Back to THEHANDOFF
No items found.

Episode 28: Using big data to address nurse staffing

September 22, 2020

Episode 28: Using big data to address nurse staffing

Listen on your favorite app
September 22, 2020

Episode 28: Using big data to address nurse staffing

September 22, 2020

Dan:
Welcome to the show Therese.

Therese:
Thank you so much. I'm happy to be here with you, Dan.

Dan:
Thank you. Therese, why don't you give us a quick overview of your history and where you're focused now, related to all the cool things you're working on?

Therese:
Well, I have been system chief nurse for a number of years, and in that role, had an opportunity to meet some wonderful, incredibly wicked smart, operations for search scientists who introduced me to the whole notion of optimization and linear programming. In fact, one of those leaders is currently a professor of math and business at Stanford University. And what I realized was that the really pernicious problems related to staffing that I had spent a career as an executive trying to solve, other industries had figured out a way to do that. So I actually moved my career and became an entrepreneur and started two companies around workforce optimization. In each of those companies, my partners were operations research scientists, had tons of fun doing that and then got a PhD in sociology. So my midlife crisis was to become a sociologist and figure out, our undergrad degrees, as nurses, we focus on body systems and individuals and their families as healthcare related systems. At the graduate level, I focused on health systems and how do all the pieces of hospitals work?

Therese:
So I figured it was time to figure out where healthcare fits into the larger social structure. That led to going to work for a large multinational company, doing optimization work across all disciplines, across the world, actually. And along that path took a position in faculty, in the college of nursing at the doctoral level, at the University of Illinois in Chicago, and subsequently moved to the School of Public Health where I teach in the MHA program. I am currently working with a financial consulting company as one of the few nurses within a company full of all sorts of CFOs and financial analysts, looking at healthcare, doing lots of merger and acquisition work and building workforce optimization within that organization.

Dan:
That's awesome. And I love that you're a nurse in a nontraditional setting, which I know a lot of listeners are and then that's been my career path as well. So one thing that we know is that staffing is the one thing on nurse leaders minds all the time. I think it's staffing, quality and safety, probably, maybe in that order, depending on the day, it seems like. But we hear a lot around workforce optimization, can you talk a little bit more about that? I think there's probably some dogma around that. When you ask a staff nurse, workforce optimization means higher ratios, maybe less staff, work harder kind of thing. On the hospital side, workforce optimization is maybe right person and the right fit, with the right numbers. So can you give some clarity on what that means in your mind?

Therese:
That's a great question. And I'm going to come at it from a couple of different ways. So for the purpose of our conversation today, I'd like to talk about this notion that I talk about optimization with a capital O, and that is using big data science to mathematically optimize staffing. Oftentimes you hear optimization talked about, and I'm going to call it with the small O, and that is that we are optimizing as defined as making something better, right? So we optimize our personal budgets. I optimize my time off by choosing great places to travel to, that's optimization with a small O. Exactly to your point, and I think it's an important thing for us to talk about. Oftentimes at the executive table is a chief nurse with our colleagues, our finance colleagues and our operational colleagues. We all use the term staffing, but every single person, depending on our organizational perspective has a different operating definition in their own head in terms of what staffing means.

Therese:
So if I were the CFO, oftentimes the notion of staffing means I've given three North, a budget of vacs. When I close out the month and the average is that budget of X, I'm a happy camper, we have met my staffing goals. On the other hand, when you talk about staffing in your shared governance groups, with groups of clinical staff, frontline caregivers, here's how they might define staffing. And that is as a leader, you talk to me and you told me what my model of care was going to look like. And that model of care could have been based on targets, it's obviously based on patient needs. And while it has a financial component to it, it distills down to a schedule and an assignment for me. So if you don't meet that goal, in my mind you have broken a promise to me. And that's a very different perspective of staffing. Both of those are very legitimate and we need to be respectful of each of those definitions.

Therese:
So when I talk about the notion of staffing, I talk about it as four very distinct, but highly interrelated processes. And they are as follows. The first one is the budget, the second is the scheduling system and methodology. The third system, which is a unique system, is deployment. So what happens in that 48 or 24 hours before the start of a shift. And then the fourth really important piece is assignment. And assignment really requires a clinical leadership perspective in the moment at the beginning of that work period, to understand whether it's qualitatively quantitatively, or hopefully a combination of both, what's going on with patients on the unit, in the department at that time. And I am as a leader matching the right nurse with that correct assignment, to optimize both, if you will.

Therese:
So just to lay the groundwork to the point that you made around, there are many different perspectives on this issue. So what led me to this work is the fact that having been introduced to this big math, if you will, by scientists outside of nursing, I began to realize and be introduced to case studies where logistics oriented industries were using incredible, robust math and problem solving to fix these logistics problems. So it led me down a path to really re situate staffing and scheduling and all of those processes that I described as a logistics problem. And the minute that I did that, a world of better mathematical approaches opened up to me. I ask myself the question and I often talk about this, how was it that the military, some of the best logisticians in the world can deploy hundreds of thousands of troops all over the globe, clothe them, feed them, get transportation to them, get healthcare to them. And I can't get the right nurse on three North for the night shift tonight. What are they doing?

