CONTACT: JENNIFER BROWN
Iowa City IA 52242
(319) 335-9917; fax(319) 384-4638
Release: May 25, 2001
UI researcher uses statistical analysis to improve staff scheduling for
IOWA CITY, Iowa -- A nationwide decline in the number of people training
to become anesthesiologists, operating room nurses, surgical technicians,
and nurse anesthetists has made it more difficult for hospitals to hire operating-room
staff. At the same time, hospitals are also facing reduced reimbursement,
compared to costs, for perioperative services. These two problems pose serious
challenges for hospitals providing perioperative care.
According to UI Health Care researcher, Franklin Dexter, M.D., Ph.D., UI
associate professor of anesthesia and director of the department's division
of management consulting, both problems can be addressed by increasing productivity
in the operating room.
"It is important to minimize downtime by scheduling staff hours to coincide
with when there are patients to treat," Dexter said. "But you also
want to minimize the number of hours staff have to work when they weren't
scheduled to be there, which results in higher costs."
Dexter explained that in the latter situation, the costs are both direct,
in the form of overtime payments, and indirect, in the form of frustration
on the part of staff and surgeons at having to work late in the day, which
can result in difficulties with recruitment and retention. Likewise, patients
have longer waiting times on the day of surgery.
Over the past few years, Dexter and his colleagues have developed a number
of computer algorithms for finding solutions to these kinds of staffing problems.
The computer program can generate and evaluate millions of possible solutions
A request from a large health care system to generate optimal operating room
staffing solutions for nine different, independent surgical suites, provided
the researchers with an opportunity to test the statistical analysis methods.
The suites included trauma centers, community-based hospitals doing mostly
elective surgery with some urgent cases, and ambulatory surgery centers. The
suites ranged in size from two to 17 operating rooms.
Dexter and his colleagues assessed weekday staffing at the nine surgical
suites using two years' worth of operating room data on case duration and
staffing levels. The results of the study appear in the June issue of the
journal Anesthesia and Analgesia.
"We knew that the computer algorithms would generate the optimal solution
for maximizing productivity," Dexter said. "But our question was
whether they would perform significantly better than the solutions arrived
at by the operating-room managers based on their experience."
In fact, the software package generated staffing schedules that increased
productivity and decreased costs compared to those used by the managers in
eight of the nine suites by an overall average of 19 percent. Only the smallest
suite, with two operating rooms, was being staffed as efficiently as possible.
"We had hypothesized that for that surgical suite the human solution
would likely be as good as any solution derived from our program because the
situation was not very complicated," Dexter said. "There was no
benefit to having a sophisticated mathematical way of obtaining a staffing
plan for such a small surgical suite."
However, for each of the other eight surgical suites, the statistical methods
identified staffing solutions with significantly lower costs than those currently
being used, and for seven of the nine suites, the cost of the plan identified
by the statistical method was at least 10 percent lower.
The study highlights the fact that even for suites with relatively few operating
rooms, finding the optimal staffing solution is a complicated task. The computer
software can assist operating room managers because it can generate and evaluate
literally millions of potential solutions to find the best one, many more
than a manager could do by hand.
Although the statistical methods produce the most efficient staffing solutions,
Dexter suggested that even if surgical suites don't use the program to generate
mathematically optimal solutions, the study still provides useful insights
about new strategies managers could use to increase productivity.
Most importantly, managers tended to use the same staffing schedule for every
day of the week. The software, however, showed that significant increases
in productivity could be achieved by varying staffing hours among the days
of the week.
Managers also had a tendency to use overlapping 8,10 and 13-hour shifts in
an attempt to increase productivity. In fact, Dexter's study suggested that
not only did this approach not improve productivity levels, it actually resulted
in longer delays for surgeons and patients on the day of surgery.
"A consequence of everyone working 8 hour shifts is that you run more
operating rooms and have more first-case-of-the-day starts," Dexter explained.
"Surgeons and patients prefer a first case of the day because they know
exactly when that case will start. With later cases you don't necessarily
know that they will start at the scheduled time."
Dexter suggests that the empirical results could be useful to managers and
hospitals in designing staffing solutions for operating rooms and deciding
how many operating rooms a facility should run.
The software package used in this study, CalculatOR, was developed by Medical
Data Applications, Ltd., of Jenkintown, Penn. Richard H. Epstein, M.D., associate
professor of anesthesiology at Jefferson Medical College and president of
Medical Data Applications, Ltd., and H. Michael Marsh, M.B.B.S., professor
and chair of anesthesiology at Wayne State University were co-authors of the
This work was performed as part of a consultation by the Division of Management
Consulting in the UI Department of Anesthesia.
For more information about consultations or about the CalculatOR software,
contact Franklin Dexter at email@example.com.
University of Iowa Health Care describes the partnership between
the UI College of Medicine and the UI Hospitals and Clinics and the patient
care, medical education and research programs and services they provide.