Accountable Care Organization

Episode 59: A Recipe for Reducing Length of Stay

April 2024

In less than a year, Southern Maine Health Care reduced length of stay by thirteen percent and saved the health system nearly four million dollars. MaineHealth Southern Region Associate Chief Medical Officer Michael Albaum, MD, led the effort and shares his “recipe” for success.

Mike C.: This is BACON, brief ACO news from the MaineHealth Accountable Care Organization. A monthly podcast for health care providers. I'm Mike Clark.

Heather: And I'm Heather Ward. Mike and I are practicing physicians and participants in the MaineHealth ACO. This month, we've got a great success story brought to us by the one and only doctor Mike Albaum.

Mike C.: Ah, success that warms the heart, doesn't it? Like a plate of chocolate chip cookies right out of the oven, huh? So?

Heather: Sounds good.

Mike C.: Sounds good. Let's get right to it.

Heather: One of the most consequential metrics in health care is length of stay. This measures the amount of time a patient stays in the hospital when receiving and recovering from treatment.

Mike C.: That's right, Heather, and length of stay, turns out is closely linked to care quality. Plus, it has a big impact on both patient outcomes and financial performance.

Heather: Okay, so basically if we can safely reduce length of stay, it is better for the patient, the payer and the hospital system. But actually, doing it can be complicated, not to mention challenging.

Mike C.: That's right. And you know, if only we had a recipe for reducing length of stay, an approach that worked at one hospital and could maybe be easily transferred to another.

Heather: Aha. Well, we have just the thing on this show today because joining us now is Doctor Mike Albaum. He's the associate chief medical officer for MaineHealth's Southern Region. Mike recently presented the results of a promising length of stay reduction project at Southern Maine Health Care. And it might just be the recipe that we are looking for.

Mike C.: Ah, excellent. That sounds great to me. Mike, welcome to BACON.

Mike A. : Great to be here. Thank you both.

Heather: So, Mike, we just briefly described why reducing length of stay is important. What was the impetus for your project at SMHC and more importantly, how did you actually get everybody on board with it?

Mike A. : Yeah. Great question. So really during Covid we found that our length of stay went up. So, you know, historically Southern Maine Health Care is a hospital, community hospital where the average length of stay is about four and a half days. And during Covid, for a lot of reasons, that went up and up that it eventually got to a point that it was six and a half even at some time, seven days. Some of that was related to Covid patients who, you know, could be complex and had a long length of stay, but a lot of it was related to constraints in other parts of the health care system. So it took longer to get people into a skilled nursing facility. It was harder to arrange transport for people who needed transport. Home care services took longer to get, and honestly, families weren't quite as ready to jump into the fray and support their loved ones in the way that they were able to prior to Covid. And again, because people couldn't travel, there were a lot more constraints. So we saw this trend. Many hospitals experienced this same trend. And so the impetus was we reached a point that the capacity of our hospital was being exceeded on a daily basis. So what that means is, you know, during Covid, we added rooms and all kinds of places. We brought units active that had been in mothballs.

Mike A. : But once we had done all of that, we still reached a point that we had more people that needed to be in the hospital than we had beds for, and so every day we would have to wait until 25 people got discharged so we could admit the 25 patients, which is about our average admissions per day, that we're looking to get into the hospital. And when that equation gets out of sorts, you end up with prolonged stays in the emergency room, people on stretchers in the emergency room, more people in the waiting room. So it really becomes a safety issue as well as a quality-of-care issue. So we knew we had to do something. In order to do something, you know, a hospital is a large and complex organization and so, what we did was got together stakeholders from across the spectrum. So nursing, physical therapy, case managers, hospitalists, social workers, pharmacy. And we said, what can we do? What do we think would be impactful in our facility. So, you know, we didn't do an exhaustive literature search or anything like that. But we said of the things that we already know that are out there, what do we think we could do to make an impact in rapid fashion and at scale? We also said this needs to be something that could apply across our whole hospital. So we started doing that work in, you know, spring and summer of 2020.

Mike A. : I'm going to get the year right. So, 2022. And pretty quickly got down to three main areas to focus on. One was we already had in place: an interdisciplinary care round, and that's where, you know, bunch of team members get together to talk about all the patients in house and what are their barriers to discharge. So that was already there, but it needed to be sharpened up. The second was our head of physical therapy, Margaret Ford, had attended a conference at Hopkins where they talked about early mobilization of patients as a way of getting people home faster. So that was the second major initiative. And the third was something that, you know, we learned from Maine Med called IPACE, which is Interprofessional Partnership for Advancing Care and Education. Basically, bedside rounding, an old concept that had kind of fallen by the wayside with EMRs and specialization of roles, but really getting the doctor or APP and the nurse at the bedside every day together, which again, doesn't sound revolutionary, but it wasn't happening in a reliable way. So those are the three main tactics: sharpening up our IDCR, interdisciplinary care rounds, early mobilization of patients and bedside rounding. And then it took quite a bit of time. Once we settled on the tactics to say, how do we implement it? We had work groups in each of those areas.

