be more efficient and drive improvement in your organizations, while also enhancing quality in the patient experience. No easy task. To help you get there, we are so fortunate to have with us two great speakers, who will share their insights and personal experiences with process improvement.
Maureen M.: 00:34 First, we have Frank Cohen, Director Analytics and Business Intelligence for Doctor’s Management, a healthcare consulting firm. Frank is an expert in applied statistics, predictive analytics, and process improvement strategies. He has a master black belt in Lean Six Sigma, and he’s the author and co-author of several books, including Lean Six Sigma For the Medical Practice: Improving Profitability By Improving Processes.
Maureen M.: 01:00 Next, we’ll get some valuable first-hand experience from Gaynor Rosenstein. Gaynor is the Chief Clinical Operations Officer at Crystal Run Healthcare, a large multi-specialty group practice in New York State. Gaynor has held many senior administrative roles in healthcare organizations and is also a certified Six Sigma black belt.
Maureen M.: 01:19 Before we turn the floor over to Frank, I’d like to review a few housekeeping items. Today’s webinar is scheduled for one hour, including time for Q and A at the end of the session. At the bottom of your audience console, you’ll see several applications widgets. You can submit questions at any time, using the Q and A widget, and I’ll ask as many questions as time allows.
Maureen M.: 01:41 In the resource list widget, which looks like a green folder at the bottom of your screen, you’ll find slides from today’s presentation, along with an overview of Phreesia, and one of our recent infographics about efficiency killers in healthcare organizations. You can expand your side area by clicking on the maximize icon on the top right of the slide area, or by dragging the bottom right corner.
Maureen M.: 02:04 If you have any technical difficulties, please click on the help widget. It has a question mark icon and covers common technical issues. Since it’s always one of the most frequently asked questions, this session is being recorded and we will share a copy of the recorded presentation by the end of the week. We’ll be live tweeting highlights from today’s webinar, so be sure to follow Phreesia on Twitter, @Phreesia. If your practice is active on social media, you can retweet our updates and we encourage you to share your own post about what you hear today, using #phreesiawebinars.
Maureen M.: 02:38 A quick word about Phreesia. Phreesia gives healthcare organizations a suite of applications to manage the patient intake process. Our innovative [inaudible 00:02:47] platform engages patients in their care and provides a modern, consistent experience, while enabling our clients to maximize profitability, optimize staffing, and enhance clinical care. To find out more about Phreesia, visit Phreesia.com.
Maureen M.: 03:01 With that, I will turn it over to Frank Cohen. Frank?
Frank Cohen: 03:06 Thank you. Thank you very much.
Frank Cohen: 03:08 My name is Frank Cohen, and as the introduction goes, I am a specialist in what’s called computational mathematics, which includes applied statistics and predictive modeling. As part of that, I do great deal of work in the area of process improvement, operational research, and the like. And I’ve got to tell you, I’m very excited about talking about this today. The big reason is because we’ve got this whole push towards value-based care, away from fee for service type models. Value-based care says we have to do our best to take care of our patients in the most effective way that we possibly can, and as efficiently as we can. As everybody knows, there’s going to be less dollars to go around.
Six Sigma in Healthcare
Frank Cohen: 03:53 I would start by asking you this question: what would you say is the primary responsibility of a healthcare organization? Just think about it for a second. What’s the number reason would you think the main goal or objective to what we do in practicing medicine? When I asked this at a conference, and I can see people, and they raise their hands, the majority of people will say it’s to provide quality care to our patient population.
Frank Cohen: 04:23 While I think that’s important, I’m going to say that I think the primary responsibility is to be profitable. If you’re not profitable, unless you’re the federal government, you can operate at a deficit for only so long before eventually you just go out of business. How do we increase our profitability within a practice, or at least maintain enough of a profitability so that we’re able to meet the requirements for providing quality healthcare?
