Chapter 5: Lean and Six Sigma Management: Building a Foundation for Optimal Patient Care Using Patient Flow Physics
Outline
Introduction
Lean and Six Sigma Management Defined
Lean Management System (LMS)
Introduction
Continuous Quality Improvement (CQI), Lean Management Systems (LMS), and Lean Six Sigma (LSS)
Philosophies and methods for leadership, management, improvement, and innovation
Offer approach, set of tools, and way of thinking about how to more effectively study, assess, and improve clinical flow
LMS and LSS are also broadly generalizable to other health applications
Lean and Six Sigma Management Defined
Six Sigma
Utilizes statistical methods to identify and reduce variation in processes
Developed by Motorola as an evolution of quality methods from the United States and Japan
Muda, muri, and mura
Developed by Toyota Production System
Emphasizes elimination of waste
Muda: Wasting Resources
Elimination of waste in seven resources
Transportation
Inventory
Motion during production
Waiting time spent during production
Over-processing
Over-production
Defects
Examples of Waste in Health Care
Wasted inventory
Rework
Excess waiting time
Lost time
Errors
Extra work associated with poor processing of information or outdated procedures
Waste from transporting patients unnecessarily
Muri: Overburdening Staff and Equipment
Evident when physicians and staff feel overwhelmed with the level of effort required for them to:
Provide patient care
Document care provided
Address regulatory reporting requirements
Attend required meetings
Provide all the support beyond the effort required for patient care
Mura: Uneven Process Workflow
Reason for the development of “just-in-time” production systems
Tendency to batch up work and send it in “one pile” to the next step in the process
Examples:
Batch reading of radiology exams
Morning blood draws
How Six Sigma Originated
Developed by Motorola after studying quality methodology practiced in Japan
Selected skilled staff with technical backgrounds and training them in advanced quality, engineering, manufacturing, and statistical methodology
Total Customer Satisfaction (TCS) Improvement Goals
Goals:
Reduce all defects 10-fold every 2 years
Reduce cycle time of all core processes by 50% within 2 years
Recognized that:
Variation in process performance can lead to greater defect/error rates
Speeding up processes requires the elimination of unnecessary tasks or steps
Principles of LMS Theory
The customer defines value.
Organizations should identify the value stream in every process and seek to eliminate all wasted steps that do not add value.
Organizations should make the product (management of the patient’s care) flow through the steps of the value-added process continuously.
Organizations need subsequent steps in the process that “pull” products (patients) from the previous step rather than push them forward.
Organizations should work to continuously improve in order to reduce steps, time, costs, and the information (data) needed to serve the customer (patient).
Lean Six Sigma (LSS)
Michael George wrote that Lean Sigma Six:
Maximizes shareholder value by achieving the fastest rate of customer satisfaction, cost, quality, process speed, and invested capital
The fusion of LMS and Six Sigma is required because:
LMS cannot bring a process under statistical control.
Six Sigma alone cannot dramatically improve process speed or reduce invested capital.
Implementing LMS
Jackson and Jones wrote that organizations cannot incorporate LMS overnight.
Management must build, nurture, and support the logic and machinery that drives lean production.
Combination of LMS and Six Sigma
Now being used in health care applications
Provides a synergistic methodology for analyzing and reducing or eliminating waste in health care processes
Patient Flow as an Application of LSS
Improving process “flow” is a primary goal.
Changing organizational culture is key.
Organizations must embed a culture that:
Identifies wasteful effort
Seeks to continually improve existing processes
Designs processes for those that do not currently exist
The “ideal state” can be accomplished by integrating patient flow physics—principles that govern patient flow and variation management
Applying LSS to Patient Flow in Community Hospital Emergency Department
Approximately 65% of patients admitted to the hospital arrive via the ED.
Long wait times contribute to patient dissatisfaction and staff and community frustration.
Factory Physics modeling
Used to improve understanding of patient flow through ED
Box plot
Introduced by John Tukey
Uses robust summary statistics and actual data points
Especially useful for comparing distributions across groups
Box Plot of 2002-2005 Patient Arrivals by Hour of Day
Box Plot of 2002–2005 Patient Arrivals by Hour of Day
Emerging Questions
How does this demand pattern support decisions and lead to more effective staffing up- and downstream?
What are the optimal levels of staffing to meet patient demand?
Does the nature of the demand vary over the course of the day?
Are there certain “chief complaints” or issues that are seen more often during a certain time of the day or on particular days of the week?