Mike A. : We had a wonderful facilitator, Alex Repond, who, you know, shepherded all of those workgroups. And we also chose kind of a go-live date. We said, let's not just test this forever. We said, we want to do this hospital wide. And it took us about six months to be ready to do that. And again, we did have a couple of, you know, pilots. We started with one team doing bedside rounding, then two teams, but we really went live hospital wide, actually, on Valentine's Day in 2023. We distributed chocolates to all the nurses and care team members in the inpatients with the little sticky that said, "I love bedside rounding." That helped build some goodwill, but it also allowed us to say, okay, now if we wanted to test the impact, we had a clean starting point, which sometimes can be a challenge in process improvements. A lot of support from nursing leadership, the Chief Nursing Officer and the nursing directors. Support from the hospital president. Finance was happy to see it happen, and we found out later the magnitude of the financial impact. And then the medical directors, working in collaboration with the heads of pharmacy, case management and physical therapy and rehab medicine. So I think those were kind of the core drivers with some project management resources through the quality department that allowed us to get this thing launched.

Mike C.: Wow, I this is such an enormous project with so many stakeholders and just a big lift, I would imagine. I can imagine that there were some significant challenges along the way. How did you get buy in? Because you it needed at every level, I imagine. And how did you keep things once you got it up and you kind of put that stake in the ground and said, "here's the day we're going to do this thing." How did you keep things on track as it rolled out?

Mike A. : We started by making sure that the whole house knew what was going on. So before we got started, we said, this is big, this is important. And we really made it a patient care focused initiative. When we set a target of saying, let's see if we can reduce by 10%. And the reason we chose that was then we said, okay, if our average daily census is about 100, that's ten fewer patients in beds upstairs. That's ten fewer patients in the emergency room waiting for admission. And that's an enormous difference in our 29 bed emergency room. You know, that was the thing, I think that gave everybody a really clear "why" for what we were doing and an achievable goal, you know, getting ten patients home one day earlier with an average length of stay of six and seven days, was something people could get their minds around and we could say, okay, go out and find those ten patients that we can make an impact every day. And it's also got to be, you know, it's not just a one and done. It's every day. It's got to be that. So a lot of intentionality that it was coming, a lot of visibility when it was here.

Mike A. : We have an operational huddle every day at 9:30 where all of the inpatient managers and directors are on a Teams meeting or Zoom meeting. And so, once we went live, we started posting and reviewing at that meeting once a week: how did we do in the prior week? First with process measures. Did the interdisciplinary care rounds meet their quality goals? What percentage of patients had bedside rounding? What percentage of patients with a mobility goal actually had walked towards their mobility goal? And so, we reported that every week. And then after a couple of months, we started to get month by month. What was it doing to length of stay? So that was a way of keeping organizational focus on it. We also needed to modify processes as we were going. One of the big barriers to bedside rounding, it turned out, was that one provider doctor or APP with an assignment of 15 patients could have as many as six or even seven nurses with those patients, which was interesting. But because, you know, you would think that would be hard to make it that complicated if you wanted to. And yet that was our experience.

Mike A. : I always admired the quote, which is either by Deming or Batchelder or Don Berwick about "every system is perfectly designed to achieve the results it gets." Well, if you tried to design a system that, say, make every hospitalist round with seven nurses, you couldn't design that system. A lot of work by the nurses and the hospitalists who were separately allocating patients, you know, to try to load balance how sick the patients were and said, hey, we really got to get one provider with three nurses for every 15 patients, and that required some amount of geographic cohorting. The other nice thing about getting those ratios right was one case manager takes care of about 15 patients. So now you could have one provider, three nurses, one case manager accountable for one block of 15 patients, which really improved the communication and the ability of that team to focus on what are the things that need to be done for those that are going home today, what do we need to be getting ready for those going home tomorrow? Getting that alignment together was really important thing to be successful.

Heather: That's great. I'm just going to take a second to summarize, because I think that this is really complicated but good information and I just want to kind of hit the high points here. It sounds like you picked three main areas to improve: interdisciplinary rounds, early mobilization, and you use IPACE, bedside rounding. You got workgroups and maybe started with some pilots. Everybody got buy in with chocolates and understanding "the why." Right? Like the improving 10%, reducing the patients by ten patients a day to reduce the number of patients in the ER, waiting in the hallways waiting. That sounds just like a ton of work. But hearing all of this now, I really want to know what the results were. And you know, what did you end up seeing as a result of all of this really strategic, focused work that you did?

