By Sarah Klein
Eugene Litvak, Ph.D., the CEO of the Newton, Mass.–based Institute for Healthcare Optimization (IHO), has spent nearly 20 years working to improve the quality and efficiency of health care services by optimizing patient flow in hospitals and clinics. One of his latest projects involves helping 16 New Jersey hospitals reduce variability in occupancy rates, among other strategies, so that staffing resources are consistently aligned with demand. Quality Matters asked Litvak, who trained as an operations researcher, what he has learned along the way.
Quality Matters: What drew you to focus on patient flow in the first place?
Litvak: I became interested in it and health care efficiency more generally after being warned by many colleagues in the area of operations management not to get into health care. They told me "these people don't care about cost, they don't care about efficiency, and they would not appreciate any improvements in these directions." To me, that was a red flag. I wanted to know how come—especially because the industry is one of the biggest in the world. Looking at hospital occupancy rates was an obvious place to start as staffing hospital wards is one of the main hospital expenses. My colleagues and I looked at hospitals in the U.S. and around the world—Australia, Europe, the U.K., Japan, and Africa—and found they had the same pattern of peaks and valleys in bed occupancy during the week, with the difference between two neighboring days ranging from 25 percent occupancy to 80 percent. That's not including holidays and weekends. It's practically impossible to staff with such variability, especially given current workforce shortages.
Quality Matters: So why was there so much variation?
Litvak: When we asked hospital administrators how they explained it, most seemed to believe that the peaks and valleys are given by God: They say things like "We can't control variation as it's determined by patient demand," "It's just the emergency department [ED]," or "The payers just don't give us enough funds to staff to peaks in bed occupancy." But we found that surprisingly the main cause of the variation is elective admissions. I'm not exaggerating. At many hospitals, it is easier to predict when somebody will break his or her leg and come to the emergency department than it is to predict when elective procedures will take place. For example, on any given weekday you can ask the emergency department manager how many patients they will see on the same day of the week five weeks from now and he or she can tell you fairly accurately. In the operating room [OR], typically, nobody can answer this question because it's related to the number of elective surgeries that will be performed on that day, and that's provider-centric. Quality Matters: How does the variability affect quality of care?
Litvak: There are numerous published studies that have documented that lack of adequate staffing, especially during peak periods, is closely correlated with hospital and ED overcrowding, medical errors, readmissions, infections, and mortality.
Quality Matters: Is the solution to control the number of elective surgeries?
Litvak: No. Quite to the contrary, optimal management of patient flow allows hospitals to perform many more surgeries without opening more ORs, adding more beds, or hiring additional personnel. Our methodology has three phases: First, provide separate resources to inpatient and outpatient flows as well as elective and nonelective flows. They need separate staffing, separate physical spaces—including operating rooms as they have conflicting goals. Elective flow would benefit from high utilization of the OR services, while unscheduled flow requires timely access to the OR. Once that's done, it's possible to smooth the flow of the elective cases based on their destination in the hospital, which makes the numbers of patients and nurses per patient on a unit more predictable. This enables hospitals to reduce the financial waste that occurs when there is overstaffing and lower the risk of adverse events that occur when there is understaffing. After these two phases, hospitals can precisely determine how many beds they need to accommodate current and future demand. This is important because the capital cost alone for one additional bed exceeds $1 million.
Quality Matters: Does smoothing the flow of patients from elective surgeries mean extending hours—that is, does it require asking providers to make use of operating rooms on weekends and evenings?
Litvak: No. When we separate emergency and urgent care from elective surgery, elective surgery throughput increases on weekdays. The key is to have the right number of resources—beds, operating rooms, and staff—for each type of flow. What we've seen in hospitals that do this is a remarkable increase in return on investment—from $17,000 per bed to as much as $300,000 per bed annually. I don't know any other intervention that has such outcomes.
Quality Matters: Isn't there a danger that this work will drive up costs and create opportunity to perform more unnecessary elective surgeries?