Frank Cohen: 04:58 One thing we can do, is we can take a look at improving profitability by either improving our revenue, generating additional revenue, or by reducing our expenses. If you’re in financial parts of your practice at all, or management, you know that this is the basic concept behind profitability. It’s a ratio of revenue to expenses. We have to increase revenue, we have to decrease expenses. How do we do that? Let’s talk about that.
Frank Cohen: 05:32 For reducing expenses, we could cut staff pay and benefits. Raise your hands if you’re okay with that. Right? I can’t see you, but I can’t imagine anybody has raised their hands. What happens if you lower staff compensation below market value? People leave. They quit, they go to find new jobs. When you have a high staff turnover because of that, what does that do to the quality of the continuity care? It reduces the continuity of care. When you reduce continuity of care, you also reduce quality of care.
Frank Cohen: 06:10 We can also eliminate FTEs. Great example, we had a practice we were doing a project for. What they did before we came in was, they had five coders and they all had to code, I think it was 90 encounters a day, or something like that. When right before we came in, they eliminated two of them. Now, they had three coders who had to code the same number of encounters as five coders did before that. Well, what happens? What we did, as part of a Lean Six Sigma project, we did a time assessment. What we looked at was a time scale.
Frank Cohen: 06:48 In the first half of a day, they coded the same number of encounters, the three of them, each coded the same number of encounters as they coded when there was still five of them. Come the afternoon and the crunch was on, they started coding more, and more, and more, until they were basically stumbling over each other, trying to code so many encounters during that time. The result was that the error rate increased.
Frank Cohen: 07:15 When they would get denials, do denial analysis, they would find out that these codes were being denied because they were coding them improperly. They weren’t providing the proper documentation or they weren’t coding them based on the documentation they were getting because they didn’t have the time to do so. The result was, is that the cost of the errors was more than the cost of those two FTEs. Bringing those two people back provided them with a more efficiency and more accuracy overall.
Frank Cohen: 07:45 You can also pinch capacity. You can try to restrict, I guess, the number of patients that you see, the services that you provide, but we’re still in a fee for service market. If you do that, you’re simply going to cut your nose off to spite your face.
Frank Cohen: 08:00 Reduce overhead. I haven’t met a practice administrator, probably in the last five years, who has any overhead left to cut. I think that we’re pretty much down to the basic amount that we can be. The bottom line is, quality is expensive. If you reduce your expenses below the point of where you can afford quality care, then you’re going to end up going out of business, anyway.
Frank Cohen: 08:25 We could increase revenues, right? We could increase our charges. This amazes me. Have you noticed healthcare is the only industry, I think in the whole world, where what you charge has absolutely nothing to do with what you earn, what your income is. I think that’s amazing, right? I’ve been in this industry for 40 years. I’m an old guy. I’ve got to tell you, I still sit in amazement that charges have almost nothing to do with reimbursement.
Frank Cohen: 08:57 You can increase collections, and that’s pretty hard to do, also. The payers have a natural animous towards providers. I don’t care what they say. Maybe I’ve done this long enough to be cynical enough to see the fact, but the payers … look, our primary purpose, not our responsibility, but our purpose is to get paid reasonably for providing quality healthcare to our patients. The payer’s primary goal is to not pay us for providing quality care to their subscribers. So, it’s very difficult to increase collections.
Frank Cohen: 09:35 You could negotiate better contracts. People go out and do that, but there’s a point at which you can’t negotiate any better contracts, because the payers also have a limit as to where they’re willing to go. What a lot of people are doing is, they’re merging into larger groups. When you merge into larger groups, you can share certain expenses. This is, by the way, a form of Lean Six Sigma. There are projects that are designed in order to do this where you share administration, you share billing services, you share infrastructure and hardware, and there can be certain efficiencies in that. The bottom line, though, is that the practice has very little influence over those payments, and therefore, over the revenues.