Therese:
Optimization is a methodology or a mathematical approach, that's actually in the scale of mathematical equations, pretty new math. It was invented in the World War II by a group of interdisciplinary scientists, military experts, engineers, psychologists, and so forth. And it is a rare approach, where you start with the end in mind and work backwards from there. So it requires that you develop the business outcomes that you intend to optimize. And as it relates to staffing, the three indicators that I optimize to is best coverage, that best coverage is defined by the department, that could be defined by staffing committees, in some parts of the country that could be defined by a mandated staffing ratio. But you need to understand what does my staffing model that I'm hoping to achieve look like? Second criteria is that how do I achieve that model at the lowest cost?

Therese:
And then the third indicator that I look at, is staff satisfaction with the entire process. Because as you and I know we can have the best looking schedule on an Excel spreadsheet, however, if it undermines engagement, if it undermines the staff's ability to care for patients, we haven't achieved anything, even though it may be best coverage at the lowest cost. So if in your mind's eye, when I talk about developing these models, a really good way to envision it is, even your mind's eye, picture a great big Rubik's cube. And each of those little boxes, and there's thousands of those little boxes on the cube, becomes either a constraint or a variable in the model. So one of the first things that I look at is, and we all, as nurses, know this intuitively, we still build budgets based on a methodology developed in 1964 when we invented Medicare, and that is the midnight census.

Dan:
Yep, the dreaded midnight census.

Therese:
The dreaded midnight census. So just a quick point of interest. The reason why midnight was looked at, is when Medicare was invented, this was back in the days where there were massive mainframe computers in the basements of all of our hospitals. They took up the sizable portion of most lower levels, and they needed a marker in time to drop a bill to Medicare. So as some of you may recall way back when with these big mainframe computers, you had to shut them down for a couple of hours every evening to let them cool down and do preventative maintenance. And typically, the time to shut down those computers was in the middle of the night when there was less activity. So it was not unusual for hospitals to shut down the main frames between midnight and 2.00 Or 3.00 or 4.00 in the morning. So they needed to mark their day at midnight so they could drop those bills. Well, that was 1964, it's now 2020, and we're still doing the same thing. Right?

Dan:
That's health care for you.

Therese:
Indeed. So the good news is there is a bit of a better way to come at this. So we would suggest, and as we create these optimization models, that we look at actual patients in beds at the hourly level. And I would argue that in very transactional quickly moving departments like ERs, ORs, and so forth, that you can actually, based on our access to big data, get those numbers at the 15 minute level. So we have found that when we look at three years worth of demand data at the hourly level, that it gives us about 95% reliability in terms of being able to predict that volume in subsequent time periods. So if you, in your minds, I can imagine that the demand on three North at 1:00 PM on a Tuesday is one of the boxes on the Rubik's cube, 2:00 PM on Wednesday, another box on the Rubik's cube, then the critical piece, which we have tended to ignore in how we create our budgets and our staffing models, is to quantify our various work rules.

Therese:
So what I mean by a work rule are things like, do the nurses on this unit work every other, or every third weekend? We intuitively know you'd need a very different configuration of staff when you're working every other versus every third weekend. How many consecutive 12 hour shifts can a nurse work on this unit? So all of those work rules could actually be quantified mathematically, and they are situated as the second layer of those boxes on the Rubik's cube. Then the third layer to fill in the rest of those boxes are the desired staffing model for this unit. So in the State of California, that might look like ratio and no matter what our model is, it always distills down to that. So you've now got a Rubik's cube with all of these little boxes filled, we take that Rubik's cube and drop it on top of what's called a math solver, which essentially spins the cube, not unlike the Rubik's cube, about 10,000 times a unit to produce the optimal core staff for the unit down to the actual configuration of FTEs.

Therese:
And the cool thing about linear programming, it allows you to look three dimensionally at a problem. So if you have a nurse in a flexible pool, that can actually work across three North, four North and five North, we can take the Rubik's cubes from all three of those units, understand how demand behaves across all three units collectively, and use that to rightsize and configure the float pool. Then in organizations that are looking for in the enterprise wide solution, you can actually stack the cubes, for example, for all of the med surg units and understand how demand behaves, and then use that to rightsize and configure an enterprise pool.

Dan:
I think you've hit it on the head. I've always been intrigued by the way that we've staffed in the past, which is, let's get a pool of 150 nurses, hire them into a medical surgical service line and then randomly assign them to days and shifts with some preference, and then hope that those skillsets match up with the patient population that happens to be in the hospital that day. And it seems like a game of chance almost. And I know we didn't have the data in the past, but now with electronic medical records and competency assessments, all this stuff, we have data on every end of patient needs, nurse competencies, all the variables you talked about, and we're still not optimizing to it. So I love that you described this way to actually get to that point. And ultimately, I mean, I think, and I'd love your opinion on this, could we actually get rid of service lines, which I think are another archaic model of the past, because you have to have these air control structures and billing units.

Dan:
But now that care moves across all of those service lines in some form, could you actually intimately match nurse's skillset to patient needs and not even call it for South? It's like, you're a nurse in this hospital and we're going to put you with the patients that need your skill set at this time. Is that something that we'll get to it at any point?

Therese:
That is a great question. And actually I am currently working with a large academic medical center in the Midwest that has both a large medical school, school of pharmacy and college of nursing. And we are actually exploring, have a couple of graduate students working with us on this project to understand, is there a role for a nurse hospitalist that is not necessarily, exactly to your point, attached to either a place necessarily or a particular segment of care. But we're actually looking at some of the characteristics that have made really highly functioning hospitalists models, so very effective in organizations. And as you know, hospital medicine is one of the fastest growing specialties in healthcare these days. And we're beginning to find it pretty intriguing. Is there a corollary? It may look a little bit different because obviously our role is different, our perspective with patient care is collaborative with the medical staff, but we each come at it differently, but they're actually looking at doing exactly as you described.

