June 18, 2012 Re-Design of a Pre-Admission Facility Interactive Quality Improvement Workshop Richard Bowry, MD Antoine Pronovost, MD Patricia Houston, MD
June 18, 2012 Re-Design of a Pre-Admission Facility Interactive Quality Improvement Workshop Richard Bowry, MD Antoine Pronovost, MD Patricia Houston,
Dec 14, 2015
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
June 18, 2012
Re-Design of a
Pre-Admission Facility
Interactive Quality Improvement Workshop
Richard Bowry, MDAntoine Pronovost, MDPatricia Houston, MD
June 18, 2012
St. Michael’s Hospital
Outline
1. Introduction to DMAIC methodology – Case study stem 1
2. Key concepts and facilitated discussion– stem 2
3. Process mapping exercise
4. Quantitative analysis, facilitated discussion
5. Quantitative analysis, group work– stem 3
6. Root cause analysis didactic session
7. Facilitated discussion: leading change… “what went wrong”– stem 4
8. Didactic session: key success factors for implementing and monitoring change
9. Conclusion
June 18, 2012
St. Michael’s Hospital
Disclosures
• Dr Richard Bowry– No disclosure
• Dr. Patricia Houston– No disclosure
• Dr Antoine Pronovost– Has received funding from the government of Ontario
to study and improve Pre-admission facility processes.
June 18, 2012
St. Michael’s Hospital
Objectives
• You will understand how to apply Quality Improvement techniques to the complex problem of redesigning a PAF
• You will become familiar with the five stages of DMAIC
• You will become familiar with the key principles of successful change management
June 18, 2012
St. Michael’s Hospital
Limitations and Caveats
• We will not be providing you with a “cook-book” answer for fixing problems in your own PAF– Solutions take teamwork, planning and local
insights to work
• The case study is loosely based on actual experience, but has been heavily adapted for the purpose of this session
Introduction to DMAIC
June 18, 2012
St. Michael’s Hospital
DMAIC - Define
• Reasons for action?• What are our targets?• What is within our control?
• All members need to agree on the problem• Create a purpose statement – rationale,
scope and targets• Start an A3 style grid to monitor progress
June 18, 2012
St. Michael’s Hospital
Define - A3
June 18, 2012
St. Michael’s Hospital
DMAIC - Measure
• What is our baseline?
• Acknowledge our own variation / trends?
• What happens 80% of the time?
• Root cause analysis
• Prioritization matrix
• Cause Effect Diagram
June 18, 2012
St. Michael’s Hospital
Prioritization Grid
June 18, 2012
St. Michael’s Hospital
Cause-Effect Diagram
June 18, 2012
St. Michael’s Hospital
DMAIC - Analyze
• What does our current state look like?
• Are there any wasted steps in what we do?
• How would a patient experience this?
• What are the root causes?
• Process mapping to identify NVA steps
• Holistic approach looking at all aspects
• Spaghetti Charts
June 18, 2012
St. Michael’s Hospital
DMAIC - Improve
• How should the future state look?• Use rapid process improvement cycles• Pilot and observe
• Remove unnecessary steps and create a future state
• No need to get it perfect first time• Implement pilots to assess impact
June 18, 2012
St. Michael’s Hospital
DMAIC - Control
• Re-evaluate and make ongoing changes• Monitor the new performance• Repeat the cycle as require to further
improve
• Reevaluate the changes and re-design as needed
• Repeat evaluation of process to assess impact
• Ongoing performance monitoring
June 18, 2012
St. Michael’s Hospital
Tool Matrix
June 18, 2012
St. Michael’s Hospital
Case Study Stem 1
• You have been asked to review your preadmission facility by your CMO because:– Patients are unsatisfied with long wait times– Surgeons offices are frustrated they cannot
access short-notice appointments• These are necessary to fill time released by last-
minute patient cancellations
– Staff complain of declining morale• Anaesthesiologists are reluctant to work in clinic
June 18, 2012
St. Michael’s Hospital
2. Facilitated discussion: Key concepts and tools to address this problem
• Perception shift: this is a chain, not a series of independent events
• Concepts:– Bottleneck– Batching
• Flow mapping: practicalities
June 18, 2012
St. Michael’s Hospital
This is a process, not a series of independent events
Anne M Breen, Tracey Burton-Houle, David C Aron,Applying the theory of constraints in health care: Part 1-- the philosophy, Quality Management in Health Care; Spring 2002; 10, 3;pg 40.
June 18, 2012
St. Michael’s Hospital
If each step has a measurable capacity, what determines overall throughput?
A. Average (13)
B. Highest cacapacity pacity (17)
C. Lowest capacity (8)
D. Cannot answer – need simulation model
The chain must be considered as a whole, not as a series of independent
events
Local optima don’t matter !
