Waste Not, Want Not: Inside the Virginia Mason Production System

A decade ago, the executive team of Virginia Mason Hospital and Medical Center (VMHMC) in Seattle, Washington, flew to Japan for training in the Toyota Production System (TPS), a continuous–process-improvement method pioneered by the automobile manufacturer. Lucy Glenn, MD, chair of VMHMC’s radiology department, says, “We began to understand how applicable this method was to health care. So many of the things we do in health care are processes. Everything that surrounds the physician–patient interaction is a process that can be improved.” The result of this training was the development of the Virginia Mason Production System (VMPS), which takes the principles of the TPS (also known as lean production) and applies them to health care. The radiology department was quick to adopt the new system, performing seven week-long workshops in 2002 alone to learn to use the lean tool set. “Subsequently, we learned to do rapid-kaizen (-improvement) events, which are two-day process-improvement events,” Glenn says. “Week-long rapid–process-improvement workshops are for more complex issues.” Critical to the adoption of the VMPS was the goal of putting patients first, which was introduced in the hospital’s 2001 strategic plan. Glenn explains that the objective of many of the radiology department’s process-improvement events is improving the patient experience. “In any area of radiology you can think of, there are opportunities,” she notes. “One of the things we strive for is elimination of waste, and one of the biggest wastes is time. Offering same-day access for patients is critical to making imaging a value-added service. You want to be timely in providing the imaging and timely in reporting the results.” Identifying Targets Process-improvement methods are applied to targets large and small; a key facet of lean production is an emphasis on the authority of frontline staff to understand how their own processes can be optimized. Richard Lee, administrative director of radiology operations at VMHMC, says, “We go to the people who are actually doing the work and are in immediate contact with the patients. We have a process where we ask them to generate ideas as to what could be improved, and that’s where it starts—from identifying how our processes could be made better for the patient.” At the other end of the pipeline, Lee says, administrators look at improving value streams to reach clinical, financial, and operational goals. “That macro perspective can result in process-improvement events and kaizens as we work a value stream from the top down,” he notes. “If it’s a more complicated issue, it may result in a series of kaizens. Process improvement can start anywhere in the organization.” In the VMPS, quality and cost are viewed as being inextricably linked, Glenn says, meaning that an opportunity for improvement in either area can launch a process-improvement project. “The more you improve quality, the more you reduce cost—because you are eliminating waste,” she explains. “You could reduce cost without improving quality, but that’s not what we’re after; we’re after reductions in the number of defects, in the amount of lead time, and in the waste within a process. Those will lead to cost reductions, but the ultimate goal is to improve quality.” In fact, Lee says, in lean production, cost is seen as a symptom of quality problems. “It’s symptomatic of waste and defects,” he observes. “We’re not thinking, ‘Let’s get rid of this cost.’ Instead, if a cost is out of line with what it should be, we’re considering what the implications are for the process. The tools are designed to address quality, not cost.” Radiology-department Projects In 10 years, the radiology department at VMHMC has tackled an array of issues using lean tools. Glenn says, “We’ve applied them to every area of radiology: all of the different modalities—CT, MRI, ultrasound, the breast-imaging center, and every section. We made a huge push to be able to offer same-day access to all of our modalities. We worked on our backlog to understand what really created that and how to take care of it, on an ongoing basis.”

VMPS Radiology Events In 2011, the radiology department of Virginia Mason Hospital and Medical Center (Seattle, Washington) performed two week-long rapid–process- improvement workshops and three kaizens. + Participants in a scheduling workshop (Integrated Procedural Center) planned and designed a resource-scheduling process that is transparent, integrated, standardized, and based on customer demand. They allowed for production planning of room, staff, and provider resources on the day of the procedure. + A clinical-admissions workshop (procedure patients) focused on reducing the lead time for procedure-patient admissions by eliminating waste, for patients and staff, that resulted in patients being left waiting. The workshop created and refined procedure groups and created standard processes for preparing patients for procedures. + In a kaizen for emergency-department CT flow, participants created criteria/ guidelines to ensure defect-free contrast administration to all patients. They also created signaling between the radiology and emergency departments to ensure smooth patient flow. + Participants in a kaizen for standing radiography designed a new device for obtaining foot and ankle images of standing patients. + During a kaizen for ultrasound flow management, participants improved the flow of patients, providers, information, and equipment in the ultrasound department by applying the concepts of flow management, skill–task alignment, and visual control. 2012 Lean Events In 2012, the radiology department has already performed one week-long rapid–process-improvement workshop and three kaizens. services (and that includes visual controls for resource use). + A kaizen for radiology forms and document management emphasized elimination of waste in overproduction of forms completed by patients, storage of forms, transportation of forms from the main campus to storage, and transportation and costs of storing forms with a storage vendor. + Participants in a kaizen for interventional-radiology provider flow used the principle of level loading (and principles developed earlier in 2012) to optimize the interventional-radiology scheduling template, ensuring the best procedural experience for patients and providers. + A second kaizen for emergency- department CT flow focused on reducing the time elapsed between appearance of the CT order (in the PACS) and the delivery of results to the emergency-department physician (by phone). Participants examined current workflows to streamline processes and improve efficiency by eliminating waits. They used Virginia Mason Production System tools to create standard processes for the sequencing of services (such as CT exams, laboratory studies, and electrocardiography) and to improve the signaling process between services. + A second scheduling workshop (Integrated Procedural Center) focused on planning and designing a resource- scheduling process that is transparent, integrated, standardized, and based on customer demand. It will allow for production planning of room, staff, and provider resources on the day of the procedure by standardizing the flow of information for procedure scheduling. It will allow for production planning, optimizing the flow and efficiency of the provider, understanding and meeting future IT needs, and designing a scheduling system that is integrated and is accessible to all