Dan:
That's amazing. And I would love to see the outcome of that work because yeah, I think that ends up being the holy grail of patient care is the right skill set, right time, all that kind of stuff. And it's amazingly complex, which is why it hasn't been done now. But I think with the datasets and the big data analytics we can do now, we're getting closer. We hear from nurse managers all the time, they spend upwards of 2100 hours a year just focused on staffing a unit and they may not have the supercomputers and the big data stuff at their fingertips. What are some low tech ways that nurse managers might conceptualize workforce to better optimize in the short term before they can maybe launch a big project around it?

Therese:
Probably four things come to mind. So given the fact that we are significantly all on some version of an electronic medical record, there is encountered data within that medical record that reports can easily be written to get you that data at the hourly level. And we find, and for our work, we work closely with organizations, decision support teams, in order to write the routines to get us those data. So it is easy that is data that said everyone's fingertips to be able to look at demand patterns at the hourly level. It also is not a stretch to then use those data to help you understand what variation looks like by not just hour of the day, which can be displayed in a heat map, but to look at, sort of checking our intuition in terms of demand variation by day of the week, month of the year and quarterly. So that's one thing that we can easily do.

Therese:
The second is to understand the impact of work rules on scheduling. So oftentimes we'll deploy various work roles only looking at it from the perspective of staff satisfaction perhaps, without understanding the impact of those work rules, both on the finances and on our ability to create schedule models. So that's the other thing, is to begin to understand that scheduling is very complex in healthcare and in large part, because of those work rules. The other is to use those data to go back to our lean thinking. We've all been educated in lean process improvement, where one of the characteristics of a lean process is that it's a pull process. That's quite an opposition, and I would argue it leads to a lot of the shift based chaos related to staffing. And that is because we "never want to be short".

Therese:
What we end up doing is scheduling our units to capacity, and then we push staff away. So we wait until two hours before the start of the shift and then we float nurses to other units, huge dissatisfier. We send people home, low census, even a larger dissatisfier, or we complicate our world even further, we keep that hamster wheel powered up by calling a nurse and saying, "I don't need you today, would you stay home today and come in tomorrow?" And I would argue what we've done is we've kicked the problem down the road by 24 hours. So we need to create systems and in large part, those systems are going to require highly qualified, robust float pools to help us resolve the poll issue, to create those polls systems. I think along with this notion of the nurse as a hospitalist as an actual specialty, one of the most exciting things that I am seeing on the horizon, particularly with our millennial and gen Z nurses, is the evolution of the float pool or flexible pool, or however you may want to think about that.

Dan:
Yeah. I think that's what we're seeing as well. I mean, trusted health is right in that spot. And so we're seeing a big demand too, and we're hearing it from our partners as well. They're saying, we're hiring new grads, but all of them after they get that year or two of experience, want to get on the road and have more flexibility in their schedule. They don't want to be tied into a unit necessarily if they want to potentially do float within a system or they want to actually go off and do contract travel work or per diem or whatever. I think that's definitely a workforce trend that is going to continue to grow. And what we've said is we're telling hospitals to their face, you'll never, ever be fully staffed ever again. And so stop trying to hire your full cohort of full time employees.

Dan:
Number one, it's not cost effective and you probably aren't able to afford all those position openings right now. But you have to come in with a more flexible on demand model where you have a core staff, like you said, but then it's supplemented with skillsets or people or whatever, based on demand and the patient needs. And hospitals haven't got there yet, but I think they're starting to see that it's almost impossible to fully staff a unit anymore. It's not going to work and so to have these other models of float and flex pools, those are going to be the future.

Therese:
Dan, I could not agree with you more and my work and my scholarship suggested exactly the same thing. I've done qualitative interviews with about 250 float pool nurses over the course of about the last 18 months and in different parts of the country. And on the West Coast, in East Coast, a number in the Midwest and I am finding exactly in support of your point, some fascinating and consistent trends. So I asked the question, "Did you take this position in the float pool because you were waiting for a day position to open up on three North, or was this a deliberate career choice?" Overwhelmingly they say a deliberate career choice. I asked the question, "Why did you choose this over a core position on a nursing unit?" They often talk about and very consistently talk about, I enjoy the adventure of going from unit to unit or hospital to hospital in a large enterprise.

Therese:
And this has given me an opportunity to build my career in ways that being on a single unit would not necessarily permit. I believe it is the result of that, that I am doing a lot of work with organizations around helping them better brand their float pools. For example, worked with an organization in California that branded their float pool, the hot chats. And for those of you who are Californians obviously you know that the hot shots are the nickname of the jumpers who are incredibly skilled, incredibly brave firefighters, that are parachuted into these dangerous, very critical situations. And they've likened the fact that their highly qualified float pool was exactly that. So that's how they decided to brand their pool. Worked with another organization in the Chicago area that had this nautical theme, if you will, around staffing and scheduling. And they looked at their float pool as the Navy Seals of nurses, the folks that are highly qualified, that are placed in positions that require that really advanced skillset, so they made a decision to call their float pool, the crew.

