Microsoft PowerPoint - Presentation - Madrigal SBTI

1-1 Rev. A Printed 5/2/2006© 2005 by Sigma Breakthrough Technologies, Inc.

Managing Patient Throughput by Process Improvement to Improve

Quality and Control Cost

Maria MadrigalMaria MadrigalBlack BeltBlack Belt

Sigma Breakthrough TechnologiesSigma Breakthrough Technologies

Rev. A Printed 5/2/2006© 2005 by Sigma Breakthrough Technologies, Inc.

1-2

Your TakeawaysYour Takeaways

You will take away from this presentation:− In-depth knowledge of the Six Sigma and Lean

methodologies of process improvement

− An example of how Six Sigma and Lean concepts can work together in helping hospitals solve problems and increase revenues

− Understanding of how two hospitals found outstanding results by using these concepts

− Useful ideas on beginning these types of initiatives in your healthcare facility


1-3

IntroductionIntroduction

Before we get started, we want to hear from YOU!− What would make this 45 minutes worthwhile of your time?

− What initiatives in your facility prompted you to attend this conference?

− What tools are you hoping to return with?

− Do you want to be entertained? Educated? Both?


1-4

A MeasurementA Measurement

2 308,5373 66,8074 6,2105 2336 3.4

SigmaLevels

Parts per Million Defective


1-5

Practical MeaningPractical Meaning

99% Good 99.99966% Good

Postal System

20,000 Lost Articles of Mail / Hr

Airline System

Two Short/Long Landings / Day

Medical Profession

200,000 Wrong Drug Prescriptions / Yr

7 Lost Articles / Hr

1 Short / Long every 5 Years

68 Wrong Drug Prescriptions


1-6

The Hidden Process YieldThe Hidden Process Yield

ShrinkageShrinkage(waste)(waste)

>95% >95% Customer QualityCustomer Quality

RedoRedoHidden Process

NOTOK

Yield At Measurement

OperationOperationInputsInputs MeasureMeasure FinalFinal‘Yield’‘Yield’ =

OK

Final Yield ignores the hidden process. Final yield performance is a function of post measurement.

65% 65% is notis not 95% 95% ... why not?... why not?

Process1 2 3

90%90%YieldYield

90%90%YieldYield

85%85%YieldYield =

FPY/RTY 81 % 69 %

95%95%YieldYield

4

65 %

>95%>95%‘Yield’‘Yield’

Final Measurement

Process Variation Causes A "Hidden Process"Increased Cost – Lost Capacity


1-7

Rolled Throughput in HealthcareRolled Throughput in Healthcare

ShrinkageShrinkage(waste)(waste)

>95% >95% Customer QualityCustomer Quality

RedoRedoHidden Process

NOTOK

Yield At Measurement

XX--RayRayPatient presentsPatient presentsfor xfor x--rayray

ResultsResults FinalFinal‘Yield’‘Yield’ =

OK

Final Yield ignores the hidden process. Final yield performance is a function of post measurement.

65% 65% is notis not 95% 95% ... why not?... why not?

Process1 2 3

90%90%XX--RayRay

90%90%

Correct TestCorrect Test85%85%

ResultsResults =FPY/RTY 81 % 69 %

95%95%ReturnedReturned

4

65 %

>95%>95%‘Yield’‘Yield’

Final Measurement

Process Variation Causes A "Hidden Process"Increased Cost – Lost Capacity


1-8

Variation Causes DefectsVariation Causes Defects

LSL USL

Poor Support of Robust Product

Functionality

Process Capability

No standard process for

staff membersInadequate

StaffingNeeds

Variation in Medical Supplies

No Metrics

Way off Target

Out of Spec.