Frank Cohen: 10:12 If we can only marginally affect expenses, and there’s not really much we can do on revenue, how do we improve profitability? Well, we have to become more efficient. To become more efficient, what we’re basically is that we want to do the same that we’re doing now, with less resources, or we want to be able to do more with the same amount of resources.
Frank Cohen: 10:39 I had a radiology group, we did a couple of projects for them, and we significantly increased their efficiency. What they decided, was they just wanted to work less, basically. They didn’t want to make more money. They just wanted to work less. They worked less and they still made the same amount. Their revenues were still basically the same and their compensation was the same, as a result of that.
Implementing Six Sigma in Healthcare
Frank Cohen: 11:04 The only way to become more efficient … or, maybe shouldn’t say the only way. The best way that I know of to become more efficient is to employ Lean Six Sigma projects or techniques. As you might guess, Lean Six Sigma is a composite. It comes from two basic components, or two paradigms. Six Sigma is a very robust, but very top heavy … or, it’s got large footprint. Implementing Six Sigma projects usually take many months or years. They involve large numbers of team members. They’re very expensive. I’m just going to tell you my experience. I don’t think I’ve ever seen a Six Sigma project succeed in a smaller organization.
Six Sigma vs. Lean Principles
Frank Cohen: 11:57 Granted, in large hospital systems or integrated delivery systems, I’ve seen lots of Six Sigma projects work. In smaller practices or departments, implementing Six Sigma doesn’t seem to work as well. It is very focused on the analytics and statistics, great deal of rigor in that area. Lean, on the other hand, does not rely as much on the statistical and analytical rigor. What it does, it looks more at the process model. What are we doing and how are we doing it? What steps are we taking and how can we minimize some of the functions and procedures that we go through on a daily basis, in order to do our jobs?
Frank Cohen: 12:36 Now Lean, the whole concept is to get right of waste. Waste is defined as anything that doesn’t contribute to either the value of the services being provided to the patient or of the employees of the organization. If there’s no value that can be found in a certain step, or a procedure, or a process, or a function, then unless it’s required by law, or regulation, or policy, we get rid of it, is basically what it is.
Frank Cohen: 13:06 Lean Six Sigma is this continuum. I happen to be a fan of really focusing on, or at least relying upon, analytics to a great degree in the Lean process. Look, number one, we have to define the problem. Oftentimes, without numbers, we can’t do. I’ll give you an example. If I were to ask you right now, “How many of you have patients that are late?”
Frank Cohen: 13:32 I assume everyone just about would raise their hands. I would ask you this question, “Is that a problem for you?”
Frank Cohen: 13:38 Most people will say yes. I’ve done this where I can see the audience. The question is, define for me “late.” How many minutes after what? Is it when they were scheduled? Is it after 15 minutes before the scheduled time? How do you define late? Then, I would ask you this, “Based on that definition, how many of you right now, if I asked you in a poll, could tell me what percentage of your patients are late every month?”
Frank Cohen: 14:10 I can tell you again, just from doing this live, very few people could report that. My question is, how do you know it’s a problem? If you don’t know how late people are, and you don’t know what percentage of people are late? We say that because it’s anecdotal. It makes sense, doesn’t it? It’s logical that if people don’t show up on time for their appointments, that would be problematic. But, maybe it’s not problematic in this particular practice. I believe we have to employ strong and robust analytics, even in the Lean side of it, in order for things to happen. I’ll show you what I mean in just a minute.
Frank Cohen: 14:47 The reason we want to have this problem, or this process, focused is so that primarily, we can understand how things work. We can take a look at the practice and we can look at all the different aspects of it, and we can start to understand better what the processes are, what the steps are in each of those processes. We want to be able to see it from start to finish. You know what happens a lot of times, is we’ll go into a practice and we’ll say, “Oh, let’s go ahead and start process mapping this out.”