20
June 18, 2012
St. Michael’s Hospital
If bottlenecks limit throughput, why not simply eliminate them?
• Because in real life, systems need flexibility:– Ability to catch up = excess capacity– Need for excess capacity increases with system
complexity/variability
13 13 13 13 13
June 18, 2012
St. Michael’s Hospital
So what do you do with bottlenecks?
• Identify the bottleneck• Elevate the bottleneck• Design the process around the bottleneck
– Unload the bottleneck– Keep the bottleneck busy all the time
• This means non-bottleneck resources MUST sometimes be IDLE.
June 18, 2012
St. Michael’s Hospital
Batching: a very special effect on bottlenecks
• Batching refers to the processing of many units in a single group, for example:– I change all the ceiling light bulbs at the same
time because I need a stepladder (hard to get)
– Painting all similar colours together (trim, then walls, then contrast wall)
– Porters delivering multiple samples to the lab
June 18, 2012
St. Michael’s Hospital
Batching: advantages and disadvantages
June 18, 2012
St. Michael’s Hospital
Process mapping: putting it all together
June 18, 2012
St. Michael’s Hospital
Flow Mapping: Common Concerns
• What if I don’t get it right the first time?
• How do I keep people focused?– How do I frame the hypothesis?
• How much technical stuff do I need to know to participate or lead this discussion?
June 18, 2012
St. Michael’s Hospital
What if I don’t get it right the first time?
• Don’t worry… you won’t get it right the first time – That’s part of the plan…
• It’s an iterative process, and you’ll likely need a few drafts.
• It’s a group process, and much benefit comes from team discussion:
“Oh so that’s what happens when the patient leaves my care…”
June 18, 2012
St. Michael’s Hospital
• Set clear ‘start’ and ‘end’ points• Follow a single patient through a standard
encounter• Use Post-It notes on large paper
background
• Transcribe draft into clean computer after meeting
Basic approach to frame the process
June 18, 2012
St. Michael’s Hospital
How many fancy symbols do you need to master?
Terminator Defines start/end of process (only 2 per map)
Activity This is where work happens
Decision “a fork in the road”, best phrased as yes/no question
Flow Line Connect steps
June 18, 2012
St. Michael’s Hospital
• Please use this time to develop a process map in small group settings
• Use the data from case study stem 2 (next slide) as a starting point for your process map
3. Process Mapping exercise
June 18, 2012
St. Michael’s Hospital
Case study stem 2: Clinic details
• 60 patients are seen daily;• Patients are registered, then seen by a nurse,
then by a family doctor;• 50 % of patients seen by an anaesthesiologist;• Subgroups (orthopaedic and cardiac surgery)
patients also receive group teaching;– Other patients receive DVD-based teaching;
• Most patients receive bloodwork, and EKG +/- x ray investigations while in clinic.
June 18, 2012
St. Michael’s Hospital
Define – Process Mapping Exercise
• Three groups
• Map the current state
• Brief Presentation of processes found
June 18, 2012
St. Michael’s Hospital
4. Quantitative analysis: Facilitated discussion
June 18, 2012
St. Michael’s Hospital
Initial Thoughts
• Quick Fix approach vs Root Cause Analysis– Bottlenecks– Local optima vs global optimum– Non-value add activity– Batching
June 18, 2012
St. Michael’s Hospital
Define – Process Mapping Exercise
June 18, 2012
St. Michael’s Hospital
Define – Process Mapping Exercise
• Lessons Learned– Conventions in mapping– Importance to map out whole process
June 18, 2012
St. Michael’s Hospital
Measure
• Sources of Data
• IT/IM Resource
• Presentation of information
June 18, 2012
St. Michael’s Hospital
5. Quantitative analysis Group workCase Study Stem #3
• Quantitative Data to be provided in the following slides/handouts. Please review and discuss implications of quantitative data.