Does the type of staff skills needed differ at various times of the day or on certain days of the week?
What should the organization do, if anything, to be prepared to handle the variation and demand outliers including during mass casualty or pandemic events?
Impact of Applying LSS to Patient Flow in Community Hospital ED
Incorporating process and flow data can:
Facilitate effective handling of patient demand and utilization of resources
Provide input as to where process improvements can be applied to reduce waiting time, poor staff utilization, and other wasteful aspects
Applying Factory Physics principles led to:
Better-managed patient volume
Better provider and staff scheduling
Higher patient and care provider satisfaction
Applying LSS to Patient Flow in Community Hospital Mammography Service
Beginning of effort:
Took 8 weeks to move through the diagnostic journey to treatment
After assessment and utilization of LSS:
Journey reduced to one week or less
Significant because earlier identification and treatment of cancer increases the likelihood of clinical outcomes, satisfaction, and overall quality of life
The Patient Flow Physics Framework
Recognizes all processes consist of:
Demand
Transformation
Inventory
Must be able to characterize and manage patient flows in three arenas:
Flow into the patient care process
Flow through the patient care process
Flow out of the patient care process
Patient Flow Into the ED
Consequence of patient flows from:
Emergency medical services
Nursing homes
Transfers from other hospitals
Direct patient admits (outside physician referrals)
Walk-in patients
Patient Flow Through the ED
Function of:
ED resource utilization
Variability associated with differences between ED providers
Physicians, physician assistants, and nurse practitioners
Beds are a key ED resource.
Rule of thumb:
Vertical patients should remain vertical.
Horizontal patients should remain horizontal.
Patient Flow Out of the ED
Crucial to avoid overly long stays
ED boarded patients
Admitted to hospital, yet “blocked” from being transported to in-patient units
Reasons for ED boarding:
Lack of transport personnel
Unavailability of consulting physicians (consults)
Insufficient beds
Significant issue during busy winter season
Patient Flow and the VUT Equation
Average patient flow care time is increased by anything that:
Increases variation (V)
Increases the utilization of resources (U)
Operates in processes designed to take a long time (T)
Average Patient Flow Care Time equals:
Variation (V) Utilization (U) Process Time (T)
Variation
Two types:
Common cause variation or natural variation
Special cause variation or artificial variation
Causes can be detected with control charts
Analysis must be applied to a specific process, individual or group, period of time, or the use of specific material or equipment
Causes of Variation
Demand for patient flow into the processes
Differences in patient needs
Differences in how care is delivered by clinical staff
Reliability of patient care and handoffs to avoid errors, and rework
Availability of resources outside current patient care processes
Strategies for Managing Variation
Learn the patterns of demand.
Standardize care for typical patients.
Put population data to work.
Learn from other organizations dealing with high variability.
Create buffers.
Utilization
Impacted by:
Available staffing capacity each hour to handle variation in demand
Availability of resources
Sufficient availability of inventory
Financial resources to access services
The time to complete clinical and administrative activities
Resources for training and other staff development
Capacity
Must be available across the basic value stream
Can address by building in “slack”
Uncommitted staff, space, and equipment
Generic High-Level Health Care Value Stream
Strategies for Keeping Utilization of Key Resources Low
Keeping key resources utilized less than 80% of the time
Understanding variation in demand and patients
Applying box plots of variation
Time
Includes activities or tasks within each step and the space between steps
Delays can occur during handoffs and transitions.
Group boundaries also contribute to “white space” management challenges.
Reducing transition time between steps helps:
Reduce patient anxiety
Determine clinical concerns earlier
Reduce the effort for treatment needed if caught earlier
Honor the patient’s personal time
Impact of the Process on Time
The design of a patient care process can determine the minimal amount of time a patient will spend, flow through, and out of a process if there were no impediments or variation in the process.
Addition of nonvalue-added steps in the process may only add time, potentially increases variation, and increased opportunity for errors.
Reducing the Length of Time Spent in the Patient Flow Care Process
May require staff to:
Eliminate or reduce nonvalue-added process tasks or steps, such as unnecessary waiting times
Implement LMS to draw attention to time spent on identifying and reducing nonvalue-added process steps
Create value-stream managers that control all resources needed for their service lines
Lean Management System (LMS)
Major functions of management in Lean organizations:
Maintaining or controlling the existing processes
Improving existing processes
Performance management
Occurs through continuous improvement
Emphasizes staff development, socio-technical (belief) systems, and change management
Lean Management System Model
Lean Management System Model
Four P Model
Problem solving for continuous organizational learning and alignment to objectives and plans
People are respected, challenged, engaged and developed
Processes are improved and stabilized based on customer defined “value”
Philosophy of long-term focus
Leader Standard Work
In LMS, best practices or standards are clearly spelled out in detail and often posted in activity areas.
