Pathology! by pptfiles

VIEWS: 147 PAGES: 54

									Pathology
23rd July Paul Walley, Kate Silvester, Martyn Cooke

K. Silvester Pathology KS V2 170704

Objectives of this session:
• Pathology is central to the clinical process
– Not „support service‟!

• Introduce the concepts behind service improvement
– Systems design to eliminate waste

• Measuring improvement
K. Silvester Pathology KS V2 170704

The process view:
• Number of steps from request to result being „acted‟ on for the patient. • High level ( helicopter view) • Detailed level: each step performed by one person in one place at one time

K. Silvester Pathology KS V2 170704

Process maps:
Test requested Lab

Action for patient

Lab 2 Impact on time, cost and quality of the whole process – not just the lab bit
K. Silvester Pathology KS V2 170704

High level view: 4 to 8 steps
Request test Take specimen Transport specimen Register specimen

Process specimen
Report result Transport result Act on result
K. Silvester Pathology KS V2 170704

High level view: 4 to 8 steps
Supplier
Doctor/nurse/patient Doctor/nurse/phleb Porter/air-tube/transport Lab reception Lab reception Lab technician lab tech/pathologist/doctor/ nurse Porter/air-tube/transport
K. Silvester Pathology KS V2 170704

Customer Request test
Doctor/nurse/phleb Porter/air-tube/transport Lab reception

Take specimen Transport specimen Register specimen
Process specimen Report result Transport result Act on result

Lab technician
Lab technician lab tech/pathologist/doctor/nurse Porter/air-tube/transport Doctor/nurse/patient

What does this show?
• What should happen: the basic process However • Variation in system!
– How can we simplify this?
• Can we got down to one customer/supplier relationship for each step?

• Who we need at the party to sort the system out!
K. Silvester Pathology KS V2 170704

Input / outputs: 4 to 8 steps
Input
Form, patient details, test required Completed form, equipment Specimen, completed form, transport Specimen, completed form,

Customer Request test
Completed form

Take specimen

Specimen, completed form, transport
Specimen, completed form, Specimen, completed form, previous specimen details
Processed specimens, completed form, previous specimen details completed form, previous reports, report previous reports, report Changed or unchanged diagnosis / treatment plan

Transport specimen Register specimen
specimen result

Specimen, completed form, Process previous specimen details, kit

Processed specimens, completed form, previous specimen Report details, previous reports, kit previous reports, report
K. Silvester Pathology KS V2 170704

Transport result Act on result

previous reports, report

Input / output method shows
• What is required at each basic step • Time cost and quality:
– – – – – Where does each bit come from? Who supplies it? When is it supplied /delivered How is it supplied / delivered How much does it cost
• Unit cost, storage cost, waste cost

– Quality of supply
• errors
K. Silvester Pathology KS V2 170704

Detailed process mapping
• We know what the 4 to 8 big steps are • We have got every one to a process mapping session. • We know what the key inputs and outputs are • We know where main errors/delays are NOW…
K. Silvester Pathology KS V2 170704

Detailed process mapping
• What happens in each „big step?‟
– what one person does – in one place (or one piece of kit) – at one time

• Identifies:
– Data items required – Kit required to perform each step
K. Silvester Pathology KS V2 170704

High level to detail Start Request test Take specimen

Process specimen

Register specimen

Transport specimen (& Request)

Report result

Transport report

Act on report

End result

K. Silvester Pathology KS V2 170704

detailed level: activities performed by one person in one place at one time Start trigger

End result Count the changes between the people involved
K. Silvester Pathology KS V2 170704

27 changes between 11 roles to write and send a letter 500 yards down a corridor = 6 to 10 weeks

Hot spots:
Where are the inventories / delays Time for whole process
Potential solutions

Start trigger

End result What the common errors are and where they occur
K. Silvester Pathology KS V2 170704

Potential solutions

Detailed process mapping shows:
Waste • Repeated steps • Delays • „inventory‟
– stored specimens, stored results…..