Litvak: No, for the same reason that any improvement in hospital efficiency should not be viewed as an enabler of additional unnecessary admissions. I think that improvements in health care efficiency and reducing unnecessary procedures are very different issues that should not be viewed as conflicting activities. In addition, there are patients that are waiting for a long time to have much-needed surgery. They are the main beneficiaries of increased hospital efficiency.
Quality Matters: What are some of the challenges you encounter when you attempt to optimize patient flow?
Litvak: It requires changing the culture and this requires changing when surgeons and those in the cardiac catheterization labs work. If you don't make changes, what you see is hospitals typically have peaks on Mondays, Tuesdays, and Wednesdays for elective surgeries. One cardiac surgeon I talked to said, "How dare you tell me when to operate on my patients? Only I am in a position to know their acuity and health status." After I left, I did some analysis and said look, "I'm no longer a Harvard professor. I'm a student and as such would like to learn: what cardiac disease are your patients exposed to that manifests itself every Tuesday?" I call this phenomenon "weekday–related disease."
Quality Matters: You've worked with the Mayo Clinic, Johns Hopkins Medicine, and Cincinnati Children's Hospital, among other hospitals. And you're now working with a group of 16 hospitals in New Jersey collaborating with the New Jersey Hospital Association under the Partnership for Patients program funded by the Centers for Medicare and Medicaid Services. What kind of outcomes have you seen from this work?
Litvak: We've shown that reducing variability directly changes quality of care. We've seen substantial improvements in mortality, length of stay, access to care, and patient-per-nurse staffing ratios coupled with reduced cost in the participating New Jersey hospitals in just 15 months. These outcomes, however, require a lot of work and hospital leadership. There's an investment required for data collection, data cleaning, analyses, and making and implementing recommendations. To get started, many hospitals chose one unit to focus on.
Quality Matters: What would it take to spread this work to more hospitals?
Litvak: I think the recognition that it is difficult if not impossible to achieve other important goals—reducing infections, readmissions, and mortality—without addressing variability may help. Some hospital administrators may also be persuaded of the merits of optimizing patient flow by the financial returns. Even if you are being super conservative and you assume that each U.S. hospital got only 10 percent of the return on investment that Cincinnati Children's got, that would be almost $60 billion reduction in capital costs alone across all hospitals.
Quality Matters: What drew you to focus on patient flow in the first place?
Litvak: I became interested in it and health care efficiency more generally after being warned by many colleagues in the area of operations management not to get into health care. They told me "these people don't care about cost, they don't care about efficiency, and they would not appreciate any improvements in these directions." To me, that was a red flag. I wanted to know how come—especially because the industry is one of the biggest in the world. Looking at hospital occupancy rates was an obvious place to start as staffing hospital wards is one of the main hospital expenses. My colleagues and I looked at hospitals in the U.S. and around the world—Australia, Europe, the U.K., Japan, and Africa—and found they had the same pattern of peaks and valleys in bed occupancy during the week, with the difference between two neighboring days ranging from 25 percent occupancy to 80 percent. That's not including holidays and weekends. It's practically impossible to staff with such variability, especially given current workforce shortages.
Quality Matters: So why was there so much variation?
Litvak: When we asked hospital administrators how they explained it, most seemed to believe that the peaks and valleys are given by God: They say things like "We can't control variation as it's determined by patient demand," "It's just the emergency department [ED]," or "The payers just don't give us enough funds to staff to peaks in bed occupancy." But we found that surprisingly the main cause of the variation is elective admissions. I'm not exaggerating. At many hospitals, it is easier to predict when somebody will break his or her leg and come to the emergency department than it is to predict when elective procedures will take place. For example, on any given weekday you can ask the emergency department manager how many patients they will see on the same day of the week five weeks from now and he or she can tell you fairly accurately. In the operating room [OR], typically, nobody can answer this question because it's related to the number of elective surgeries that will be performed on that day, and that's provider-centric. Quality Matters: How does the variability affect quality of care?