Mike A. : The good news is that it did impact our outcomes. First we started by celebrating the quality of process measures in each of those areas. We audited the interdisciplinary care round. We had a tool, and we had a consultant that had come in and had already been something that was planned. Although we were well along our way with time, the consultant got here, but they helped with what's a checklist for the interdisciplinary care. We would have leaders audit those rounds to see how they were doing. Were they hitting the desired targets for a high quality IDCR round? We also reported, as I mentioned, the patient, the percentage of patients on a daily basis that had experienced bedside rounding. We said for all of these, one of the target of 80%, we said we're not aiming for perfect, but we want to see enough of the effect so that we can judge the outcomes. And similarly for the early mobility, it was really just targeting the patients that were either independent or required minimal assistance. And it wasn't an audacious goal. It was just was a goal set. And was any progress toward that goal recorded? Not even did they hit it or not. But you know, it's a starting point. So what we found was we launched in February. So we started we collected data for six months after that. So March through August was kind of our performance period. And to have something to compare to, we compared to the prior March through August. So from our baseline period in 2022, pre-intervention, our average length of stay was 6.3 days.

Mike A. : Post intervention it went to 5.4 days. So you know, a pretty significant drop. It exceeded our 10% goal. It was, depending on how you measure it, a 13 or a 14% drop in the length of stay. We were pretty happy with that. And the other thing that we reported regularly, and this was based on advice from Omar Hassan, the chief quality officer, was that as soon as you start talking length of stay, doctors start arguing how sick their patients are. We did look at some other measures that adjusted for complexity. We used geometric mean length of stay, which adjusts for the admitting diagnosis. And more recently we've been using a tool called Midas that adjusts for observers expected length of stay. And the trending was all in the same direction as you added in those other things. So we got our 13 to 14% reduction. Again, we weren't doing the cost accounting for it, but in the background, Rick Olsen, our CFO, and Lu Inzana at Maine Med came up with a methodology to say, how do you translate shorter length of stay into savings. And so at SMHC, that reduction equated to a savings of $3.7 million over the full year that we did that, that was that was fiscal year 23. And honestly, fiscal year 23, we didn't launch until February. Our fiscal year starts in October. That was really based on only about eight months of you know, the intervention being in place. So who knew, but shortening length of stay is a tremendous savings for the health system.

Heather: Wow. Fantastic.

Mike C.: Incredible work. As with any successful project, sustaining it, right, keeping it going is, is sort of the bug. What have you found at this point in the project? How are you sustaining it now? How is the buy in for the key stakeholders as you continue with these same goals?

Mike A. : Right. Well, one thing that's heartening is when I go to nursing huddles and ask, how's bedside rounding doing? They depend on it as part of their daily work. Now, in fact, if a provider doesn't show up for bedside rounding and we still have some issues with, you know, scheduling that, you know, one doc has got multiple teams he's trying to get to. They will track down the doc. And Andrew Goldman, the head of our hospitalist service, will help to ensure that those docs are there and available. So it's become part of our standard work, which is a good thing. And that's one of the other reasons to, you know, and this is another Institute for Healthcare Improvement principal you designed for scale, you know, so you want to design your projects at the scale you're trying to end up at, so that you don't just have a demonstration project that never really sustains. We still review our measures once a week at the huddle board. So once a week on Mondays, Alex Repond, who was the program manager, shows us how we're doing on our process measures, shows us how we're doing on the larger measures. We celebrate the successes on a regular basis. We showcase this stuff to MaineHealth leadership, to other hospitals. And so that's a good thing.

Mike A. : Alex, Nick Flavin and I put together a poster that we brought to IHI. Don Berwick looked at it and said that's great. We said, can you sign our poster for us? And he did. I've got a kind of a capstone for my career to have Don Berwick, you know, sign our project. We keep talking about it and we're keeping our eye on all of the measures. Because honestly, I think that if the process measures, the three initiatives don't stay reliable, we'll start losing ground on the overall measure. The other thing that we try to keep in mind is there's always a reason why an improvement project goes off the rails. We have more orientees, our census is high. But another, I think good marker for a process improvement is when things are unstable in the environment, these processes to be more reliable. So, when there's a strain on the system, that's the real test of whether you've made it part of your procedures. It's like you want to lean into these things and say, if census is high, these are the things we need to make sure are still happening. Otherwise, we're not going to work our way out of it.

Heather: Yeah, that is so true. I love that when there's a strain on the system, that's when these processes should be functioning the best to keep us on the right track.