• What is often missing –impacts quality • We have „brilliant people working in „…p‟ processes‟.
K. Silvester Pathology KS V2 170704

Simplify the system Remove steps! Reduce errors and delays Start trigger

End result

K. Silvester Pathology KS V2 170704

Lots of different ways of doing it.
Important that everyone sees the process from end to end • Brown paper and post its • Walk the process
– See it! – (photograph each step)
K. Silvester Pathology KS V2 170704

Process map review
• Martyn go through photos
– Haematology – Chem path – Microbiology – Histo

• Exercise 1: Spot the waste!
K. Silvester Pathology KS V2 170704

The wastes
Overproduction Waiting Transporting Inappropriate Processing Unnecessary Inventory (backlog) Unnecessary Motion Defects
K. Silvester Pathology KS V2 170704

Untapped Human Potential Inappropriate systems Energy and Water Materials Service and Office Wastes Customer Time Defecting Customers

Complexity mapping
• Walk the „floor‟ from end to end • Visibility of the system
– If we want to reduce steps, how could we do it?

K. Silvester Pathology KS V2 170704

Complexity system for A&E department showing processes for patient with chest pain and patient with injured wrist Out of hours GP service reception Resuscitation
Dr desk

loo
Waiting area with chairs triage

Teaching room

Shared resource Triage nurse
offices Plaster room in out K. Silvester Pathology KS V2 170704 patients

ecg

Blind alley full of filing cabinets

2 Plain film x-ray rooms

Observation ward

waiting
Trolley, beds or examination couches/chairs

Complexity system for A&E department after change showing processes for patient with chest pain and patient with injured wrist Out of hours GP service reception Resuscitation
Dr desk

loo
Waiting area with chairs

Teaching room

ecg offices Plaster room in out K. Silvester Pathology KS V2 170704 patients Observation ward

2 Plain film x-ray rooms

waiting
Trolley, beds or examination couches/chairs

Exercise 2: Floor plan
• Map the flow • Where do the samples get held up?
– (backlog)

K. Silvester Pathology KS V2 170704

Paul Walley

K. Silvester Pathology KS V2 170704

Linking process requirements to systems design
1. Sense of value for different streams for work for different customers
time: GP, ward patient , chemo, serious trauma quality: microbiology see & treat versus see, culture and treat.

2. Volume and variety: Pareto analysis
Flow control

3. Service blue printing the patient‟s experience
K. Silvester Pathology KS V2 170704

Service Blueprinting
Patient can’t find parking space Can’t find blood clinic Patient told they are late

F What the patient sees

F F

W

F

Patient bled after watching 15 bloods taken

W

Patient not told of likely wait

W
Result lost

What pathology sees F Failure point

Sample arrives late

W
Blood processed

Report left in letter rack

F

W Waiting point
K. Silvester Pathology KS V2 170704

Performance Objectives
Different segments of work may require very different emphasis on performance. It is useful to articulate this before you make process design decisions: How does speed requirement Cost vary for FBC: GP request? Chemotherapy? Speed Quality Serious trauma? Do some have to be more dependable? Flexibility
K. Silvester Pathology KS V2 170704

Dependability Can we mix these segments in the same process?

The relationship between volume and variety dictates resource configuration to some extent at least… High
Project Jobbing

Batch Variety

Mass
Continuous

Low Low

K. Silvester Pathology KS V2 170704

Volume

High

But we immediately find that we have a compete mix of volumes. Simple “Pareto analysis” usually gives us this profile…

Cu. volume

80% of the work 20% of the causes

Ranking by volume
K. Silvester Pathology KS V2 170704

Here is one way of summarising how volume affects process configuration… Line Job/batch Cellular processes

Increasing volume
K. Silvester Pathology KS V2 170704

Server

Server

Server

K. Silvester Pathology KS V2 170704

Queue type A

Server

Queue type B

Server

Server

Server

Server

Demand and capacity
Why do we get queues? • Demand > capacity • Variation mismatches between D & C • To keep us 100% utilised (unit cost)

K. Silvester Pathology KS V2 170704

www.steyn.org.uk
100 90 80 70 60 50 45 40 35

50 40 30 20 10 0

25 20 15 10 5 0

1

21

41

61

81

101

121

141

161

181

week demand waiting list waste

K. Silvester Pathology KS V2 170704

201

demand/waste

30

waiting

Variation mismatch = queue
Queue

Demand

Capacity

Can‟t pass time unused capacity forward to next week
K. Silvester Pathology KS V2 170704

Equilibrium
River flows in
Demand Did Not Attend Cancellations Evaporation Deaths

Water level stays the same River flows out

Activity
K. Silvester Pathology KS V2 170704

Process requirements:
• Who are the different customers? • What service specification do they need?
– GPs – Laboratory
• urgent • Routine

– Patients – & Etc
K. Silvester Pathology KS V2 170704

…?