Litvak: There are numerous published studies that have documented that lack of adequate staffing, especially during peak periods, is closely correlated with hospital and ED overcrowding, medical errors, readmissions, infections, and mortality.
Quality Matters: Is the solution to control the number of elective surgeries?
Litvak: No. Quite to the contrary, optimal management of patient flow allows hospitals to perform many more surgeries without opening more ORs, adding more beds, or hiring additional personnel. Our methodology has three phases: First, provide separate resources to inpatient and outpatient flows as well as elective and nonelective flows. They need separate staffing, separate physical spaces—including operating rooms as they have conflicting goals. Elective flow would benefit from high utilization of the OR services, while unscheduled flow requires timely access to the OR. Once that's done, it's possible to smooth the flow of the elective cases based on their destination in the hospital, which makes the numbers of patients and nurses per patient on a unit more predictable. This enables hospitals to reduce the financial waste that occurs when there is overstaffing and lower the risk of adverse events that occur when there is understaffing. After these two phases, hospitals can precisely determine how many beds they need to accommodate current and future demand. This is important because the capital cost alone for one additional bed exceeds $1 million.
Quality Matters: Does smoothing the flow of patients from elective surgeries mean extending hours—that is, does it require asking providers to make use of operating rooms on weekends and evenings?
Litvak: No. When we separate emergency and urgent care from elective surgery, elective surgery throughput increases on weekdays. The key is to have the right number of resources—beds, operating rooms, and staff—for each type of flow. What we've seen in hospitals that do this is a remarkable increase in return on investment—from $17,000 per bed to as much as $300,000 per bed annually. I don't know any other intervention that has such outcomes.
Quality Matters: Isn't there a danger that this work will drive up costs and create opportunity to perform more unnecessary elective surgeries?
Litvak: No, for the same reason that any improvement in hospital efficiency should not be viewed as an enabler of additional unnecessary admissions. I think that improvements in health care efficiency and reducing unnecessary procedures are very different issues that should not be viewed as conflicting activities. In addition, there are patients that are waiting for a long time to have much-needed surgery. They are the main beneficiaries of increased hospital efficiency.
Quality Matters: What are some of the challenges you encounter when you attempt to optimize patient flow?
Litvak: It requires changing the culture and this requires changing when surgeons and those in the cardiac catheterization labs work. If you don't make changes, what you see is hospitals typically have peaks on Mondays, Tuesdays, and Wednesdays for elective surgeries. One cardiac surgeon I talked to said, "How dare you tell me when to operate on my patients? Only I am in a position to know their acuity and health status." After I left, I did some analysis and said look, "I'm no longer a Harvard professor. I'm a student and as such would like to learn: what cardiac disease are your patients exposed to that manifests itself every Tuesday?" I call this phenomenon "weekday–related disease."
Quality Matters: You've worked with the Mayo Clinic, Johns Hopkins Medicine, and Cincinnati Children's Hospital, among other hospitals. And you're now working with a group of 16 hospitals in New Jersey collaborating with the New Jersey Hospital Association under the Partnership for Patients program funded by the Centers for Medicare and Medicaid Services. What kind of outcomes have you seen from this work?
Litvak: We've shown that reducing variability directly changes quality of care. We've seen substantial improvements in mortality, length of stay, access to care, and patient-per-nurse staffing ratios coupled with reduced cost in the participating New Jersey hospitals in just 15 months. These outcomes, however, require a lot of work and hospital leadership. There's an investment required for data collection, data cleaning, analyses, and making and implementing recommendations. To get started, many hospitals chose one unit to focus on.
Quality Matters: What would it take to spread this work to more hospitals?
Litvak: I think the recognition that it is difficult if not impossible to achieve other important goals—reducing infections, readmissions, and mortality—without addressing variability may help. Some hospital administrators may also be persuaded of the merits of optimizing patient flow by the financial returns. Even if you are being super conservative and you assume that each U.S. hospital got only 10 percent of the return on investment that Cincinnati Children's got, that would be almost $60 billion reduction in capital costs alone across all hospitals.