Frank Cohen: 15:18 I’ll get into that in just a minute. And, what people will do is, they will develop a map of how they think it should be. Right? Because they know there’s problems. It’s really amazing. I’ll go to a conference and I’ll have 1000 people in the room, and I’ll ask this question, “How many of you right now, if I give you a piece of paper, could write down at least three or four major problems within your organization?”
Frank Cohen: 15:41 And everybody would raise their hand. So, write them down. Write down these three problems. Then my question would be, “If you know that it’s a problem, why is it still a problem?”
Frank Cohen: 15:50 If I could write down and tell you, for example in my business, the problem is that I don’t follow-up appropriately with my clients, and therefore, I lose the opportunity for repeat business. That’s crazy. If I write that down on a piece of paper, and I can see that’s my problem, why in the world haven’t I fixed that problem? A lot of times is because we don’t have this process focus. We don’t get to the root cause of things. Instead, what we do is, we blame people for messing up, rather than understanding the idea that maybe we don’t have the most efficient process that there is.
Frank Cohen: 16:21 There’s a combination of art and science in this. Lean, there’s a very artistic component to Lean. If you do projects correctly, they’re actually quite beautiful. Just like mathematics can be very beautiful at it’s core. There’s also the science part, which we’re going to talk about with regard to the analytics. Let’s start with idea of discovery.
Frank Cohen: 16:44 There’s three components we’ll talk about here. We’re talking about classification, correlation, and cause and effect. Those three associate to the first three, or the three most important, what I believe to be tools, which is process mapping, value stream mapping, and then doing fishbone diagrams, or cause and effect analysis.
Frank Cohen: 17:06 We start with this classification. What we’re trying to do, is just take a long around and see what things look like. How do people move? What redundant events do we see going on? Are there things that are being manually that can be automated? For example, this is one area where we find, a lot of times, that in areas where you have more sophisticated populations of patients, you can use kiosks instead of coming in and getting a clipboard, and everybody writing all that information down. It’s a pain to do it that way. That’s one thing that you can see … things you can see right away.
Frank Cohen: 17:41 Here’s an example. One time, whenever I do a project, I like to sit in the waiting room. Just hours, I’ll sit there for three, four, five, six hours, with a pencil and a piece of paper. I’m taking notes and I notice something. I notice that, particularly for new patients that were coming in, they get this whole big packet of information. They would get up and they would go to the desk several times, filling out the packet, and they would come back and sit down.
Frank Cohen: 18:05 Eventually, I went up to the desk and I asked this guy who was behind the desk, in the intake part of it, and I said, “I see all these people coming up, asking questions. What are they asking?”
Frank Cohen: 18:16 He said, “Well, a lot of them, they don’t understand the financial forms. They want us to clarify for them.”
Frank Cohen: 18:22 Some of them, they can’t see … they can’t read what it says. Why is that? Well, it’s an older population and they forget their reading glasses. It was really quite amazing. I got up, I went to a Sam’s Club, they had one down the street, and I bought 20 pairs of reading glasses in different strengths, and a really nice little basket. I put the basket on the counter. There was a little sign that said, “Need reading glasses?”
Frank Cohen: 18:48 Of course, we made the print, the wording, very big so the people who didn’t have glasses even could read it. Reading glasses, take one. It was amazing. We cut what we believe, and we measured, about seven minutes off the time it took for someone to do an intake for a new patient visit, just by observing what went on within the waiting room.
Frank Cohen: 19:11 Another example was, we had a case where … this was in Florida, elderly patients. A normal adult will walk at somewhere around 5.2 to 5.6 feet per second. Where an older person, or somebody who may have some debilitation, walks down towards the 2.6 to 2.8 feet per second range. What we noticed was that all the physicians’ offices were up in the front of this big U-shaped building and the treatment rooms were in the back.
