Loading ...

The recent projection that national hospital bed occupancy will reach dangerous shortage levels as soon as 2032 perhaps rang a familiar alarm for those seeking to solve the persistent puzzle that is hospital throughput.

Just a few years ago, fresh off the height of the COVID-19 pandemic, some hospitals looked again to methods that arose alongside the hospitalist model. They go by many similar names (interdisciplinary rounds [IDRs], intradisciplinary rounds, or multidisciplinary rounds [MDRs]), and recently, a pair of teaching hospitals reported new success with MDRs (or, if you prefer, IDRs).

Looking ahead, though, emerging research is pointing toward approaching throughput optimization via indirect methods.

A University of Miami researcher has used an operations perspective to create a mathematical model for optimizing hospitalist workload. It was inspired by a simple and often reported data point: Caseloads among hospitalists can vary from less than 10 to more than 20 patients. So Masoud Kamalahmadi, PhD, an assistant professor of management science at the Miami Herbert Business School, University of Miami, Coral Gables, Florida, created a model for optimizing physician workload to minimize patients’ length of stay and maximize hospital throughput.

“In our field, we love variability. Whenever you see variability, it means that they don’t know what the right answer is,” he said. “If there is variability, then each person is just coming up with their own solution.”

Texas Case Study: MDRs 2.0

The gynecologic oncology inpatient unit at MD Anderson Cancer Center in Houston has about 50 beds and serves both medical and surgical patients.

Formal or informal MDRs have long been a part of the unit, including one iteration a few years ago that was used by every discipline and department in the cancer center.

“We quickly found in 2023 that a structured MDR that is unified or uniform was not equivalent in every department,” said Kiara Whitney, FNP-C, APRN, MSN, inpatient nurse practitioner on the gynecologic oncology team at the time. “We can’t have the same MDR process in surgery that we have in leukemia for our liquid tumors vs for our solid tumors. So then we implemented MDRs 2.0.”

This latest version came alongside a throughput initiative that called for discharge orders to be completed before 9 AM and nurse-provided patient discharge instructions completed by noon. The goal was for a new patient to be coordinated by 2 PM.

To accomplish these goals, the provider team huddled at 7 AM, and a MDR meeting was held at 10 AM. And that’s exactly 10 AM, said Tara Tatum, RN, MBA, an associate director of nursing at MD Anderson Cancer Center.

At first, the meeting took an hour, but now it holds steady at 25-30 minutes.

“When we’re discussing a patient, it should be only about how to get the patient discharged and what is our plan of care. It is not about all the other personal things that are going on with this patient. We have to eliminate those kind of discussions and get really to the core of what we’re doing,” Tatum said. “Yes, it does matter the social things that are going on with the patient, but we keep that at a minimum” 

Rounding at 7 AM due to the nature of being a surgical team set the team up for success (although most patients were medical patients), said Whitney, who applauded Nicole Fleming, MD, her inpatient medical director, for championing the initiative.

“The biggest thing is that we want to ensure that our patients are educated prior to the day of discharge. We don’t want to focus so much on orders in by 9 o’clock. We want to focus on how do we prepare patients before day of discharge for discharge.”

The results were presented at the 2024 ASCO Quality Care Symposium and showed that discharge data accuracy rose 2.9%, there was a 60.5% increase in orders completed before 9 AM, and 20.4% more patients left the hospital before noon.

The structured discharge process helped ensure nothing was overlooked and helped identify key areas uniquely important to the unit, such as including nutrition planning because their patients tend to be at high risk for small bowel obstruction. Tracking readmission, which was prominent on a dashboard the team used, helped incorporate that knowledge into care and planning.

“I’m not saying that you’re going to be looking for that goals-of-care conversation with every patient. But it really just opened up our eyes and we might say, ‘This patient has been admitted a few times. Is this something that we’re doing as providers, like maybe we’re not setting them up for success at home.’ Maybe they’re not having home health coming in. Maybe they lack education on how to take care of a dressing change, and that’s why they keep on coming in with these recurring infections.”

“So I think the MDRs definitely assisted with hospital throughput, but then also with how we approach patient care,” Whitney said.

Illinois Case Study: Refinement

Plans to improve rounding approaches at Evanston Hospital Endeavor Health arose through a classic improvement planning process, said Marina Kovacevic, MD, an internal medicine physician and lead hospitalist at the tertiary care hospital outside Chicago.

They have arrived at a split rounding approach that was developed both through planned iterations and an unplanned one. Setting your organization up to be able to identify a potential need to finesse MDRs or IDRs is an important factor after the heavy lifting of initial implementation is complete.

