In this session, I will review some ideas and principles with regards to leveling work in healthcare. Creating smooth workflow is called heijunka by the Japanese who employ Toyota production system. How do we achieve smooth flow? Using a stream as an analogy, how do we create unobstructed workflow, shown in the upper image, as compared to flow that has multiple obstructions and erratic flow, as shown in the lower image? Why should we aspire to achieve even production or level workflow? Because wide swings in workload create stress and encourage multi-tasking. As we discussed in fixing healthcare 1.0, both of these conditions increase the likelihood of human errors. Whenever there is sudden extreme increases in work without appropriate adjustments, this condition endangers our patients. When it comes to smoothing out work, there are two conditions in healthcare that require unique approaches, emergency patients and elective patients. Let's first look at the best practices for leveling workflow for emergency patients. For this condition, production needs to be designed to match customer demand. Walk-in clinics and emergency rooms employ these approaches. These areas need to adjust staffing to patient demand and this can prove difficult. Ideally, emergency patients should not have to wait. To achieve this goal, workflow needs to be more flexible. And personnel and processes need to increase and decrease in number and speed to match demand. When it comes to emergencies, we can't control the number of patients, but we can match the workers to the workload. Ideally, health systems should match the number and types of workers and other resources to deliver value on demand without waste. Overstaffing can result in low per employee productivity, while understaffing can lead to stress, multi-tasking and errors. Unfortunately, staffing is often based on convenience. And traditional eight and 12 hour staffing patterns usually do not match demand in the emergency room environment. Studies of demand over time in emergency rooms and clinics reveal surprisingly predictable patterns of demand, and once they are understood, worker numbers and resources can be adjusted to match these patterns' demand. As an example, let's look at a typical large health system emergency room. The vertical axis for this set of bar graphs shows the number of patients waiting in the Every hour of a typical weekday. As you can see, demand is low early in the morning, and progressively rises, peaking in the early evening, maximum demand occurring in the 8 hour period between 3 PM and 11 PM. How can we level this workload? I work as an admitting officer of the day approving and performing second checks, error proofing, for all patients that the Team wants to admit to the internal medicine service. My shift is from 7 AM to 7 PM, shown as the number one at the top of this graph. In the morning hours until about 2 PM, I can easily handle the volume. But after 2 PM, the workload far exceeds my capacity. Recognizing this reality, our hospital service assigns a second physician beginning at 2 PM and going until 12 PM. And to manage the peak demand, assigns a third physician from 4 PM to 12 PM. This more flexible approach to physician assignment usually results in a leveling of work. To summarise this session, leveling creates even workflow, reducing stress, multi-tasking and errors. This approach produces the right number of workers at the right time to respond to customer demand. For emergency patients, the number of workers needs to match the daily demand patterns. Thank you.