Therese:
I think this notion as I think about designing the float pool of the future, as one of my colleagues says, this is not your mother's float pool, that you must have the millennials and gen Z is on your committees and your task forces, because they will tell you how these float pools need to work.

Dan:
Yeah, I couldn't agree more. And one thing I've been talking about as well as there's been a stigma from the past around travel nursing, as a way to skirt the system, or maybe be less quality. But what we're seeing is these are nurses that have worked at every major academic medical center in the country and are some of the most prepared nurses on the planet. And we've been talking with chief nurses about it, and then we say, "How do you engage with your travel nurses? Do you talk to them? Do you get feedback?" And they say, "Not really, it hasn't been something we've done." And I said, "They're a source of innovation." And these float pool nurses are the same thing, they see everything in your hospital. You want to lead innovation, talk to these people. Have seen every single unit, the variation, the issues, the best practices, the worst practices, and how do we harness that energy, even more to create change inside the system. I think there's a lot of exciting opportunities there.

Therese:
I couldn't agree with you more.

Dan:
I wanted to dive in really quickly into what you've seen lately with COVID. I don't know, optimization doesn't seem like a word that meshes well with pandemic, but I think what we've seen is that it's shown the cracks in the system and it's widened those and so made them more visible for us to maybe go in and fix. So I'd love to hear how COVID-19 the pandemic, has accelerated or changed the work that you were doing pre pandemic.

Therese:
Much of the COVID either mid or in some community post COVID strategic thinking right now is around the fact that, where in the past we could look at three years of data and be able to predict what the next 12 months might look like. We need to take those models now and create various scenarios with what happens where we saw our volume dropped precipitously. What happens if volume only comes back 10%, 20%, 30%. What we are finding is that many organizations don't have the mathematical modeling capability to do those various scenarios. So that has consumed much of my work. I am also working a case study of a seven hospital system in the Northeast that I'm working with, had started to do some really deliberate planning around an enterprise wide float pool prior to the start of COVID. And they've realized that having this really well trained, highly competent group of nurses that they had already experienced moving around the system, actually saved them during COVID.

Therese:
That they could deploy staff, their version of the hot shots in a way all those systems and processes had already been worked out. So I believe that we're going to see a larger shift of our workforce into those flexible categories, including contract travel nurses as well, who are a critical portion of that workforce. I think we're going to see more and more of that activity going forward. I think we're realizing that to be very flatfooted and have too much of our staff baked into the core, has left organizations really vulnerable during the pandemic.

Dan:
I agree with that as well. And that's something we've seen. We've seen interesting surges in requests for nurses. Right now, as we speak in Arizona there's requests for upwards of 500 nurses to enter the State. And what we learned from New York is those models may not be as accurate as they could be. And so we're going in a little toe at a time to test it out, to make sure that we're not over delivering and then these nurses get canceled. And a lot of the upset pieces that came along with the first wave of things. And yeah, I think the more we can help hospitals get access to the data so that they can make evidence-based leadership decisions, the better off the industry is going to be. And ultimately nursing as a profession will be happier because I think a lot of the great comes from the way that work is scheduled and occurs on old models. So I'm excited about that.

Dan:
So wrapping up here, one of the things we like to do is hand off information to the next shift, that's why we've named this The Handoff. And Therese, what is one or two key points that you'd want to hand off to healthcare leaders as they think about workforce optimization staffing and the future of work?

Therese:
I think a couple of key learnings for me is the data's out there. These data that are going to allow us to a point that you made earlier, Dan, to think about things in a very holistic way, those data exist. As nurses, we have to become quant oriented, so that we don't necessarily are the ones that are extracting those data, but we need the ability to frame the issue and frame the problem and work with our decision support colleagues, to be able to extract those data. I might also add to that, that we can't schedule our way out of this problem, we've been trying that. There are wonderful scheduling approaches out there, yet we still continue to experience the same situation. I think we need to understand that this is a complex problem, it's going to take a complex holistic solution.

Therese:
And then the third thing I would say is we need to engage our emerging leaders in this process and our new workforce, our millennial and gen Z workforce. They have much to teach us about what the workforce of the future needs to look like and we need to marry that future with analytics.

Dan:
I totally agree. We need data and we need to look at the signposts of the future of work and what the next generation of nurses want out of their career and help build the profession that keeps them there. So, Therese, thank you so much for being on the show today. Where can we find you? Where's the best place to get ahold of you if we want to dive more or seek your services, or even just connect and share ideas?

Therese:
So I can be reached on the website, Kaufmanhall.com, K-A-U-F-M-A-N hall, H-A-L-L.com. And you'll be able to find some thought leadership pieces on workforce optimization there and be able to reach me electronically as well.

Dan:
Love it. Thank you so much for your insights today. We may need to dive more there. I know there's a lot more we can talk about, so we'll have to see if we can do a part two on this, but really appreciate your time. And we'll put all those contact information into the show notes so people can reach out.

Therese:
Great. Thank you so much. It was a pleasure, Dan.

Description

Staffing has long been one of the Achilles heels of the nursing profession, confounding nurse managers and nurse leaders alike, and taking up an inordinate amount of time. Our guest for this episode is an expert on the topic and has taken her cues from other industries to treat this as a data science problem. 

Early in her career, Therese Fitzpatrick was a chief nursing officer for various hospitals and health systems in the midwest before taking a turn as an entrepreneur, receiving her PhD in sociology and becoming a professor in the University of Illinois at Chicago’s School of Public Health. 