The department also looked at imaging quality on a larger scale, undertaking multiple projects aimed at gaining a better understanding of the evidence for certain highly utilized studies. One such project dealt with the use of MRI to diagnose back pain, which was found to be almost always inappropriate in a recent study.¹ “We realized that the way we had been handling back pain was this: The patient would see the primary-care provider, who would order an MRI exam, and then they would send the patient to a physical therapist. Maybe a month down the road, the physical therapy might be started,” Glenn says. “We realized that the patient needed to be started on physical therapy right away, and then—if there’s still an issue after six weeks—to get an MRI study.” This is an example of value-stream mapping and its results: By rearranging the flow of clinical events, the department was able to reduce MRI utilization, with the end result of eliminating waste, improving patient care, and reducing costs. “It’s making sure that we are appropriate in our imaging, and it’s creating access for those patients who really need those exams,” Glenn says. VMHMC’s radiology department also tackled radiation-dose reduction for CT exams as part of its evidence-based–medicine initiative. Radiologists routinely critique CT exams in terms of appropriateness; in addition, members of the department collaborated on a project to revamp every protocol, in every CT system, to ensure that dose was reduced as far as possible. “Whenever we have a process-improvement event, we make sure we use everyone involved in that process,” Glenn notes. “That includes a scheduler, a technologist, and a radiologist, if need be. We also try to involve the patient’s voice in most of what we do. If we can’t have an actual patient, we’ll be sure to have the patient’s voice represented by surveys performed ahead of time.” In 2011 and 2012, the radiology department has performed three week-long rapid–process-improvement workshops and six kaizens (see box). Patient Safety Among the key benefits of the VMPS, 10 years into its application, is the augmentation of patient safety that it has made possible throughout the organization. Glenn says, “One of the things we’ve worked on for many years is this culture of safety—getting to a point where every staff member feels safe bringing up every patient-safety issue, no matter what it is. We don’t care what the patient-safety alert is or where it came from; we just want to know about it so we can address it.” A patient-safety alert results in an examination of the issue from two perspectives: root-cause analysis and culture. “Root-cause analysis deals with whether the problem is a process issue or simply an individual error or behavior issue,” Glenn says. “The cultural analysis looks at behavior that was a human mistake or an at-risk behavior—you’re going 65 miles per hour when you know the speed limit is 55.” Patient-safety alerts in radiology have often focused on certain contrast media and their association with kidney problems in at-risk patients. “Once you realize what the components are in a patient-safety alert, you can deal with correcting them,” Glenn notes. “Usually, they are handoff or communication issues—problems with information flow that lead to human error. There are things that can be done to improve the processes surrounding a patient-safety event.”

FigureThe Integrated Procedural Center’s floor plan.

Most critical to improving safety is creating a culture of vigilance, Glenn and Lee note, in which employees feel protected enough to report any problems. “The whole idea that you cannot talk to just anyone, because of a power gradient, is something we work on continually,” Glenn says. Results are measured using an annual Culture of Safety survey distributed to employees. “Last year, 78% of our employees said they felt safe reporting a patient-safety issue; at many organizations, it’s around 40%,” Lee notes. “That’s not good enough for us, though. Our goal is 100%.” Facility Design VMHMC recently opened the doors of its new Integrated Procedural Center (see figure), which brings together the catheterization, interventional-radiology, and gastrointestinal laboratories. “This center directly reflects the VMPS principles,” Lee says. “We listened to the voices of the patients, physicians, nurses, and technologists to design a unique work area, in which we really worked to remove waste.” The facility was designed using a lean tool kit for space design that focuses on production, preparation, and process (3P). “3P allows you to look at a process before you start a service line and bring the frontline staff together to map it out,” Glenn says. “You look at information flow, supply flow, provider flow, and patient flow; you determine the best way to optimize these and have a facility that will make the work easier, not harder.”

