Reducing Patient Falls With a Lean Management System: An Executive’s Challenge

An interview with Diane Miller, vice president at Virginia Mason, executive director of Virginia Mason Institute and an executive sensei

What role did you play in the work to decrease patient falls at Virginia Mason?

DM: I cosponsored the work with Charleen Tachibana, Virginia Mason’s CNO and senior vice president of quality and safety. As a vice president, I am certified in VMPS [the Virginia Mason Production System] and understand the importance of the sponsor’s role in challenging the team to achieve results that may not be seen initially as possible. As a sponsor, I asked the team: “What are you testing this week?” “How are we moving forward to achieve zero defects in our work for patient falls?” “What are the barriers?” “Do we have standard work, did we follow standard work, and what are we learning?” The nurse managers on the teams talked about what was helping in their units to share the learning.

What did your team discover in their work?

DM: We found that often the staff had competing priorities, and while we had standard work, falls can happen when we don’t embed the standard work in our workflow. Not following standard work isn’t the error, but it was key for the team to address why we didn’t follow it and then work toward removing barriers or identifying countermeasures to ensure we could follow it. Sometimes, however, the standard work isn’t the right standard work and it needs adjustments based on learning from the PDSA [plan-do-study-act] cycles.

When a fall occurred, we asked, “Were we able to follow the standard work?” If not, we needed to ask why not, discuss what we had learned, and determine if we needed to modify our standard work or remove the barriers to following it.

The team also identified root causes and implemented targeted interventions. As we continued to study the reasons for falls through the data, we saw that a large percentage of falls were related to delirium, which led to a more focused organizational goal related to delirium work. There’s now specific, detailed standard work about delirium assessments and medications, alcohol withdrawal, etc. We also discovered that the shower floors needed better nonskid surfaces, that the color of the emergency cords in the bathroom blended in with the paint color, and that the bed and chair alarms sounded too much like other alarms. The team therefore improved the floor surfaces, put in red emergency cords, and altered the bed and chair alarms to sound distinctly different from other alarms, so that the bed and chair alarms could immediately alert the care team to what was happening at that moment.

What barriers did you find to making improvements?

DM: A big hurdle we had to overcome was the belief that you can’t stop falls. If we think we can’t get to zero, why would we try to do better? And we know we can’t get to zero in one week. We need continual studies and PDSAs to discover the causes of falls and try to fix the problems, test new ideas and implement best practices into our work.

And we also needed to talk about why we need to stop falls. We want to stop falls because of the cost to patients and the organization. Falls affect length of stays, surgical procedures and patient experience. We also understand the deeply emotional impact on our care teams when a patient in their care experiences a fall.

There’s another component we needed to address while learning how to get to zero. We asked: “Until we reach zero, how can we reduce the harm caused when a patient does fall?” And even as we were working to reduce the harm, we needed to continue to believe that we can eliminate all falls — that we can get to zero.

“There’s another component we needed to address while learning how to get to zero. We asked: “Until we reach zero, how can we reduce the harm caused when a patient does fall?” And even as we were working to reduce the harm, we needed to continue to believe that we can eliminate all falls — that we can get to zero.”

– Diane Miller

What new processes did the team implement as a result of the PDSAs?

DM: The team established hourly rounding with patients — they found it was important for the patient to know that the nurse or PCT would be coming each hour. Their new standard work required the care team to discuss six things during each round. Then they studied these rounds to refine the standard work. They also worked with patient partners to develop new strategies to educate and involve patients and their families, including the creation of a fall-prevention video and written fall-prevention tips. To keep staff training current, they created annual training for hospital providers, pharmacists, nurses, assistant nurses and physical therapists, and they implemented annual bed-alarm training for nurses, assistant nurses and physical therapists.

Were there any changes at a higher level?

DM: Yes. In our Patient Safety Alert System process, a fall used to be recorded as a yellow alert, which meant it was collected in the aggregate. Then we agreed that each fall should be an orange alert. In other words, each fall had to be reviewed in real time and not simply be collected as part of a trend (although each was still part of a trend). As an executive I carried a pager, so I knew about each fall. The manager in charge would immediately huddle with the care team to assess the causes and create any necessary immediate action for the patient.

In addition, on a weekly basis the team reviewed the log of actions. We asked: “Was the standard work being followed?” and “Do all staff know how to follow it?”