Mike A. : We picked that up. It was one of the surveyors, believe it or not, when the state was here torturing us once over one thing or another. And I think it was during the sign out from one of those might have been joint commission, but they were like, yeah, the real test of your processes is, are they durable when they are stressed? And so that's kind of been a mini mantra that Helen Troy, who's in charge of quality down here, says over and over again. I think it's a good one to aspire to.

Heather: Yeah, absolutely. Okay. Then for those of us who are not at SMHC, like me, what advice do you have for other hospitals that might want to adopt this?

Mike A. : Yeah. So that's why when I presented it, you guys mentioned the recipe analogy a couple of times. So, when I've shared this with other members of the health system, I'm using the stone soup metaphor. And, you know, the stone soup allegory is, the two guys show up in the village. I think it's in Eastern Europe. And they start boiling water and they throw two stones in and everyone's like, what are you doing? They're like, well, we're making stone soup. And they're like, well, how do you do that? And you invite everybody else in the village, people start showing up with their turnips and their potatoes. And so we use that metaphor because I have to admit that whenever I get an off the shelf solution and say, here, this could work for you, I'm like, ah, you don't understand us, you know? And everyone's got that. I think that the basic principles are adaptable. So bedside rounding is where we should be for a lot of reasons, you know, not just length of stay, but I think for improving care team engagement, patient experience, education about medications, you know, IPACE. And that bedside rounding is right now. Does that look the same at Waldo as it does at Maine Med? No, it's not going to.

Mike A. : In an academic center with a lot of learners in tow, it can't look the same. So I think it's worth saying the good thing there is the provider and nurse responsible for the patient at the bedside, and how you get to that good can vary place to place. Similarly with the early mobility, you know, in our hospital that involves physical therapy, helping support nursing in doing this, but not physical therapy on point for the actual setting the goal, getting them patients down the hallway, nursing worked on, you know, how do you get that goal visible to the patient, visible to the family. So they're using the whiteboards, they're using Epic, walking people earlier so that they can go home instead of going to rehab or can go home without home services instead of with services. You know, we do see that as well in our data. The people that go home without services or the shortest length of stay home with services, meaning home care agency is the next longest and people going to sniff is the longest of all. And honestly, you know, we've all seen that patients in the hospital are usually pretty frail to start with. And if they spend four days versus six days on their back, it can really make a difference in where they end up keeping them on their feet can help.

Mike A. : Taking it to other hospitals, I think that's it. You know the similarly the interdisciplinary care rounds, who participates, what time of the day. What we found really important for that was really focusing on the data discharge. One of the innovations for sharpening those huddles was to say the day the patient is admitted using their principal diagnosis to set an expected discharge date. Prior to that, people were just pulling a number out of the air. You know, the case managers would say, yeah, we feel like this is going to be five days. And the hospitals would say, I think it's going to be seven, and that's fine. But we at least started now to say, okay, this is a patient with CHF and based on the national data on that, which is available in Epic, it comes right up, that's an expected five-day length of stay. I made up that number, but the principal is there. Then they could take that and say we think this patient's going to be here longer because fill in the blank or we think it could be shorter because fill in the blank. But it started from a common expectation.

Mike A. : So again those IDCR rounds are really organized based on who are the participants in your in your facility. The principal of a group should be getting together to talk about it. For us it works to have it mid-morning and then a secondary huddle mid-afternoon. And also in our shop we stopped having the physicians and APPS go to the interdisciplinary care rounds. We used to do that. It was kind of disruptive to the morning, which is the key time for getting discharges.  We found that as long as they had spoken with their case manager before those rounds, those rounds were effective and the care teams could be out discharging patients, which is what you need them doing in the morning. So that, again, it may not be the same in every location, I know other hospitals are working on trying to get the providers to the interdisciplinary care rounds. I know Maine Med is working on that now. So again, who participates, what time things happen and how you're organizing those you know deserve some local influence for what does it take to work in the local facility? So that's why this is more like making stew than baking a cake, where if you're if you're half a teaspoon off on the salt, you're in trouble.

Mike C.: That's a great analogy. Wow. Congratulations, Mike, for you and the team at Southern Maine Health Care and for and thank you for bringing this not just a report of success but inspiring us with the possibilities that that this could be scaled to other organizations. So thank you for being with us today, Mike.

Mike A. : Great. Enjoyed being here.

Speaker2: Thanks for listening to BACON this month. You can find all our podcasts on your podcast app and at our web page And if you have questions, comments or suggestions, we'd love to hear from you! Please email us at That's

Speaker1: BACON is produced by the MaineHealth Accountable Care Organization. Thanks for joining us this week. We'll see you next time.

Speaker2: See you next time.