1-9

Time

Perf

orm

ance

Breakthroughs inProcess Performance

GOOD

BAD

3 Sigma (CpK = 1)

6 Sigma (Cpk = 2)


1-10

Changing the Decision Making ProcessesChanging the Decision Making Processes

Decision MakingGrowth Path

Types of ProblemsYou Will Normally Solve

1. Intuition, gut feel, I think …..

2. We have Raw Data and look at it

3. We make graphs / charts of the data

4. We use advanced statistical toolsto evaluate the data

Simple

Complex

How Many Times Have Your Heard This ? “I Think The Problem Is…”


1-11

Overall ApproachOverall Approach

Practical ProblemPractical Problem Statistical ProblemStatistical Problem

Statistical SolutionStatistical SolutionPractical SolutionPractical Solution

y f x x xk= ( , ,... , )1 2


1-12

The Funneling Effect

Optimized Process

30+ Inputs

8 - 10

4 - 8

3 - 6

Found Critical X’s

Controlling Critical X’s

10 - 15

All X’s

1st “Hit List”

Screened List

MEASURE

ANALYZE

IMPROVE

CONTROL


1-13

Triage TeamTriage TeamA.K.A. “Muda Busters!”A.K.A. “Muda Busters!”


1-14

Accelerated Change Event, “ACE”Accelerated Change Event, “ACE”

Time Period

Team Given Clear Objectives

Team Composition

Support

Timeline

• Lead time reduction • Cost reduction• Yield enhancement• Meet customer demand

• 1/3 are process specialists from area• 1/3 are people related to the area • 1/3 are those from outside of area

• 5 Days Ground Rulesg Cannot damage the customerg Cannot permanently “harm”

internal operationsg Use creativity before capital

AM

PM

MON TUE THU FRIWED

TRAINING

CURRENT STATEMAP

CONCEPT & DETAILDESIGN

“TAKE ACTION”

IMPLEMENTIMPROVEMENTS

REFINE

Stabilize / MeasureIMPROVEMENTS

REFINE CHANGESPREPARE PRESENTATION

PRESENTATION& CELEBRATION

“TAKE ACTION” 14

DESIRED STATEGAP ANALYSIS


1-15

What is Lean?What is Lean?

Lean is the identification and elimination of “Waste” with in a process

Waste is define as anything that adds cost but Does not add value to the process.


1-16

TYPES OF WASTETYPES OF WASTE

PEOPLE

TYPESOF

WASTE

Processing

Motion

Waiting

FixingDefects

Making TooMuch

MovingThings

Inventory

QU

ALITY

QUANTITY


1-17

Lean Tools UsedLean Tools Used

Value Stream Mapping

SIPOC

Spaghetti Mapping

Shift Transition Score Card

Flow Analysis

Time / Value Mapping

5S

C & E Matrix

Kanban

Visual Workplace

Consensus Building


1-18

ACE Events Focus on ACE Events Focus on Non ValueNon Value--Added Activity To Reduce Lead Time Added Activity To Reduce Lead Time

Reducing value-added (VA) activities usually results in only minor gainsReducing Non Value-added (NVA, aka Waste) usually results in major lead time reductions without adding resources or capital

Typical company VA NVA Original lead time

Traditional cost reductions VA NVA

Minor improvement

VA NVA Major improvement

Don’ t make capital investments on improving VAs until the NVAs achieve equilibrium with the VAs

ACE waste reduction


1-19

The Value Stream MapThe Value Stream Map


1-20

Reducing Length of StayReducing Length of Stay

Team’s Actions Potential Time Savings

2 minTaking specimen to lab

30 minWait to be tasked

30 minWait to be seen by MD

2 minTriage paper to Reg/MD

5 minPt waiting to be triaged

2 minRN/tech to take vital signs

2 minRN/tech start blue form

5 minCentral Location for Stretchers and Wheelchairs

Total Potential Savings of 1’18”


1-21

Changes In MTCChanges In MTC

Change: − Clear criteria at Meeter/Greeter

and in MTC as to which patients can be transferred directly to MTC

Benefits:− Reduction in ALOS for MTC

− Reduction in load on ER Triage


1-22


Change: − Standardize daytime registration & Fast Track after 11pm

Benefits:− Reduction in patient ALOS

− Cuts registration from 2-3 hours to 15 minutes


1-23


Change: − Visible performance tracking metrics

MTC volume

ALOS for discharged

ALOS admitted patients

5s scores

Benefits:− Sustain the gains

− Pride of ownership


1-24

Changes In MTCChanges In MTCChange: − 5s in MTC (everything has a place, everything in its place)

