Figure 1. Baseline system performance: surgery prep time.

Figure 2. Funnel plot of surgery prep time showing variance by surgeon.

Figure 3. Chief of Urology’s individual performance for surgery prep time.

Figure 4. All surgeons’ performance for surgery prep time.

Figure 5. Last case end time delta showing improvement with cases ending early.

Figure 6. Adaptive clinical management process.

U.S. hospitals and health systems lost between $53B and $122B of revenue due to COVID-19 in 2021. Sixty-eight percent of a high-performing hospital’s revenue and up to 60% of their operating margin comes from perioperative services.

Many strategies have been suggested for optimizing operating room (OR) performance, eg, reorganizing scheduling, rethinking block utilization, changing leadership structures, and instituting daily team huddles. These strategies do have merit, but they do not in themselves produce the kind of efficiency gains that are urgently needed to increase capacity.

Adaptive Clinical Management (ACM) is a structured process of robust decision-making based on data supplied by active monitoring systems. ACM is a method to change a system and learn about the system. A critical step is the ability to evaluate results and adjust actions (adapt) on the basis of what has been learned. Barriers of implementation of ACM in health care include not collecting or analyzing data and not providing access to data to the front lines.

Increase Capacity in Urology OR by Optimizing “Surgical Prep Time”

The aim of this project was to increase the capacity of the OR by changing practice patterns that are in our control as surgeons (surgery prep time). We will define success as being able to care for more patients without extending work hours.

Surgery prep time is defined as the time interval between anesthesia readiness to start of the surgical procedure. The tasks to be accomplished during this time include positioning the patient, performing the safety time-out, and prepping the surgical field before incision. The time stamps for anesthesia readiness and the procedure start are routinely recorded data elements in the electronic health record. Baseline data were automatically extracted and analyzed by SPC charts using Adaptx (Seattle, Washington). Adaptx is software that extracts readily available data from the electronic health records and displays the data as SPC charts in real time. Variation between surgeon performance was analyzed using funnel plots. Two providers were identified as having the lowest prepping times outside 3 standard deviations from the rest of the group. Their prepping time protocols were reviewed. Common factors in their practice were identified including their role during prepping time, solutions used for prepping, and the roles of other team members. A standardized prepping workflow was then first implemented by myself as the team leader and, once deemed successful, it was disseminated to other members of my division. Data for clinical performance were fed back weekly to the surgeons so they could continue to iterate and adapt their workflows to achieve higher levels of efficiency.

The full 3-minute scrub is performed before start of day. This then allows for a shorter scrub using an alcohol based surgical hand disinfecting agent. Both the surgical and the anesthesia pause are done simultaneously. After induction of anesthesia and placement of regional blocks, the surgeon helps in positioning the patient. The surgeon then preps with betadine while the trainee scrubs. The trainee then drapes the patient while the attending scrubs. When draping is complete, the surgeon is scrubbed and gowned, ready to start the case. If no trainee is available, the same workflow is performed using the scrub tech.

Figure 1 shows the baseline performance for surgery prep time from January 2019 to September 2020. These data are plotted on an x-bar chart using standard 3-sigma upper and lower control limits. System performance is 13.7 minutes and stable (lack of special cause variation). Note there are no data for April 2020 due to OR closures related to COVID-19.

A funnel plot of the individual surgeons shows the variance and allowed the team to identify two surgeons (labeled A and B) as having the best performance (Figure 2). Surgeon A has surgical prep times averaging 11.2 minutes, which falls outside the 3-sigma lower control limit and is considered a special cause variation. Surgeon B shows similar performance (11.2 minutes) but has performed fewer surgeries in that time frame and, therefore, does not fall outside the lower control limit.

As the champion, I then adopted these best practices, which became my new standard clinical work in September 2020. Figure 3 shows my performance from January 2019 through February 2020 with just common cause variation and stable process. The chart has a change annotation on September 2020, which reflects the new clinical standard workflow. There is special cause variation (8 points below center line)—a shift down is detected from September 2020 through April 2021—indicating individual improvement for surgical prep time. In addition, there are breaches of the lower control limit in September 2020 and also in April 2021, all signaling improvement.

As a result, the entire group was asked to adopt the same standard clinical workflow and strive to be more efficient. Figure 4 shows how the entire team was able to improve their performance following adoption of these protocols in September 2020. There was a special cause variation (8 points below the centerline) which justified replotting the center line at 11.01 minutes.

As a result of this improved efficiency in combination with other efforts to improve patient flow (including the same process for anesthesia prep time), this facility was able to increase the case volume (Figure 5) from 70 cases a month to an average of 85 cases a month. It should be noted that March 2021 and April 2021 show a record number (106 and 107, respectively) of surgeries performed. This additional volume represents approximately $2M-$2.5M additional revenue per year.

The balancing measure chosen to ensure staff are not working past the end of their shifts was “last case end time” (Figure 5). Here, the time difference was calculated between the last patient leaving the OR and the official end of block time. During measurement of baseline performance, the last patient out of OR time was 15.7 minutes after the end of block time, meaning the room was finishing late. After September 2020 the last patient was leaving the OR on average 20.8 minutes before the official end of the surgical block, meaning the room was running ahead of schedule.

Conclusions

All OR processes can be measured using routinely captured data in the electronic medical record and continuously monitored using statistical process control charts. Variation in performance can be quantified using funnel plots of individual clinicians, allowing best practice to be identified and scaled. During and following implementation of new clinical standard work, the effect on individual and team performance can be monitored, allowing the team to learn more about their system and continue to adapt their work to achieve even higher levels of performance.