Today she is an executive at consulting firm Kaufman Hall, where she helps hospitals across the country assess their clinical and operational performance, and optimize their staffing.

In this episode Therese walks us through how she applies big data to the problems of staffing and workforce optimization, some low-tech ways that nurse managers can approach scheduling, and how COVID-19 will impact the evolution of the nursing workforce. Therese believes that robust float pools driven by Millenial and Gen Z nurses are the future and encourages nurse leaders to include them in conversations around staffing and workforce.

Links to recommended reading: 

Transcript

Dan:
Welcome to the show Therese.

Therese:
Thank you so much. I'm happy to be here with you, Dan.

Dan:
Thank you. Therese, why don't you give us a quick overview of your history and where you're focused now, related to all the cool things you're working on?

Therese:
Well, I have been system chief nurse for a number of years, and in that role, had an opportunity to meet some wonderful, incredibly wicked smart, operations for search scientists who introduced me to the whole notion of optimization and linear programming. In fact, one of those leaders is currently a professor of math and business at Stanford University. And what I realized was that the really pernicious problems related to staffing that I had spent a career as an executive trying to solve, other industries had figured out a way to do that. So I actually moved my career and became an entrepreneur and started two companies around workforce optimization. In each of those companies, my partners were operations research scientists, had tons of fun doing that and then got a PhD in sociology. So my midlife crisis was to become a sociologist and figure out, our undergrad degrees, as nurses, we focus on body systems and individuals and their families as healthcare related systems. At the graduate level, I focused on health systems and how do all the pieces of hospitals work?

Therese:
So I figured it was time to figure out where healthcare fits into the larger social structure. That led to going to work for a large multinational company, doing optimization work across all disciplines, across the world, actually. And along that path took a position in faculty, in the college of nursing at the doctoral level, at the University of Illinois in Chicago, and subsequently moved to the School of Public Health where I teach in the MHA program. I am currently working with a financial consulting company as one of the few nurses within a company full of all sorts of CFOs and financial analysts, looking at healthcare, doing lots of merger and acquisition work and building workforce optimization within that organization.

Dan:
That's awesome. And I love that you're a nurse in a nontraditional setting, which I know a lot of listeners are and then that's been my career path as well. So one thing that we know is that staffing is the one thing on nurse leaders minds all the time. I think it's staffing, quality and safety, probably, maybe in that order, depending on the day, it seems like. But we hear a lot around workforce optimization, can you talk a little bit more about that? I think there's probably some dogma around that. When you ask a staff nurse, workforce optimization means higher ratios, maybe less staff, work harder kind of thing. On the hospital side, workforce optimization is maybe right person and the right fit, with the right numbers. So can you give some clarity on what that means in your mind?

Therese:
That's a great question. And I'm going to come at it from a couple of different ways. So for the purpose of our conversation today, I'd like to talk about this notion that I talk about optimization with a capital O, and that is using big data science to mathematically optimize staffing. Oftentimes you hear optimization talked about, and I'm going to call it with the small O, and that is that we are optimizing as defined as making something better, right? So we optimize our personal budgets. I optimize my time off by choosing great places to travel to, that's optimization with a small O. Exactly to your point, and I think it's an important thing for us to talk about. Oftentimes at the executive table is a chief nurse with our colleagues, our finance colleagues and our operational colleagues. We all use the term staffing, but every single person, depending on our organizational perspective has a different operating definition in their own head in terms of what staffing means.

Therese:
So if I were the CFO, oftentimes the notion of staffing means I've given three North, a budget of vacs. When I close out the month and the average is that budget of X, I'm a happy camper, we have met my staffing goals. On the other hand, when you talk about staffing in your shared governance groups, with groups of clinical staff, frontline caregivers, here's how they might define staffing. And that is as a leader, you talk to me and you told me what my model of care was going to look like. And that model of care could have been based on targets, it's obviously based on patient needs. And while it has a financial component to it, it distills down to a schedule and an assignment for me. So if you don't meet that goal, in my mind you have broken a promise to me. And that's a very different perspective of staffing. Both of those are very legitimate and we need to be respectful of each of those definitions.

Therese:
So when I talk about the notion of staffing, I talk about it as four very distinct, but highly interrelated processes. And they are as follows. The first one is the budget, the second is the scheduling system and methodology. The third system, which is a unique system, is deployment. So what happens in that 48 or 24 hours before the start of a shift. And then the fourth really important piece is assignment. And assignment really requires a clinical leadership perspective in the moment at the beginning of that work period, to understand whether it's qualitatively quantitatively, or hopefully a combination of both, what's going on with patients on the unit, in the department at that time. And I am as a leader matching the right nurse with that correct assignment, to optimize both, if you will.

Therese:
So just to lay the groundwork to the point that you made around, there are many different perspectives on this issue. So what led me to this work is the fact that having been introduced to this big math, if you will, by scientists outside of nursing, I began to realize and be introduced to case studies where logistics oriented industries were using incredible, robust math and problem solving to fix these logistics problems. So it led me down a path to really re situate staffing and scheduling and all of those processes that I described as a logistics problem. And the minute that I did that, a world of better mathematical approaches opened up to me. I ask myself the question and I often talk about this, how was it that the military, some of the best logisticians in the world can deploy hundreds of thousands of troops all over the globe, clothe them, feed them, get transportation to them, get healthcare to them. And I can't get the right nurse on three North for the night shift tonight. What are they doing?