June 18, 2012
St. Michael’s Hospital
• Re-Design of a
• Pre-Admission Facility
Stem #3: Quantitative Data (Continuation)
RN Wait Time RN Encounter Time
Mean 13.7 min 32.5 min
Median 10 min 30 min
Standard Dev. 9.9 min 12.9 min
Resource Availability
8 Nurses
Throughput 14.8 patients/hours
June 18, 2012
St. Michael’s Hospital
Stem #3: Quantitative Data (Continuation)
FMD Wait Time FMD Encounter Time
Mean 21.5 min 7.6 min
Median 20 min 6 min
Standard Dev. 17.1 min 3.9 min
Resource Availability
1 FMD
Throughput 7.9 patients/hour
June 18, 2012
St. Michael’s Hospital
Stem #3: Quantitative Data (Continuation)
Anaesthesia (AN) Wait Time
AN Encounter Time
Mean 27.6 min 12.3 min
Median 20 min 10 min
Standard Dev. 21.9 min 5.7 min
Resource Availability
1-2 AN
Throughput 4.9 patients/hour (1 AN)
June 18, 2012
St. Michael’s Hospital
Stem #3: Quantitative Data (Continuation)AN wait time by scheduled time of day
June 18, 2012
St. Michael’s Hospital
Throughput balancing: find the bottleneck
June 18, 2012
St. Michael’s Hospital
Measure – Data Interpretation
• Wait-time and value-add times
• Satisfaction
• Capacity analysis
• Scheduling
• Variability
June 18, 2012
St. Michael’s Hospital
6. Root Cause analysis
June 18, 2012
St. Michael’s Hospital
Analyze – Root Cause Analysis
Lack of Consistent
Triage Process
Redundancy in Validation
of Patient Information
Lack of Standardized
Forms
Gaps in Patient Education
Merging of Patient
Information
Triage/Wtg Rm Traffic Directed by
RN
Many ways to get info for Pt Reg
Multiple Phone Calls, Interruptions
Data Entry
Multiple Competing
Duties
Repetitive Collection of Pt Demo
Rework, competing priorities, and
interruptions at triage slows down the overall process and adversely affects staff and patient
satisfaction.
Excessive waiting time along with a confusing
process for patients affects patient
satisfaction within the ED.
Redundancy in information gathering along with seeking out
information through different channels, causes delays and
frustration for staff and patients. There is an
increased risk for errors.
Multiple Competing IT
Systems
Patient Registration
Seeking Add’l Info
Continuous EDIS vs ADT Reconciliat’n
Multiple Entry Points for ED
Patients
June 18, 2012
St. Michael’s Hospital
Analyze – Theory of Constraints
1. Identify the Constraint
2. Exploit the Constraint
3. Subordinate everythingto the Constraint
4. Elevate the Constraint
5. Repeat for the new Constraint
June 18, 2012
St. Michael’s Hospital
Analyze – Computer Simulation
June 18, 2012
St. Michael’s Hospital
7. Facilitated discussionCase Study Stem #4: Le denouement
• Suggestions are implemented, but results are not anticipated– Wait times increase– Throughput decreases
• Morale deteriorates significantly– Staff, especially RN’s leave their positions leaving
unfilled vacancies– Much finger-pointing/blaming ensues
June 18, 2012
St. Michael’s Hospital
8. Key success factors for implementing and monitoring change
June 18, 2012
St. Michael’s Hospital
Improve – Stakeholder Engagement
• Engage in issues that matter
• Use Engagement to drive decisions
• Engage the right stakeholders
• Engage empowered representatives
• Seek shared values
• Agree on the rules of engagement
• Manage expectations –provide adequate resources
June 18, 2012
St. Michael’s Hospital
Improve – Stakeholder Engagement
• What stakeholders need:– Fairness– Listen– Build Trust– Be open– Be accountable– Evaluate
June 18, 2012
St. Michael’s Hospital
Improve – Change Management
• Establishing a Sense of Urgency• Forming a Powerful Guiding Coalition• Creating a Vision• Communicating the Vision• Empowering Others to Act on the Vision• Planning for and Creating Short-Term Wins• Consolidating Improvements and Producing
Still More Change• Institutionalizing New Approaches
Kotter, Leading Change 1996
June 18, 2012
St. Michael’s Hospital
Control - Sustainability
June 18, 2012
St. Michael’s Hospital
Improve – Unintended Consequences
• Balanced Scorecare
June 18, 2012
St. Michael’s Hospital
Improve – Measuring Success
June 18, 2012
St. Michael’s Hospital
Control – Control Charts
CTAS 1-3 Performance (percentage met EDLOS < 8 hours)CTAS 4-5 Performance (percentage met EDLOS < 4 hours)
Apr '08 to Aug '10
89
76
82
5250
55
60
65
70
75
80
85
90
95
100
Apr
-08
May
-08
Jun-
08
Jul-0
8
Aug
-08
Sep
-08
Oct
-08
Nov
-08
Dec
-08
Jan-
09
Feb
-09
Mar
-09
Apr
-09
May
-09
Jun-
09
Jul-0
9
Aug
-09
Sep
-09
Oct
-09
Nov
-09
Dec
-09
Jan-
10
Feb
-10
Mar
-10
Apr
-10
May
-10
Jun-
10
Jul-1
0
Aug
-10
%
CTAS 1-3 CTAS 4-5
June 18, 2012
St. Michael’s Hospital
Conclusion
• DMAIC Methodology
• Stakeholder Engagement
• Leading Change
• Measuring Success
• Importance of Value Add
June 18, 2012
St. Michael’s Hospital
• The following slides can serve to supplement case discussion.
