Flow Mini Course M11

1
Flow Mini CourseM11
Paris 25th April 2008

• Carol Haraden PhD
• Kate Silvester BSc MBA FRCOphth
• (With thanks to Richard Steyn MS FRCSEd(C-Th)
  FIMCRCSEd MRCGP)

2
Agenda

• Running the UK NHS
• Exposing the problem 20 minutes
• Understanding the problem 40 minutes
• Solving the problem 15 minutes
• Questions and Discussion
• Close at 5.30

3
UK Healthcare System (April 2002)
4
Exposing the ProblemRunning the UK NHS
Computer Model Demonstration
1
2
3
4
5
Go to www.steyn.org.uk/models/patflow.ppt
You cant use these models for real life data
5
UK Healthcare System July 2005
Deaths
Long term care
Intermediate Care
ER
Assessment Units
Wards
General Practice
Procedures
Planned admissions
Imaging and Tests
Specialist Clinics
6
Results for the UK NHS?

• Cost
• 61 Billion to 136 Billion p.a.
• Staff
• Frustration and employment issues
• Patients
• persistent delays
• 18 weeks for planned care
• 4 hours emergency care
• UK Government targets

7
You will never solve the problem with the mindset
that created it
Albert Einstein
8
Understanding the Problem

• Why do we get lines, queues and waiting lists?

9
Demand and Capacity Definitions
10
Demand and Capacity
For model go to www.steyn.org.uk/models/demand
analysis.xls
You cant use these models for real life data
11
Av. Demand gt Av. Capacity
For model go to www.steyn.org.uk/models/demand
analysis.xls
You cant use these models for real life data
12
Variation Mismatch
For model go to www.steyn.org.uk/models/demand
analysis.xls
You cant use these models for real life data
13
If av. Demand av. Capacity, variation mismatch
queue
Cant pass unused capacity forward to next week
14
Demand capacity for a clinic
15
Hospital admission / discharge variation mismatch
16
Responding to the Queue

• Increase capacity
• Short term
• Overtime Waiting list initiatives
• Agency staff
• Long term
• Business case how much?
• Prioritising
• Carve Out

17
Impact of carve out
For model go to www.steyn.org.uk/models/demand
analysis.xls
You cant use these models for real life data
18
Carve out in CT
Scanner 1
Scanner 2
Mathematically impossible to balance this number
of queues
19
The Flaw of Averages
Plans based on averages
Short term increased capacity
Increased cost
20
What should we do instead?
For model go to www.steyn.org.uk/models/demand
analysis.xls
You cant use these models for real life data
21
Pooling
For model go to www.steyn.org.uk/models/demand
analysis2.xls
You cant use these models for real life data
22
Pareto Analysis
15
30
Glenday Sieve
23
Planning the right capacity
For model go to www.steyn.org.uk/models/demand
analysis.xls
You cant use these models for real life data
24
Erlang
Service Failure Waiting or Defaults
Cant run a system beyond 80 to 85 utilisation
without service failure Good practice SNCF
French trains
20
40
80
100
60
utilisation
25
Trade Off

• Cost of slack capacity
• Versus
• Cost of managing the queue
• Clinical risk

26
Reduce variation to reduce cost
For model go to www.steyn.org.uk/models/demand
analysis.xls
You cant use these models for real life data
27
Reduce Variation Mismatch

• Understand Demand
• Predictable variation?
• match capacity
• Unpredictable variation?
• slack capacity

28
Predictable and Unpredictable Variation in Demand
Weekly Emergency Room Attendances
Total attendances
Minors Discharged
Majors Require admission
29
Reduce Variation

• Reduce variation in
• CAPACITY
• Demand (look at upstream capacity)

30
Reduce Variation in Capacity better DC match
2 big clinics Mon Wed
3 small clinics Mon, Tues, Wed
Total number of
referrals (Demand)
Number of slots
available (Capacity)
Result 3rd next available appointment within 5
days
31
Good management is all about reducing variation

• Deming

32
Solving the Problem

• Improving Patient Flow 2

33
Patient Flow
Computer Model Demonstration
1
2
3
4
5
Go to www.steyn.org.uk/models/patflow.ppt
You cant use these models for real life data
34
UK Healthcare System The way forward
Long term care
Intermediate Care
Home
Assessment Units
Wards
ER
General Practice
Procedures
Imaging and Tests
Wards
Specialist Clinics
Procedures
35
More Information

• www.steyn.org.uk
• Presentations
• Models
• Guides
• Feedback, questions and discussion
• Reading list
• Links to other practitioners
• Contact us