skip to Main Content

Solving Nurse Allocation Issues: How AI Effectively Prepares Infusion Centers

Mohan Giridharadas

Mohan Giridharadas

Founder & CEO, LeanTaaS

Louis Pasteur famously said, “Chance favors the prepared mind.” In highly specialized environments such as infusion centers, which are unpredictable by nature, it often seems impossible to adequately prepare. Scheduling is highly complex, wait times are long, and nurses are usually rushed. Burnout and job dissatisfaction are systemic, and as a result, the patient care experience is suboptimal. Yet, new technologies make it possible to reduce uncertainties, adapt to change, and fuel a new level of preparedness that tips chance heavily in favor of patients and staff while enabling centers to function efficiently.

Teeing Up The Typical Day

The person who reminded me of Pasteur’s quote is the director of nursing in a cancer center for one of the nation’s premier medical facilities. At a conference, she reported on the progress of a major initiative she and a colleague had been tasked with — overhauling operations in the center’s infusion clinics. This particular administrator described the nursing environment when they took over as analogous to a golf tournament. Everyone showed up in the morning, exchanged pleasantries, maybe swapped some stories, and then teed off for the day independently with no visibility into anything else happening beyond their own schedule. One nurse might have 15 patients, another only three, and there was no sharing of information that would balance the load to work better for nurses and patients.

Her colleague, a clinical nurse specialist overseeing operations, shared a slightly stronger analogy. She thought it was more akin to a war zone. When the pair took the job of revamping infusion, patient wait times could be as long as 525 minutes. Nurses arrived with patients already waiting in the lobby. Nothing was ready for their appointments, which were then delayed as nurses fought over patients. Days quickly spiraled out of control as the nursing staff started the day behind on their already demanding schedules.

Through a tremendous amount of work, these two administrators and a team they had assembled overhauled the processes in place to better prepare for appointments in advance — forms signed, labs completed, drugs mixed, rooms and chairs ready — everything was wrapped up in a bow to start the day. But no nurses had been assigned. Even with five nurses (one per pod of 11 infusion chairs) coming in an hour early to make the assignments, they were still arguing over the day’s schedule. By 10 or 11 a.m., everything was chaos, and by 2 or 2:30 p.m., the center was empty. This was a consistent pattern, one commonly experienced by infusion centers across the country.

No End In Sight

Another issue plaguing this team as well as hundreds of other teams around the country is that once the schedule is thrown off, the day extends well beyond the scheduled close, often late into the night. In many infusion centers, nurses cannot remember the last time they made it out at the scheduled end of their shift. Walking out on patients in the middle of their treatments is not really an option. As a result, many infusion centers pay a significant amount in overtime every month, and nurses grow weary and resentful.

According to a 2018 survey of 22,000 nurses, half of nursing professionals work overtime at an average of nine extra hours each week. Not only does that take a toll on the staffing budget, but it leads to high turnover rates — and time lost trying to fill an open position places a further burden on existing staff.

Bringing Intelligence To Staffing

Health systems across the country are now experiencing a massive wave of innovation. In addition to medical advances, the underlying systems that enable health centers to function are being overhauled. Artificial intelligence (AI) in particular is transforming operations — from scheduling appointments to facility utilization to staff allocation — with the desired effects of happier patients and staff.

AI-based platforms have done wonders for appointment scheduling, taking into account how long appointments last, what is involved in that appointment, whether the patient has other linked appointments that need to be considered, and the patient load for the day. Systems can predict cancellations or appointments that need to be moved based on lab work so that whole days aren’t routinely upended. They also can allow for emergency situations when patients need to be accommodated. While no day is perfect, systems are constantly learning based on what happens in a specific infusion center each day, and the algorithms continuously improve based on the learning from the actual data in order to adjust future schedules.

Where it gets really exciting is in terms of how AI then can influence nurse allocation. New systems “read” the patient schedule, examine which physicians and nurses are scheduled, and account for acuity. Infusion treatments require focus from the nurse in a manner that is similar to an aircraft requiring focus from its pilots — 100 percent concentration at the start and the end of the journey. Intelligent allocation of patients to nurses can ensure that the nursing workload at the start and the end of each appointment is fully dedicated to a single patient while the “mid-flight” workload of a nurse can be reasonably spread across a small handful of patients that are seated near each other.

Applying AI to infusion centers therefore becomes a bit like running air traffic control at a busy airport. It connects multiple dependencies with precision, even in the face of variables that could change at any moment. Powerful tools help personnel make crucial decisions based on data that ultimately affect people’s lives — and they can do it in near real time with the appropriate HL7 data feeds.


AI-based solutions for infusion centers are no longer hypothetical as some of the world’s leading high-volume care centers have integrated them successfully over the past few years. Feedback across the board has been a reduction in overtime, an increase in nurse satisfaction, a leveling of workload — across the day and across staff — and a decrease in staff turnover.

More specifically, the colleagues in the example highlighted above instituted an AI-based system approximately two years ago. In the time since, they’ve developed a sophisticated model for determining acuity, which feeds into their scheduling software to help it make better determinations about how long appointments will take and what their staffing needs look like.

The nurses are much happier with the allocations generated by the software and have clear rules for making tradeoffs that will no longer negatively affect the schedule or patient care. The infusion centers also cap nursing assignments at eight patients a day to prevent staff from getting overwhelmed. And, four years after stepping up to transform their infusion centers, they’ve accomplished the seemingly impossible task of reducing wait times from a high of 525 minutes to less than 30.

AI-based technologies will not deliver a perfect solution to every challenge facing an infusion center, where the environments change every day, but systems do learn based on what happens. Predictive analytics are constantly improving, and as a result, each day becomes significantly closer to the standardized ideal, effectively preparing them for whatever chance throws their way.

About The Author

Mohan Giridharadas is an accomplished expert in lean methodologies. During his 18-year career at McKinsey & Company (where he was a senior partner/director for six years), he co-created the lean service operations practice and ran the North American lean manufacturing and service operations practices and the Asia-Pacific operations practice. He has helped numerous Fortune 500 companies drive operational efficiency with lean practices. As founder and CEO of LeanTaaS, a Silicon Valley-based innovator of cloud-based solutions to healthcare’s biggest challenges, Mohan works closely with dozens of leading healthcare institutions including Stanford Health Care, UCHealth, NewYork-Presbyterian, Memorial Sloan Kettering, MD Anderson and more. Mohan holds a B.Tech from IIT Bombay, MS in Computer Science from Georgia Institute of Technology and an MBA from Stanford GSB. He is on the faculty of Continuing Education at Stanford University and UC Berkeley Haas School of Business. For more information on LeanTaaS, please visit, and follow the company on Twitter @LeanTaaS, Facebook at and LinkedIn at

Back To Top