In 2001, Virginia Mason Medical Center was searching for a new management method that could help us achieve our vision to be the quality leader in health care. Through a chance meeting, a Virginia Mason executive began chatting with a Boeing executive about the Toyota Production System. The two discussed how Toyota's system of creating a product without defects and the concept of kaizen, or continuous improvement, could be adapted to the world of health care. And so began our journey. Soon our leadership team was actively researching the Toyota Production System and planning a trip to Japan to learn more. Seven years and several trips later, foreign words such as muda (waste) and poka-yoke (mistake-proofing) are now part of our everyday language. Our gastroenterology department was one of the first departments to benefit from the continuous improvement concepts using what has became known as the Virginia Mason Production System. The results have included dramatic improvements in GI procedure volumes. Here's how we did it.

1 Standardize each procedure room.
Staff can easily find what they need when supplies and equipment are located in the same place in each room. They also spend less time searching for items and can turn rooms over quicker.

2 Improve patient and provider flow.
After we measured the distance between procedure rooms and counted the number of steps it took patients to walk through the entire process, we redesigned the flow of patients and providers and reduced walking time. From arrival until discharge, patients now walk 42 percent less (89 feet, compared to 153 before). Admit nurses now walk 88 feet, compared to 371 feet before. Recovery nurses walk 230 feet, compared to 1,244 feet before — a reduction of 82 percent. We weren't done there.

We drew the routes that nurses, techs and physicians used to get supplies before, during or after a procedure. Seeing it on paper made it easy to eliminate extra steps by moving equipment or supplies. For example, we used to store narcotics in a medication room near recovery, which is nowhere near procedure rooms. Now each morning we distribute the medications for the day's procedures to each procedure room and store them in a locked device.

3 Free nurses from room turnover.
Our nurses used to do everything in the department, from accompanying the patient through the process to changing over a procedure room for the next patient. We created the role of room turnover technician. Freeing the nurses from room turnover duties let them focus on patient care, rather than cleaning a dirty room.

4 Instantly communicate room status.
As part of this new system, though, we needed to be able to easily determine the status of a room. The solution was a light system that tells staff on the floor who or what is needed in a room. A green light indicates a team member is in the room and a procedure is in process; a white light indicates the room is clean and ready for the next patient; a blue light signals the need for a supply or technician; a red light signals an emergency; and a yellow light means the room is dirty and ready for cleaning. This system lets us direct the appropriate person to the room to provide service. It helped our department reduce average room turnover time from 35 minutes to less than 18 minutes; sometimes room turnovers are as short as 10 minutes.

5 Stop pushing and start pulling.
We created a "pull" system that draws patients through the process rather than pushing them. For example, we admit the first GI/endo patient of the day in the procedure room. This eliminates wait time for the procedure room nurse and lets the admit nurse concentrate on the next patient in line. Once the first procedure is finished in each room, the procedure nurse pulls the next patient from the admit area into the appropriate procedure room.

6 Increase doctors' efficiency.
We also studied how physicians spent their time in the department and where they went. We discovered that physicians were often going back and forth between the GI/endoscopy suite and the adjacent clinic during gaps in procedure start times. To eliminate this, we scheduled procedures consecutively and asked doctors to document each case at the end of the procedure rather than at the end of the day. Keeping the physicians in the GI/endoscopy suite throughout their blocks of time eliminated nearly a quarter-mile of walking per physician each day. Finally, we eliminated post-procedure visits for patients with normal test results, which saved doctors even more time.

7 Improve scheduling.
In addition to improved block scheduling for physicians, we freed up more scheduling slots for the "unexpected" cases we could actually anticipate each day. The unexpected really is predictable. We reserve these slots for same-day urgent and emergent scheduling needs and don't overschedule planned procedures.

8 Small-batch lab results.
Besides patient flow, we also concentrated on the flow of information in and out of the GI department. Lab results used to be delivered once a day. Included in those reports were "pending" reports that provided no added value. We instituted direct printing of final pathology and lab results to the department, and eliminated printing all pending lab results, which had wasted much paper, ink and time. Only final results are printed now. Results now arrive three times a day in smaller batches that are processed quickly and passed on to physicians and patients throughout the day. The result is less of a backlog and a 61 percent reduction in the lead time for results, from 5.7 days to 2.23 days. Not printing cases with pending results has resulted in a 70 percent decrease in printing costs in the department.

9 Segregate admission and recovery.
Before we started our kaizen process, we admitted patients in the same area where they recover. This created a bottleneck of admit and recovery patients moving through a single central hallway. To relieve the logjam, we created a five-chair admit area in two rooms that were used only four hours a week for sigmoidoscopies. The reconfiguration created a U-shaped layout for the department with specific process areas for admission, procedures and recovery.

Not done yet
Without increasing hours of operation, we've been able to do around 160 more cases per room annually, resulting in increased revenue of more than $300,000 per procedure room per year. Because the department is more efficient, the net margin per room has increased by 60 percent. We've made significant gains in screening and therapeutic procedure volumes, but we have more work to do in complex interventional procedure schedule planning. We won't be satisfied until we completely eliminate waste and inefficiencies.