Design the ideal lab! (ideal lab)
• What do we mean by ideal?
– 24 hrs turnaround: – Efficient but takes a week

• Flexibility
– Tesco experience of lean – Batching and arrivals

• 1 or 2 process streams?
K. Silvester Pathology KS V2 170704

What are we trying to improve?
• Measures for improvement

K. Silvester Pathology KS V2 170704

Measure for Judgement
Process times for sample
80 70 60 Time (Days) 50 40 30 20 10 0
K. Silvester Pathology KS V2 170704

70

35

Laboratory A

Laboratory B

So what?
• Can „customers‟ be sure that either laboratory has improved? • Do they care about LaboratoryB?
• How does this help LaboratoryA or LaboratoryB improve?
K. Silvester Pathology KS V2 170704

Improvement in process time (Laboratory A)
80

70
60 Time (Days) 50 40 30 20 10

70

35

0
Before Change
K. Silvester Pathology KS V2 170704

After change

Can patients be sure LaboratoryA has improved?

Improvement in process time (LaboratoryA)
100 90 80 70 60 50 40 30 20 10 0

Time (Days)

Change Implemented
Oct Aug Nov Jun May Mar date Sep Feb

Yes: definite improvement as a result of change
K. Silvester Pathology KS V2 170704

Dec

Apr

Jan

Jul

Improvement in process time (Laboratory B)
80 70 60 Time (Days) 50 40 30 20 10 0
Before Change
K. Silvester Pathology KS V2 170704

70

35

After change

Can patients be sure LaboratoryB has improved?

Improvement in process time (LaboratoryB)
100 90 80 70 60 50 40 30 20 10 0

Time (Days)

Change Implemented
Oct Aug Nov Jun May Mar date Sep Feb Dec Apr Jan Jul

LaboratoryB was getting better despite the change!
K. Silvester Pathology KS V2 170704

Improvement in process time (Laboratory C)
80 70 60 Time (Days) 50 40 30 20 10 0
Before Change
K. Silvester Pathology KS V2 170704

70

35

After change

Can patients be sure Laboratory C has improved?

Improvement in process time (LaboratoryC)
100 90 80 70 60 50 40 30 20 10 0

Time (days)

Change Implemented
Oct Nov Jun May Aug date Sep Dec Apr Jan Mar Feb Jul

What is going on at Laboratory C ?
K. Silvester Pathology KS V2 170704

Measures for improvement versus measures for judgement
Improvement: • Continuous over time • Show the variation • Show if the change made a significant difference Judgement • One off measures • One number will always be bigger than the other number • No „variation‟ • Did the change make a difference?

K. Silvester Pathology KS V2 170704

Understanding variation
“Every process displays variation”
• Common cause variation
– stable, consistent pattern of variation – constant causes all the time

• Special cause variation
–“assignable special cause” – pattern changes over time

K. Silvester Pathology KS V2 170704

Richard‟s trip to work
120 100 80
Mean

Min. 60
40 20 0

Consecutive trips
K. Silvester Pathology KS V2 170704

Chemotherapy Waiting Times (JPH)
45 40 35

30
Days 25 20 15 10 5 0 Consecutive patients

Significant change for cancer patients at JP Hospital?
How did staff do it?
K. Silvester Pathology KS V2 170704

Chemotherapy Waiting Times (JPH)
Change 1 Baseline Carve Out Change 2 3-5/2002 6/2002 Booked Appts. -from 7/2002 Recalculation Special cause variation

45 40 35 30 Days 25 20 15 10 5 0

Consecutive patients waiting time
K. Silvester Pathology KS V2 170704

mean

upper process limit

lower process limit

George Eliot examples
• Martin Cooke

K. Silvester Pathology KS V2 170704

Statistical process control cf Quality assurance
Statistical Process Control: • Monitors the process • Continuously over time • Shows the „common‟ variation • Shows the „special „ cause • Shows when a change occured Quality Assurance • Monitors the output • Compares results between labs – comparison and judgement • Not continuous over time • Doesn‟t „say‟ anything about the underlying process

K. Silvester Pathology KS V2 170704

Radiology example
• Rate limiting step = taking the film • Functional bottleneck = carve out = radiologist • Taking out the carve out - result

K. Silvester Pathology KS V2 170704


								
To top