“The key points are focusing on the frontline team, getting feedback from everybody, sharing the understanding of the importance of discharge planning and discharge coordination, and keeping patients at the forefront,” Kovacevic said, noting that estimated discharge time and transportation considerations were among the details that rose to the surface.

The IRD approach was first piloted on a single adult general medicine nonteaching unit that tended to have low patient communication scores; discharge coordination problems; and broad swath of diagnoses, including sepsis, heart failure, pneumonia, liver failure, kidney failure, and alcohol withdrawal.

After problems arose from including too many team members in rounds, they tried splitting rounds to discharge-focused interdisciplinary and then clinical rounds, and then honed that approach, which was ultimately adopted.

Length-of-stay metrics were already met, so earlier discharge became the focus. The new IDR approach reduced discharge time by a median of 1 hour and 10 minutes, from 3:25 PM to 2:15 PM, the team reported.

But after those results were achieved, it was important to keep listening to the team members and provide ways for them to bring ideas and observations forward, Kovacevic said.

A difference became apparent between the teaching teams and the nonteaching teams.

“We recognized that on our teaching units, we may have a little bit of delay in care,” Kovacevic said, so feedback from the different team members such as nurses and case managers was gathered. “There were some anecdotal observations that we might need to work on care coordination and discharge planning a little more. And then we heard also from our resident doctors saying, ‘I don’t know how to decide when a patient is ready for discharge. I am not sure how to coordinate safe transition out of the hospital. I’m not sure what are the components of all of this.”

“And and that’s actually recognized as a key competency,” she added.

Attending physicians were added to the IDRs, although they remained resident-led. Length of stay dropped significantly, from 3.92 days to 4.44 days, although there was no significant change in median discharge time, which did have a nonsignificant drop from 3:57 PM to 3:45 PM. An additional qualitative survey measure indicated team members saw improvement across measures such as discharge time and medical plan clarity.

Thinking Discharge Indirectly

Strategies to improve discharge times and throughput often resurface every few years, said Marisha Burden, MD, MBA, a practicing physician and hospitalist who is also a professor of medicine and division head of Hospital Medicine at the University of Colorado Anschutz Medical Campus, Aurora, Colorado.

Burden describes herself as “really just dedicated to addressing real-world challenges in hospitals and health systems, and then my research focuses on designing sustainable workforce models, ideally to enhance clinician well-being, optimize patient care, and align all of that with organizational priorities. So I’m trying to hit the win-win-win spot, which is kind of hard.”

She also was the lead researcher of a 2023 report that showed in a randomized controlled trial, prioritizing morning discharge didn’t result in significantly earlier discharges or reduced length of stay.

“What we did see in the qualitative comments, though, is that these initiatives suggested that they could even cause harm, such as rework or frustration,” Burden said. “The take-home is we have to be really thoughtful about what our organizational initiatives are around this very important issue. While well-meaning, they may not work and may actually cause harm.”

The study was prompted by her own experiences.

“I’ve both received and sent texts urging busy clinicians to prioritize discharging patients first, when experiencing peak hospital capacity,” she said. “And everyone knows it’s important to discharge patients, but there’s a lot of competing demands. And so these reminders, while well-intentioned, often fail to drive meaningful change but remain a go-to strategy for leaders.”

Burden noted that a pivotal 2014 paper showed that length of stay increased as workload increased, sometimes by as much as 2 days during occupancies below 75% but amid high workload.

Instead of asking that all of the day’s work be finished before noon, Burden is now studying ways to optimize the workload. She pointed to a pivotal analysis published by the business researcher Kamalahmadi, which showed the ideal case load is between 10 and 14 patients, and the mix of complexity matters.

“There appears to be an interaction between workload and patient complexity that influences length of stay and ultimately throughput, which means that work design matters more than ever,” Burden said.

Kamalahmadi’s operations-focused paper, published in 2023, estimated that a community hospital and an academic hospital could each save $1.5 million annually by increasing staffing, although the driving mathematical model arrived at the optimal increases for different reasons. The community hospital needed more hospitalists to expand hospitalist coverage, while the academic center needed to reduce caseload.

Hospital leaders may be hesitant to increase staffing because it’s challenging to measure indirect links, Kamalahmadi said.

“I chose this approach because staffing is a fundamental factor that affects every aspect of discharge planning. Many studies focus on rounding and meetings, but even the best plans and strategies can’t be effective without enough staff to implement them,” he said. “The challenge is not just having the right people at the table but ensuring there is enough capacity to carry out those plans. Others may have overlooked staffing as a key factor because it’s a more complex issue, requiring a connection between staffing levels and hospital throughput.”

Burden called Kamalahmadi’s work “so impressive” and said it “suggests that getting work design right — which means optimal workloads — can improve throughput and even improve organizational financials, which is critical.”