So instead of just the safety office looking at details and finding trends, the managers were addressing the falls and looking for answers to get us to zero defects. They continually looked for patterns, analyzed the system and checked to see whether everyone across the organization had common definitions and understood the nuances.

Did the dynamics of the team change over time?

DM: Yes. We initially focused on the inpatient units but began to also see falls in the ER, so we invited the ER manager to the weekly meetings as well. The managers learned from each other in these meetings and kept each other on track.

Did the standard work for patient falls vary in different departments?

DM: In every patient care unit, there’s a lot that is the same in our standard work — there’s rounding, huddles, etc. If there’s a special population in a unit, there’s tweaking.

The bundle was developed over many weeks of this work. We had to look at national data and figure out what assessment we were using to assess risk and thereby apply the appropriate interventions. We asked, Does the work need to involve a nurse? In our RPIWs [Rapid Process Improvement Workshops] and other work, how are we using the full, consistent assessment and language?

How important was data in developing the standard work?

DM: It was important. Some national falls work was helpful to review, but it didn’t tell us everything about root cause. Without hard data, how do you know what makes a difference? It’s not one thing; it’s a collection of changes. We can have theories about why falls are happening, but we really had to peel it back. That’s how we were led to our delirium work. It took the discipline of our method to get to the root cause. It involved a PDSA process over and over.

For example, the use of a gait belt can prevent patient falls, so we needed to find out why a gait belt wasn’t being used by staff when we agreed it should be. After a fall, we asked: “Did the patient have a gait belt on?” If the answer was no, we asked why. And sometimes the answer was “I couldn’t find a gait belt.” The same was true for walkers for the patients who needed them. We might hear: “I couldn’t find a walker.” So then the question for the team focused on how our staff — who knows they should be using a gait belt or walker with certain patients — can always have quick access to the tools they needed to prevent falls. It was very important for us to make it easy for staff to do the right thing.

How important is educating the staff about the best practices? 

DM: Education is important, of course, but education and slogans are not enough. We need mistake-proofing to prevent falls from happening. The root cause is rarely “I don’t remember.” If there’s a sequence of six things to do every hour, and they must be done in that sequence every hour, then it makes sense to put a chart on the wall to ensure that all six things are done in that sequence every hour. Standard work is important. Real time is important. Hourly rounding is important, and it needs to be reliably done.

Why is it important for an executive to be part of a team’s work to achieve major organizational goals? 

DM: As an executive in this work, I was outside the day-to-day work of the team. At the meetings, I would listen to the team’s assessment of data and would ask, “What’s the decision?” My involvement in this way showed that I was holding people accountable. I was always asking questions from a VMPS perspective: Are we following the standard work? Is this the right standard work? Do we need to revise it? When will this revision happen? And what are the barriers we can remove to ensure you can provide the perfect patient experience?

Diane Miller, MBA, is a vice president at Virginia Mason, the executive director of Virginia Mason Institute and an executive sensei. As a key member of Virginia Mason’s executive team, she helped design the infrastructure to support the organization’s adoption of the Virginia Mason Production System as its management method. Today she leads rapid process improvement events for clients worldwide and designs and develops education and training services for health care leaders and providers. She also oversees the Center for Health Care Solutions, an innovation unit working to create a new model of health care delivery that achieves reliable evidence-based care with high patient satisfaction and a lower cost. Prior to coming to Virginia Mason Institute, Diane was a consultant who worked with organizations to apply the Toyota Production System to health care. She also worked in a variety of positions at Virginia Mason, including a leading role in organizational development when Virginia Mason first began its lean journey in 2001. Diane is certified in the Virginia Mason Production System®.


  • Jacqueline McKenna says:

    This is a very interesting article about the influence an executive can bring to an improvement initiative. In our improvement collaborative programmes one criteria to be accepted is an executive sponsor. Do you envisage that without an executive sponsor the outcomes of a programme would be the same?

  • Cathy Corrie says:

    This raises some interesting questions around how you embed improvement as part of business as usual and the challenges of doing so where the problem is perceived to be too big to fix. The role of high quality, real time data in identifying root causes, targeting solutions and getting clinical buy-in feels like a widely transferable learning.

Leave a Reply

Your email address will not be published. Required fields are marked *