Appendix
June 18, 2012
St. Michael’s Hospital
Theory of Constraints asserts that in the real world a ‘balanced plant’ will self-destruct
Statistical variability: Throughput at each step varies around a mean
+
Dependent events: a downstream process cannot occur before its upstream precursor
=
Small gaps build up to infinity unless there is reserve capacity
June 18, 2012
St. Michael’s Hospital
Consider the famous example of a group of hikers
• Scouts are heading on a 5 mile hike
• They must walk single file– They cannot pass each other (dependent
events)
• Each hiker walks at a similar pace, but there is some variation– Each time a scout stumbles or slips, he loses
some ground
June 18, 2012
St. Michael’s Hospital
Diagram of the ‘Goldratt’ hike
Direction of hike
S S S S S S S S S S S S S S S S S S S S S S S
S S S S S S S S S S S S S S S S S S S S S S S
Start
After 1 hour
June 18, 2012
St. Michael’s Hospital
Conclusions from the hiking example
1. Over time, the scouts will continue to spread;
2. To keep the group compact, one must place the slowest hiker (bottleneck) at the front.
June 18, 2012
St. Michael’s Hospital
So how do you identify bottlenecks?
• In the hiker example, you look for a large gap in front of a scout
• In a plant, you might look for a large pile of inventory in front of a particular station
• In a hospital, you could look for a large number of (angry looking) patients in a waiting room
June 18, 2012
St. Michael’s Hospital
Operational management requires awareness of two key elements
• Variability: Statistical variation and dependent events
• Bottlenecks: Bottlenecks are neither good nor bad
June 18, 2012
St. Michael’s Hospital
Batching: a very special effect on bottlenecks
• Batching refers to the processing of many units in a single group
• All units have the same start/finish times
• Batching is highly effective when setup costs/setup time are high
June 18, 2012
St. Michael’s Hospital
Batching cupcakes:
June 18, 2012
St. Michael’s Hospital
As a cupcake-baker, batching is great because:
• I mix one batch of batter, drop it into moulds, place in the oven, and I’m done;
• I only have to run the oven once (lower energy costs );
• This is a ‘locally optimal’ solution.
June 18, 2012
St. Michael’s Hospital
As a cupcake-decorator, batching is terrible:
• At first, I have no work to do while the cupcakes are baking
• Then I suddenly have 20 cupcakes to decorate.
June 18, 2012
St. Michael’s Hospital
How does this come together?
• Assume baking a batch of 20 cakes takes– 15 minutes prep + 45 minutes baking
• Assume decorating takes 5 minutes per cake
• How long would it take to make a single batch of 20?
June 18, 2012
St. Michael’s Hospital
Answers:
A. 5 minutes/cake x 20 = 100 minutes
B. 3 minutes/cake x 20 = 60 minutes
C. 60 minutes + 5 minutes/cake x 20 = 160 min
June 18, 2012
St. Michael’s Hospital
Answer is D 160 minutes
• This results in cupcake cycle time of 160/20 = 8 minutes per cake
• That doesn’t seem so bad…
June 18, 2012
St. Michael’s Hospital
When was the first cupcake ready?
• 60 + 5 = 65 minutes
June 18, 2012
St. Michael’s Hospital
When was the last cupcake ready?
60+100 = 160 minutes
Time for 10th cupcake
60+(10x5) = 110 minutes
June 18, 2012
St. Michael’s Hospital
Why might this be a problem?
• Assume cupcakes are shipped from the kitchen in batches of 20:– What if a walk-in client wants to pickup 6
cupcakes:• It takes almost 3 hours for the first (and last) cake
to be ready
– What if the cupcakes sell best when they are fresh (< 45 minutes from the oven)
June 18, 2012
St. Michael’s Hospital
What are possible solutions?
• Have the cake-decorator start/finish 1 hour after the cake-baker
• Have a cake ‘reserve’ for the decorator– ‘buffer’ in operations– parallel in health care: waiting room for
patients
• Make smaller batches– The ultimate small batch is a single unit– Might reduce batch size after decoration
June 18, 2012
St. Michael’s Hospital
What is the product at the end of the 8-hour day?
Baking
• 8 hours/batch x 20 cakes/batch = 160 cakes
Decorating
• 7 hours (1 lost hour) x 12 cakes/hour = 84 cakes
Total
• 84 finished cakes
• 76 ‘waiting’
Related Documents
![A4B4-Murakami-PP [Read-Only] - Stevens Point€¦ · • Armstrong Institute for Patient Safety and Quality (Dr. Peter Pronovost) • Center to Eliminate Cardiovascular Health Disparities](https://static.cupdf.com/doc/110x72/5b1cb3767f8b9af2348bf612/a4b4-murakami-pp-read-only-stevens-point-armstrong-institute-for-patient.jpg)
![Pronovost [ppt 918kb]](https://static.cupdf.com/doc/110x72/554b7bd9b4c90564168b58b8/pronovost-ppt-918kb.jpg)