Frank Cohen: 19:38 They had to take the patients and walk them all the way to the back to the treatment rooms. We estimated that it cost a little over two minutes per patient visit. When you have several hundred patient visits a day, that’s the equivalent of seeing an additional eight or ten patients a day. What we did was, we rearranged the building, moved the offices around, and were able to solve that problem. We want to look and see what things look like. That’s what process mapping is about.
Frank Cohen: 20:04 I mapped the process. The patient checks in. If it’s a new patient, for example, they have to do intake forms. If it’s not, excuse me, then it’s a doctor’s visit. Is there a visit check-in? No? Then, maybe they just have to go to the lab or something. Whatever the process might be. Same thing here. This is for a lab study itself.
Frank Cohen: 20:28 We map out the process, which simply identifies each of the individual steps in the entire process, as to what we’re doing now. A lot of times what people will do, is we’ll want to map the whole patient visit process. We want to do from when the patient walks in the door to when the patient leaves. What you have to know is, that within that, there are sub-processes.
Frank Cohen: 20:50 For example, this whole idea of check-in can be broken down into a sub-process. If you’re listening to me right now, and you work in the check-in or at the front desk, you know for a fact it’s not just one little bubble check-in, and that’s it. It is broken down into many parts. There’s insurance verification. Does the patient owe any money from the last visit? Is there a co-pay that has to be collected this time? Are there any particular issues? Are their demographics changed? Has their insurance policy changed? Just a whole host of things.
Frank Cohen: 21:23 Just be careful, when you do this, not to map too much. In this case, we’re just looking at an overview map. We don’t have time in this presentation, but we get into detailed map. In a patient visit cycle alone, the through put cycle, there can be a dozen different detailed processes that need to be looked at. Just keep that in mind as we go through this.
Frank Cohen: 21:47 The next thing we want to do is, we want to understand this concept of correlation. Correlation is actually a mathematical conclusion. Correlation uses a calculation, in order to determine how closely related certain events might be. For example, engine size versus gas mileage would be negatively correlated. The larger the engine, the less gas mileage I would get. That’s pretty easy to see, right?
Frank Cohen: 22:18 For example, I can tell you that the longer the time it takes for a new patient to get an appointment at the practice, the higher the number the no-shows. After a three week period, if it takes more than three weeks for a new patient to get in to see the doctor, then that no-show period is going to expand significantly. Those things are important to me.
Frank Cohen: 22:45 Now, the question is, does that time cause the no-show? Or, is there some other confounding variable involved in that? That’s something that we look at when we look at the cause and effect part of it. But, this is a part. Okay?
Frank Cohen: 23:07 I’ll give you an example. This is where we can actually save ourselves time and effort. I had a rural practice, I think it was out in Missouri, and it had got a new administrator. It was a dozen physicians, so it was a pretty good size. This new administrator was this … had an MBA or MHA, and had worked at a large practice in the city, and came with all these great ideas. The first thing that he wanted to do, was to take their intake process for new patients and move it to the internet. They spent a huge amount of money on a HIPPA certified, secured website, and all that. You could go online and fill out the intake forms online, and they were pushed through.
Frank Cohen: 23:55 The problem was, nobody was doing it. Maybe 30%, 20% of the patients. That’s not enough when you spend that kind of money. When we came in to look at doing some of the work for them, we did a survey. We found out that fewer than 50% of the patients is, I think it was 41%, had access to the internet. Of those, only half accessed it regularly. Obviously, this is many years ago and it was a rural area. But had they stopped and even thought about looking at some of those aspects, asking those questions in advance, they would have correlated those two items. Then they would have had a much better chance at saving that money.
Frank Cohen: 24:34 In building this correlation part, we want to also build inspection points. They’re actually kind of … excuse me, kind of interesting. Here’s where … an order from chemo, and then it says order printed from the Mosaic system. The codes are handwritten on the order to each drug. Pre-med solutions you move from Lynx. Lynx is one of their systems. It goes to the nurse before compounding. This is for chemotherapy, by the way. This is a big cancer center that we worked on.