Therese:
Optimization is a methodology or a mathematical approach, that's actually in the scale of mathematical equations, pretty new math. It was invented in the World War II by a group of interdisciplinary scientists, military experts, engineers, psychologists, and so forth. And it is a rare approach, where you start with the end in mind and work backwards from there. So it requires that you develop the business outcomes that you intend to optimize. And as it relates to staffing, the three indicators that I optimize to is best coverage, that best coverage is defined by the department, that could be defined by staffing committees, in some parts of the country that could be defined by a mandated staffing ratio. But you need to understand what does my staffing model that I'm hoping to achieve look like? Second criteria is that how do I achieve that model at the lowest cost?

Therese:
And then the third indicator that I look at, is staff satisfaction with the entire process. Because as you and I know we can have the best looking schedule on an Excel spreadsheet, however, if it undermines engagement, if it undermines the staff's ability to care for patients, we haven't achieved anything, even though it may be best coverage at the lowest cost. So if in your mind's eye, when I talk about developing these models, a really good way to envision it is, even your mind's eye, picture a great big Rubik's cube. And each of those little boxes, and there's thousands of those little boxes on the cube, becomes either a constraint or a variable in the model. So one of the first things that I look at is, and we all, as nurses, know this intuitively, we still build budgets based on a methodology developed in 1964 when we invented Medicare, and that is the midnight census.

Dan:
Yep, the dreaded midnight census.

Therese:
The dreaded midnight census. So just a quick point of interest. The reason why midnight was looked at, is when Medicare was invented, this was back in the days where there were massive mainframe computers in the basements of all of our hospitals. They took up the sizable portion of most lower levels, and they needed a marker in time to drop a bill to Medicare. So as some of you may recall way back when with these big mainframe computers, you had to shut them down for a couple of hours every evening to let them cool down and do preventative maintenance. And typically, the time to shut down those computers was in the middle of the night when there was less activity. So it was not unusual for hospitals to shut down the main frames between midnight and 2.00 Or 3.00 or 4.00 in the morning. So they needed to mark their day at midnight so they could drop those bills. Well, that was 1964, it's now 2020, and we're still doing the same thing. Right?

Dan:
That's health care for you.

Therese:
Indeed. So the good news is there is a bit of a better way to come at this. So we would suggest, and as we create these optimization models, that we look at actual patients in beds at the hourly level. And I would argue that in very transactional quickly moving departments like ERs, ORs, and so forth, that you can actually, based on our access to big data, get those numbers at the 15 minute level. So we have found that when we look at three years worth of demand data at the hourly level, that it gives us about 95% reliability in terms of being able to predict that volume in subsequent time periods. So if you, in your minds, I can imagine that the demand on three North at 1:00 PM on a Tuesday is one of the boxes on the Rubik's cube, 2:00 PM on Wednesday, another box on the Rubik's cube, then the critical piece, which we have tended to ignore in how we create our budgets and our staffing models, is to quantify our various work rules.

Therese:
So what I mean by a work rule are things like, do the nurses on this unit work every other, or every third weekend? We intuitively know you'd need a very different configuration of staff when you're working every other versus every third weekend. How many consecutive 12 hour shifts can a nurse work on this unit? So all of those work rules could actually be quantified mathematically, and they are situated as the second layer of those boxes on the Rubik's cube. Then the third layer to fill in the rest of those boxes are the desired staffing model for this unit. So in the State of California, that might look like ratio and no matter what our model is, it always distills down to that. So you've now got a Rubik's cube with all of these little boxes filled, we take that Rubik's cube and drop it on top of what's called a math solver, which essentially spins the cube, not unlike the Rubik's cube, about 10,000 times a unit to produce the optimal core staff for the unit down to the actual configuration of FTEs.

Therese:
And the cool thing about linear programming, it allows you to look three dimensionally at a problem. So if you have a nurse in a flexible pool, that can actually work across three North, four North and five North, we can take the Rubik's cubes from all three of those units, understand how demand behaves across all three units collectively, and use that to rightsize and configure the float pool. Then in organizations that are looking for in the enterprise wide solution, you can actually stack the cubes, for example, for all of the med surg units and understand how demand behaves, and then use that to rightsize and configure an enterprise pool.

Dan:
I think you've hit it on the head. I've always been intrigued by the way that we've staffed in the past, which is, let's get a pool of 150 nurses, hire them into a medical surgical service line and then randomly assign them to days and shifts with some preference, and then hope that those skillsets match up with the patient population that happens to be in the hospital that day. And it seems like a game of chance almost. And I know we didn't have the data in the past, but now with electronic medical records and competency assessments, all this stuff, we have data on every end of patient needs, nurse competencies, all the variables you talked about, and we're still not optimizing to it. So I love that you described this way to actually get to that point. And ultimately, I mean, I think, and I'd love your opinion on this, could we actually get rid of service lines, which I think are another archaic model of the past, because you have to have these air control structures and billing units.

Dan:
But now that care moves across all of those service lines in some form, could you actually intimately match nurse's skillset to patient needs and not even call it for South? It's like, you're a nurse in this hospital and we're going to put you with the patients that need your skill set at this time. Is that something that we'll get to it at any point?

