Daily Kaizen

In the past months, we’ve had the opportunity to lead a group of functional leaders through value stream mapping the revenue cycle of our organization.

Two books were foundational to this: “Learning to See” and “Value Stream Mapping for the Lean Office”. However, even with these great books, there is a gap between a) the theory and straight forward case studies presented and b) the complexity that must be navigated to be successful in reality.

From this experience, we have a few learnings that weren’t in either book, but proved successful for our team in bridging this gap. We’d like to share these in hopes that it aids your future mapping.

1)      Increasing the relevance of data – Within our revenue cycle, all functional areas are not aligned around the same customer. As a result, the leaders of various functional areas often see the horizontal linkage of measures as apples to oranges, and have a hard time drawing enough meaning to allow their thinking to be truly changed. As daily management systems continue to evolve and become connected horizontally, this should improve. However, in the current state, having a robust data component from current state mapping through future state remains a challenge. Given these constraints, the following helped in maximizing the effectiveness of our mapping:

•  Collect data to test the hypothesis, rather than attempting to collect all data available without understanding why. The hypothesis should span a large section (a “chunk”) of the value stream. (In our Revenue Cycle , we had three “chunks”: Non-dues revenue, premium revenue, and cash receipt, that each had a unique hypothesis to be tested)
• Anticipate major future state hypotheses, and begin collecting data to test up front. If daily management systems were present throughout the organization, data could fairly easily be collected in real time, making it easy to test hypotheses. Lacking that, we needed to allow enough lead time to collect, organize, and synthesize the data. Creating a hypothesis for each future state idea was critical; as it allowed us to clearly identify what specific data would be needed to test the hypothesis.
• Engage the team (operational leaders) in data collection – This is a significant change in role. Involving leaders in a structured plan teaches them why and how data is collected and used within value stream mapping, and exposes them to the complexity and problems with the current state. This is critical for the team to understand and own the current state, and in building engagement for the ongoing implementation and PDCA of daily management throughout the organization.

2)      Utilize an A3 with your value stream leader / pacesetter – In the initial week of work, we encouraged our leader to develop an A3 to bring focus to the problem we were solving, and to help communicate the story. We started with the background: “why is this important?”, “what’s so wrong with the current state that taking on mapping is critical?”.  Throughout current state mapping, we encouraged the leader to stay on the left side of the A3, as she developed hypotheses, used poignant data to concisely illustrate the current state, and asked “why?” to work toward root cause. The A3 helped the value stream leader succinctly communicate the issues with leaders throughout the organization, increasing their awareness and building alignment. By the end of current state mapping, alignment on the left side had been reached, and the focus shifted to developing the right side. The leader then used the A3 to share the story; from problem, to root cause, to countermeasures aimed at closing the gap on root cause. The engagement and alignment gained prior to the future state mapping was critical for that event to be powerful and effective, and to allow the event to move beyond the high level vision and into the detailed planning of implementation.

3)      Getting alignment on future state through nemawashi in current state. On the last day of current state mapping, the value stream team reviewed and finalizing their understanding of the current state– including a summary of the waste and improvement opportunities that they’d identified throughout the walks.
We then took the team through a process of drafting their vision for the future state, by asking them to use what they just confirmed about the current state to:

• Individually, either draw or list out their vision to meet the deliverables, and stay within the guardrails of the future state charter.
• In small groups, present their vision components and use a nominal group process to identify like ideas, resolve differences, and agree on a shared vision.
• In one large group, share the small group visions and identify common components and differences.

Through facilitation, the differences were resolved to arrive at one vision for the large group. This vision became the foundation of their future state map, to be finalized a few weeks later.

There were several advantages to this approach:

• Team members were still immersed in customer requirements and the wastes within the current state. When we gathered two weeks later to map the future state, there was an observable regression toward the status quo; they were back in their functional roles / mind sets. However, their words and ideas gave us the basis to challenge them to a more transformational future state.
• In the two weeks between current and future state mapping, the Pacesetter continued to build alignment around the vision, using the A3 for nemawashi.
• With a vision 80% in place at the start of the future state mapping event, the team could get further: refining the vision, defining details, agreeing on required improvement activities and sketching out an implementation plan.