Frank Cohen: 25:01 All items are placed in a basket and set on the counter. What the purpose of this is, we can look at what happens in each one of those individual steps. We want to be able to create contingency plans, in case things go wrong. We want everyone to be on the same page and have a better understanding. It’s not often, really, if you think about even from a management perspective, where we break things down to this level of granularity, and it’s really important to do so.
Frank Cohen: 25:28 Here’s an example of an INR clinic, where they’re doing anti-coagulant blood testing. And you can see, we added some numbers. For example, 30% of appointments are no-shows. Okay? We found out that 60% of people come by car and 40% by public transportation. One of the big problems is that the parking lot is full, so what do people do? They leave and they go home. Right? There may be waits of 15 to 30 minutes to get this done. Or, in the waiting area, once they’re in the waiting area, it could be an hour, hour and a half, that they have to wait for the results, to determine what the dosages are going to be.
Frank Cohen: 26:09 This is a situation where we found out that the biggest problem was transportation. Right? People either couldn’t park or they couldn’t afford it, or they had to take public transportation. It wasn’t overly reliable. Here’s what we did. First, we checked with the OIG, make sure this wasn’t incentivizing against our rules. What we did, we went to the cab company, local taxi company, and we said, “Hey, we want to make a deal with you. We want to arrange a set of concentric circles.”
Frank Cohen: 26:37 Here’s the hospital and we draw a circle three miles out. We draw another circle seven miles out, and we draw another circle 10 miles out. Most people were within a 10 mile radius of the facility. What we said was, “If they can’t make it in, we would call to verify the appointment. If they can’t, if they have problems with transportation, then you go pick them up and we’ll pay X amount of dollars for the three mile, X amount of dollars for the seven, X amount of dollars for the 10. We worked this deal out with the cab company.
Frank Cohen: 27:06 It cost us about $45000 the first year, but we lost $300000 years before in no-shows, the organization did. It was a tremendous boost for them. How did this happen? We just looked around. We just saw things and we said, “Hey, how come people aren’t showing up for these appointments?”
Frank Cohen: 27:24 The only way you’re going to know that is if you call them and ask them. So, we did. We did a survey of those folks and we came up with this information. Finally, we want to do this cause and effect relationship. We want to be able to link things together. I know you’ve heard this before, the typical, or the stereotypical example is, the rooster crows and the sun rises. You know? Does the rooster crowing cause the sun to rise or does the sun rising cause the rooster crow? Or, is there no relationship between them at all?
Frank Cohen: 27:54 For example, your payer mix changes. Maybe you lose a certain payer and the payer mix shifts, and you lose revenue at the same time. That may be correlated, but did the change in that payer mix cause that change in revenue, or were there other variables or confounding variables, involved?
Frank Cohen: 28:11 We talked about the length of time to appointment, right? Its longer and our no-shows increase. Does the length of time to appointment cause those no-shows? The way we find these things out, is by using teams. We get those subject matter experts, the people within the organization, who know the most about what’s going on. We get them together and we ask them those questions, and we try to determine it.
Frank Cohen: 28:34 Here’s a simple one. We have a long wait time for new patients, and we’re trying to figure out why is that happening? Well, is it a wave scheduling problem? Maybe we’re not doing phone triage. People call with really complex problems and we put them in the same 15 minute or 30 minute time slot as somebody who calls that doesn’t have complex problems. Maybe it’s the time of the visit. We tested that. We did an analysis, where we took one patient at the beginning of each hour, for a whole week, and then we did it for a month. We went across. We said, “Does the time of day have any relationship to being late?”
Frank Cohen: 29:09 And we found out, no, it didn’t. One of the big problems were unnecessary forms. They had an ABN, and in every one, people would read through it and sign it. I don’t even think that’s legal. You know? They also did manual insurance validation when people got there instead of ahead of time. They’re sitting on the phone for a long time, waiting to