Therese:
That is a great question. And actually I am currently working with a large academic medical center in the Midwest that has both a large medical school, school of pharmacy and college of nursing. And we are actually exploring, have a couple of graduate students working with us on this project to understand, is there a role for a nurse hospitalist that is not necessarily, exactly to your point, attached to either a place necessarily or a particular segment of care. But we're actually looking at some of the characteristics that have made really highly functioning hospitalists models, so very effective in organizations. And as you know, hospital medicine is one of the fastest growing specialties in healthcare these days. And we're beginning to find it pretty intriguing. Is there a corollary? It may look a little bit different because obviously our role is different, our perspective with patient care is collaborative with the medical staff, but we each come at it differently, but they're actually looking at doing exactly as you described.

Dan:
That's amazing. And I would love to see the outcome of that work because yeah, I think that ends up being the holy grail of patient care is the right skill set, right time, all that kind of stuff. And it's amazingly complex, which is why it hasn't been done now. But I think with the datasets and the big data analytics we can do now, we're getting closer. We hear from nurse managers all the time, they spend upwards of 2100 hours a year just focused on staffing a unit and they may not have the supercomputers and the big data stuff at their fingertips. What are some low tech ways that nurse managers might conceptualize workforce to better optimize in the short term before they can maybe launch a big project around it?

Therese:
Probably four things come to mind. So given the fact that we are significantly all on some version of an electronic medical record, there is encountered data within that medical record that reports can easily be written to get you that data at the hourly level. And we find, and for our work, we work closely with organizations, decision support teams, in order to write the routines to get us those data. So it is easy that is data that said everyone's fingertips to be able to look at demand patterns at the hourly level. It also is not a stretch to then use those data to help you understand what variation looks like by not just hour of the day, which can be displayed in a heat map, but to look at, sort of checking our intuition in terms of demand variation by day of the week, month of the year and quarterly. So that's one thing that we can easily do.

Therese:
The second is to understand the impact of work rules on scheduling. So oftentimes we'll deploy various work roles only looking at it from the perspective of staff satisfaction perhaps, without understanding the impact of those work rules, both on the finances and on our ability to create schedule models. So that's the other thing, is to begin to understand that scheduling is very complex in healthcare and in large part, because of those work rules. The other is to use those data to go back to our lean thinking. We've all been educated in lean process improvement, where one of the characteristics of a lean process is that it's a pull process. That's quite an opposition, and I would argue it leads to a lot of the shift based chaos related to staffing. And that is because we "never want to be short".

Therese:
What we end up doing is scheduling our units to capacity, and then we push staff away. So we wait until two hours before the start of the shift and then we float nurses to other units, huge dissatisfier. We send people home, low census, even a larger dissatisfier, or we complicate our world even further, we keep that hamster wheel powered up by calling a nurse and saying, "I don't need you today, would you stay home today and come in tomorrow?" And I would argue what we've done is we've kicked the problem down the road by 24 hours. So we need to create systems and in large part, those systems are going to require highly qualified, robust float pools to help us resolve the poll issue, to create those polls systems. I think along with this notion of the nurse as a hospitalist as an actual specialty, one of the most exciting things that I am seeing on the horizon, particularly with our millennial and gen Z nurses, is the evolution of the float pool or flexible pool, or however you may want to think about that.

Dan:
Yeah. I think that's what we're seeing as well. I mean, trusted health is right in that spot. And so we're seeing a big demand too, and we're hearing it from our partners as well. They're saying, we're hiring new grads, but all of them after they get that year or two of experience, want to get on the road and have more flexibility in their schedule. They don't want to be tied into a unit necessarily if they want to potentially do float within a system or they want to actually go off and do contract travel work or per diem or whatever. I think that's definitely a workforce trend that is going to continue to grow. And what we've said is we're telling hospitals to their face, you'll never, ever be fully staffed ever again. And so stop trying to hire your full cohort of full time employees.

Dan:
Number one, it's not cost effective and you probably aren't able to afford all those position openings right now. But you have to come in with a more flexible on demand model where you have a core staff, like you said, but then it's supplemented with skillsets or people or whatever, based on demand and the patient needs. And hospitals haven't got there yet, but I think they're starting to see that it's almost impossible to fully staff a unit anymore. It's not going to work and so to have these other models of float and flex pools, those are going to be the future.

Therese:
Dan, I could not agree with you more and my work and my scholarship suggested exactly the same thing. I've done qualitative interviews with about 250 float pool nurses over the course of about the last 18 months and in different parts of the country. And on the West Coast, in East Coast, a number in the Midwest and I am finding exactly in support of your point, some fascinating and consistent trends. So I asked the question, "Did you take this position in the float pool because you were waiting for a day position to open up on three North, or was this a deliberate career choice?" Overwhelmingly they say a deliberate career choice. I asked the question, "Why did you choose this over a core position on a nursing unit?" They often talk about and very consistently talk about, I enjoy the adventure of going from unit to unit or hospital to hospital in a large enterprise.

Therese:
And this has given me an opportunity to build my career in ways that being on a single unit would not necessarily permit. I believe it is the result of that, that I am doing a lot of work with organizations around helping them better brand their float pools. For example, worked with an organization in California that branded their float pool, the hot chats. And for those of you who are Californians obviously you know that the hot shots are the nickname of the jumpers who are incredibly skilled, incredibly brave firefighters, that are parachuted into these dangerous, very critical situations. And they've likened the fact that their highly qualified float pool was exactly that. So that's how they decided to brand their pool. Worked with another organization in the Chicago area that had this nautical theme, if you will, around staffing and scheduling. And they looked at their float pool as the Navy Seals of nurses, the folks that are highly qualified, that are placed in positions that require that really advanced skillset, so they made a decision to call their float pool, the crew.