Popularity: 14% [?]

Understanding the cause and effect relationship between processes and outcomes is a fundamental part of effectively managing and improving a business. Without a good understanding of this relationship you can put a lot of effort and resource into actions that add no value or worse may lead to harm. This is also an area of a lot of frustration for many of us deep in the trenches of trying to transform healthcare. This frustration comes from the extreme difficulty we face in trying to meaningfully measure clinical outcomes and then tie those outcomes back to specific actions that did or did not take place within our clinical systems.

Over the last two years at Group Health we have worked relentlessly integrating and improving our clinical process across the continuum of care. At a macro level the results are heading in the right direction with many of our outcome measures (admit rate, ER rate, etc.) showing significant improvement. Yet, we don’t often know why the improvement is occurring or when things are not improving fast enough we often don’t know what is not working. For example, when we see our Emergency Department rates starting to decline we don’t know if it is because we have improved access to care, improved our care planning process, improved our outreach process or a combination of all of these processes. In other words, the reason for someone staying healthy or getting sick is elusive.

As a countermeasure we have worked hard to try and find intermediate outcomes to help us better understand out performance. More importantly we have taken a relentless focus on standardizing our processes. We might not be able to prove in the short-term that having a comprehensive care plan will keep a patient out of the ED, but out intuition and a lot of clinical experience says it will. Thus across our system we are getting clear about which processes we believe will have the greatest clinical impact and we are improving them. Basically, the bet is that collectively if we improve enough processes across the value stream the results will follow.

While this strategy seems to be working it is not ideal. It is hard to tell a group of providers who have decades of analytical training and are very busy to trust that the results will follow if they just focus on the process. In the long-term groups like the Healthcare Value Leaders Network are looking to improve the measurement capabilities of clinical care. In the mean time we will keep focusing on the processes and learn a lit bit from each change.

Popularity: 7% [?]

As and organization transitions to a Lean management system many of existing structures and practices change. Teams and individuals begin to work together in new, more disciplined and more effective ways. One of the most important changes early on in the transformation is the requirement that team member participate in a Daily Huddle. Having a Daily Huddle is a simple change, but for manage teams and managers it a profound change. Many of these team members have never been asked to do more then their daily work activities and many of these managers have never been asked to lead their team in daily improvement. Thus the Daily Huddle represents a new way of working which like all changes can be difficult to get started.

A Daily Huddle is time each day that the team comes together to work through the PDCA cycle. The team discusses the plan for the day and makes sure everyone is on the same page. Additionally the team checks the results from the previous day/week/month and problem solves what can be done more effectively. Finally, it is a chance for the team to make adjustments and improvements to their daily work practices including training on new processes, etc

When teams first get started Dailey Huddles are often awkward and teams don’t know what to do. My experience shows it often takes several months for teams to get engaged. The good news is that over time teams learn to become reliant on Huddles and most would tell you it is time well spent.

Below are a couple of tips to consider when you first get started:

•  Make sure that the huddle topics are relevant to the team. This means that most of what is discussed needs to relate to the daily work. For example: who is here today, who called in sick, what is our expected demand, what problems came up yesterday, etc. Often managers spent too much time talking about improvement projects and teams lose interest.
•  Reviewing visual systems is an important part of an effective Huddle. Often early on in the Lean transformation teams done have robust visual systems. Don’t dismay, having the huddle will provide a context for getting the team to take ownership for Visual Systems. Ask them what information they need to know to manage day to day and then incrementally improve the visuals.  
• Keep the Huddles going, even if it is really busy. Often I see managers put a hold on Huddles when it gets busy. This sends a wrong message to the team. In fact, the busiest days should be the days that Huddles are the most important.  
• Finally, as team members to rotate responsibility for leading the Huddle. It should not always fall to the manager. This helps teams begin to self manage.

 

Popularity: 42% [?]