A lack of standardized work increases variation and can produce very strange data distributions.
A generalized approach can restrict flow and slow down processes.
Leader Standard Work for Activities Completed Many Times Each Day
Leader Standard Work for Activities Completed Many Times Each Day
Modified from Creating a Lean Culture (Mann, D., 2015, pp. 58–59).
Leader Standard Work for Activities Completed at Least Once Daily
Leader Standard Work for Activities Completed At Least Once Daily
Modified from Creating a Lean Culture (Mann, D., 2015, pp. 58–59).
Leader Standard Work for Activities Completed Weekly or Less Frequently
Leader Standard Work for Activities Completed Weekly or Less Frequently
Modified from Creating a Lean Culture (Mann, D., 2015, pp. 58–59).
Direct Contact Between Management and Frontline Workers
Builds trust
Provides opportunities for much learning, data collection, and problem identification before it is filtered by middle management
Allows managers to directly observe how buffers are managed
Helps managers understand the value streams
Motivates staff to stay focused and continually improve
Visual Controls
Should always begin with some type of “as is” value stream or process map
Process map or flowchart
Visual representation of the care process created with information provided by team members
Helps clinicians:
Clarify through visualization what they know about their environment
Determine what they want to improve about it
ED Process Map
Emergency Department Process Map Example
Health Care Metrics
Ideal health care metrics might include:
Cycle time
Patient wait times
Medical errors
Scheduling errors
Patient volumes
Provider volumes
Takt times
Every metric should include a numerical target.
Need to be “stretch goals” to help motivate improvements
Data Collection
Ideally, data would be automatically collected by the electronic health record (EHR).
Data points need to be collected for every patient or transaction for any desired time period.
Data needs to be easily exported into a standard data format that facilitates data analysis.
Huddle Board Template
Huddle Board Template
Run Chart
Average ED process cycle times for sampled patients on November 15, 2016
Average ED Process Cycle Times for Sampled Patients on November 15, 2016
Control Chart
XbarR door-to-doctor times for sampled patients on November 15, 2016
XbarR Door-to-Doctor Times for Sampled Patients on November 15, 2016
Box Plot
Box plot for physicians to assess their decision times
Box Plot for Physicians to Assess Their Decision Times
Pareto Chart
Missing chart elements from a coding QA review
Missing Chart Elements From a Coding QA Review
Pitch Chart
Pitch chart for RCM chart reviews
Pitch Chart for RCM Chart Reviews
Training Matrix
Staff skills and cross-training matrix
Staff Skills and Cross-Training Matrix
Overlay Run Chart
Overlay run chart for number of patients and staff by times
Overlay Run Chart for Number of Patients and Staff by Times
Staff Suggestions and Their Implementation
Staff Suggestions and Their Implementation
Trend Chart for Suggestions Made and Implemented by Staff
Trend Chart for Suggestions Made and Implemented by Staff
Daily Accountability
Management actively engages with staff:
When they “go to the gemba”
During huddles
During regular follow-up discussions on clinical improvement projects
Going to “where the work occurs” is the best place to collect real time data.
Waste Walk Tool
Waste Walk Tool
Questions to Ask During Gemba Walks
What are your team’s targets or goals today?
How well are you meeting these targets (facts)?
What is your plan or planned actions to close the performance gap?
How can I help you implement your ideas?
Huddle Boards
Focus of 10- to 15-minute stand-up meetings that might involve different staff at different levels
Opportunity to:
Briefly review the highlights of KPI data
Focus on positive and negative variation
Examples of Huddle Boards
Huddle Board Example 1
Huddle Board Example 2
Coaching
Preferred way method of staff development and improving accountability
Managers show workers respect by:
Asking how something should be done rather than telling them the correct way to complete a task
Modeling appropriate behaviors
Using shaping and modeling to help staff perfect new skills
Conclusions
LMS and LSS offer an approach, a set of tools, and a way of thinking about how to more effectively assess and study clinical flow.
These methodologies are widely applicable in studying and improving quality and efficiency of health processes.
Improving clinical teamwork is an important factor in improving patient outcomes.