Therese:
I think this notion as I think about designing the float pool of the future, as one of my colleagues says, this is not your mother's float pool, that you must have the millennials and gen Z is on your committees and your task forces, because they will tell you how these float pools need to work.

Dan:
Yeah, I couldn't agree more. And one thing I've been talking about as well as there's been a stigma from the past around travel nursing, as a way to skirt the system, or maybe be less quality. But what we're seeing is these are nurses that have worked at every major academic medical center in the country and are some of the most prepared nurses on the planet. And we've been talking with chief nurses about it, and then we say, "How do you engage with your travel nurses? Do you talk to them? Do you get feedback?" And they say, "Not really, it hasn't been something we've done." And I said, "They're a source of innovation." And these float pool nurses are the same thing, they see everything in your hospital. You want to lead innovation, talk to these people. Have seen every single unit, the variation, the issues, the best practices, the worst practices, and how do we harness that energy, even more to create change inside the system. I think there's a lot of exciting opportunities there.

Therese:
I couldn't agree with you more.

Dan:
I wanted to dive in really quickly into what you've seen lately with COVID. I don't know, optimization doesn't seem like a word that meshes well with pandemic, but I think what we've seen is that it's shown the cracks in the system and it's widened those and so made them more visible for us to maybe go in and fix. So I'd love to hear how COVID-19 the pandemic, has accelerated or changed the work that you were doing pre pandemic.

Therese:
Much of the COVID either mid or in some community post COVID strategic thinking right now is around the fact that, where in the past we could look at three years of data and be able to predict what the next 12 months might look like. We need to take those models now and create various scenarios with what happens where we saw our volume dropped precipitously. What happens if volume only comes back 10%, 20%, 30%. What we are finding is that many organizations don't have the mathematical modeling capability to do those various scenarios. So that has consumed much of my work. I am also working a case study of a seven hospital system in the Northeast that I'm working with, had started to do some really deliberate planning around an enterprise wide float pool prior to the start of COVID. And they've realized that having this really well trained, highly competent group of nurses that they had already experienced moving around the system, actually saved them during COVID.

Therese:
That they could deploy staff, their version of the hot shots in a way all those systems and processes had already been worked out. So I believe that we're going to see a larger shift of our workforce into those flexible categories, including contract travel nurses as well, who are a critical portion of that workforce. I think we're going to see more and more of that activity going forward. I think we're realizing that to be very flatfooted and have too much of our staff baked into the core, has left organizations really vulnerable during the pandemic.

Dan:
I agree with that as well. And that's something we've seen. We've seen interesting surges in requests for nurses. Right now, as we speak in Arizona there's requests for upwards of 500 nurses to enter the State. And what we learned from New York is those models may not be as accurate as they could be. And so we're going in a little toe at a time to test it out, to make sure that we're not over delivering and then these nurses get canceled. And a lot of the upset pieces that came along with the first wave of things. And yeah, I think the more we can help hospitals get access to the data so that they can make evidence-based leadership decisions, the better off the industry is going to be. And ultimately nursing as a profession will be happier because I think a lot of the great comes from the way that work is scheduled and occurs on old models. So I'm excited about that.

Dan:
So wrapping up here, one of the things we like to do is hand off information to the next shift, that's why we've named this The Handoff. And Therese, what is one or two key points that you'd want to hand off to healthcare leaders as they think about workforce optimization staffing and the future of work?

Therese:
I think a couple of key learnings for me is the data's out there. These data that are going to allow us to a point that you made earlier, Dan, to think about things in a very holistic way, those data exist. As nurses, we have to become quant oriented, so that we don't necessarily are the ones that are extracting those data, but we need the ability to frame the issue and frame the problem and work with our decision support colleagues, to be able to extract those data. I might also add to that, that we can't schedule our way out of this problem, we've been trying that. There are wonderful scheduling approaches out there, yet we still continue to experience the same situation. I think we need to understand that this is a complex problem, it's going to take a complex holistic solution.

Therese:
And then the third thing I would say is we need to engage our emerging leaders in this process and our new workforce, our millennial and gen Z workforce. They have much to teach us about what the workforce of the future needs to look like and we need to marry that future with analytics.

Dan:
I totally agree. We need data and we need to look at the signposts of the future of work and what the next generation of nurses want out of their career and help build the profession that keeps them there. So, Therese, thank you so much for being on the show today. Where can we find you? Where's the best place to get ahold of you if we want to dive more or seek your services, or even just connect and share ideas?

Therese:
So I can be reached on the website, Kaufmanhall.com, K-A-U-F-M-A-N hall, H-A-L-L.com. And you'll be able to find some thought leadership pieces on workforce optimization there and be able to reach me electronically as well.

Dan:
Love it. Thank you so much for your insights today. We may need to dive more there. I know there's a lot more we can talk about, so we'll have to see if we can do a part two on this, but really appreciate your time. And we'll put all those contact information into the show notes so people can reach out.

Therese:
Great. Thank you so much. It was a pleasure, Dan.

Back to THEHANDOFF