A Lean Tool Set for the Radiology Department At the 2012 Health Connect Partners Radiology & Imaging Conference in Miami, Florida, Dan Littlefield, RPh, MBA, director of process improvement and lean consulting at Healthcare Performance Partners, presented “Improving Imaging Services Using Lean,” on May 11. Littlefield outlines a set of lean tools that are particularly useful for radiology practices and departments. Standard work methods: Littlefield defines standard work as “the current, one best way to do an activity,” he says, as documented on a one-page job guidance sheet. For a department to function optimally, he says, 80% of the work done in a given day should abide by standard work protocols, freeing staff to handle the other 20%, or outlier work, more efficiently and effectively. Single-piece flow: To differentiate between single-piece flow and its converse, batching, Littlefield offers the example of escalators versus elevators: Escalators convey people one at a time, in an even flow, while elevators convey them in batches. “Error potential increases as batch size increases,” he says, in reference to patients. Radiology departments should endeavor to make the flow of patients even, rather than dealing with them in batches. Visual management: In lean production, the highly visual sharing of information “exposes abnormalities, eliminates waste, and promotes error prevention,” Littlefield says, in addition to making quick recovery possible and supporting standardized work. Visual management also uses measurements to inspire the staff to meet goals; examples of these indicators, for radiology, might be daily turnaround times, hours worked per unit of service, and error rates. Error proofing: Littlefield observes, “Too often, in health care, we inspect at the end of the process to make sure that whatever we did has been error free.” Instead, he recommends, radiology practices and departments should seek the root causes of errors—to prevent them before they can happen. “You cannot inspect quality into a product,” he notes, paraphrasing Harold Dodge (1863–1976). “You can only build quality into a product.” Leadership: Effective leadership is “the most important tool we can use in lean,” Littlefield says. Leaders must invest personal time and attention to following through on actions related to change, and they must also be change advocates, constantly challenging the status quo of the practice or department. “If you consistently meet your goal,” he says, “then you need to change the goal.”

In applying 3P principles to the design of the center, the team at VMHMC looked at the similarities among the three procedural areas that could be used to create an effective shared space. Although the rooms for each procedural area are quite different, the admission and recovery areas are shared. A central patient area is considered onstage, while backstage, provider and technologist flow is occurring, invisible to the patient. “The patients’ experience is very calming and relaxing,” Glenn says. “They don’t see what’s going on backstage (all of the communication between the radiologists and the technologists). There’s a central space for patients, who then go into the rooms on the outside of that space; the outermost ring is the backstage workflow area.” A similar process was used to design VMHMC’s breast-imaging center. Glenn says, “We went from a large space to a smaller space, but managed to do much more within that smaller space because it was designed for better efficiency.” The 3P design-optimization process has more than increased patient satisfaction to offer the radiology department, however; by reducing overall square footage, it lowers operational and staff costs. “A lot of space is wasted in waiting rooms,” Glenn notes. “We try to design our processes so that there are few waiting states, and you don’t need those big rooms.” Lee adds, “We design the rooms so that we require fewer staff members to watch patients. Rather than having staff covering three separate areas, we’re down to one person who can control the flow.” The result is a facility that supports the organization’s patient-first strategy by freeing staff to deal with clinical, safety, and service concerns, instead of issues arising from patient flow. “We designed it to improve quality and safety and to reduce the burden of the work that the staff has to do to provide a quality service for patients,” Lee says. “It’s a design in which we strove to perfect the health-care experience for our patients.” Creating the Culture Glenn and Lee stress that the lean culture is continually supported and enforced in the organization. “The culture change didn’t happen overnight,” Glenn notes. “We’ve been doing this for 10 years, and it probably took five to seven years before we got to a tipping point where everyone in the organization understood and embraced the method.” Creating the culture requires very strong top-down support, they say; the organization’s leaders must be committed to clearing the time and resources that staff members need for true process improvement and change. At VMHMC, a centralized kaizen-promotion office helps develop process-improvement projects with the radiology department, and the department reports to the entire organization on the results of these projects at the 30-, 60-, and 90-day marks. “You have to show the rest of the organization the targets you achieved—so that there is rigor, in terms of implementation and making sure those changes stick,” Glenn explains. Every Friday at noon, hundreds of VMHMC employees join the organization’s leaders in an on-campus auditorium for what is known as reporting out, in lean terminology. In the 60- to 90-minute meeting, staff members are invited to report on process-improvement projects that have taken place that week (and, in doing so, to hold one another accountable and encourage one another to continue). “There is a strong commitment from executive leaders to implementing staff ideas,” Lee says. “Staff members have seen those efforts come through and have seen those ideas become a reality. In our organization, the commitment to helping the staff put its ideas into play has been a constant.” Lee continues, “Over 10 years, everyone becomes a participant. People are excited to share what they have accomplished. They are proud of the time and hard work they put in; as a frontline employee, I view the events as very self-reinforcing.” As health-care facilities seek efficiency opportunities to offset reimbursement declines, lean process-improvement techniques (see sidebar) are capturing the interest of health-care executives across the country. Be advised, however, that results are not achieved overnight. Time and persistence are the keys to making lean methods work for radiology and for health care in general, Glenn concludes. “This is our strategic plan. It’s our vision,” she says. “Quality is our top goal, and the strategic plan is everywhere. We start every meeting with it. It’s on our screen savers. It just takes constancy of messaging and time.” Cat Vasko is associate editor of Radiology Business Journal.