Move nurse station computers to central areaRelocate crash carts, eye-trays, Pixus, glove dispensers, slit lamp, fridges, linenRemove unnecessary itemsMend exam lightPaintedOmnicells redesigned, removed unused inventory and repositionedIncreased accessibility to & visibility of crash carts

Benefits:− Reduced travel time− Improved patient / visitor safety− Increased workspace efficiency− Decreased inventory− Increased patient throughput− Reduce patient ALOS


1-25

BeforeBefore


1-26

AfterAfter


1-27

MTC Tech Travel To SplintMTC Tech Travel To Splint


1-28

MTC ALOS ImpactMTC ALOS Impact

Before After (conservative)

Process Steps 45 40

VA Time (mins) 48-123 46-121

NVA Time 100-926 95-712

Historical MTC ALOS = 7’30”

End of the week MTC ALOS = 3’04”


1-29

QuotesQuotes

“We have More ‘Blue’ than anything” (blue being waste in the current process”

“Everything that we’re talking about we’ve been trying to change for 2 years”

“I think there are a lot of us that are working against ourselves”

“It usually takes 6 months to move a computer, we’ve moved 3 today”

“It usually takes 2 months to move an Omnicell, we did it in hours”

“These teams could put a man on the moon”

“We’ve achieved more this week than we have in the last year!”


1-30

Lean + Six SigmaLean + Six SigmaLeanLean

Waste EliminationStandard WorkFlowCustomer PULL

Six SigmaSix Sigmaσ Variation Reductionσ Scrap / Rework Eliminationσ Process Optimizationσ Process Control

SPEED STABILITY & ACCURACY

LEAN + SIX SIGMA = A POWERFUL UNIONLEAN + SIX SIGMA = A POWERFUL UNION


1-31

113-bed hospital located in Central Texas

35 miles south of Austin; 45 miles north of

San Antonio. Out migration has always been

a problem

57.7% of inpatient admissions come through the Emergency Department

Factual InformationFactual Information


1-32

VOC, VOB and Problem StatementVOC, VOB and Problem Statement

Q3 02 Q4 02 Q1 03 Q2 03

2.732.952.902.97

0

0.5

1

1.5

2

2.5

3

3.5

4 Only 27% Patients reported being “Very Satisfied”.Where do unsatisfactory delays occur?− 81 % stated before being taken

to a treatment room.− 56 % stated before being

treated by a physician.

Facility had been concerned over customer satisfaction ratings with wait times in the ED. It was thought that people from the community were heading to area hospitals in Austin and San Antonio to receive

care (with shorter waiting times)


1-33

Metrics IdentifiedMetrics Identified

ALOS (Average Length of Stay)− Goal to see a 5% reduction in ALOS

Triage to ER Bed− Goal of reducing turnaround time by 20 minutes

LWBS (Left Without Being Seen)− Goal to reduce LWBS to under 2%

Patient Satisfaction− Measured Gallup poll scores in these areas:

ED Overall SatisfactionWait TimeHelpful and Courteous StaffED EfficiencySpeed of Service


1-34

Data CollectionData Collection

LOS (minutes) Triage to ER bed

220.0

200.0

180.0

160.0

140.0

120.0

100.0

80.0

60.0

40.0

20.0

0.0

Median = 34 minutes

Mode = 25 minutes

Freq

uenc

y

300

200

100

0

Std. Dev = 29.73 Mean = 42.3N = 1023.00

Cycle Time from Triage to Bed Using Six Sigma tools, we

began collecting data on variables in the process. We

found that the average waiting time for a patient to be placed in a bed was over

42 minutes!


1-35

Data CollectionData Collection

Tl Length of Stay (LOS) in minutes

675.0

625.0

575.0

525.0

475.0

425.0

375.0

325.0

275.0

225.0

175.0

125.0

75.0

25.0

Median = 128 minutes

Mode = 55, 90, and 95 minutes

Freq

uenc

y

300

200

100

0

Std. Dev = 88.84 Mean = 149.6N = 1285.00

ALOSWe also found that the

Average Length of Stay for our ED patients was 2.5

hours.


1-36

Using Lean Tools to Drive ResultsUsing Lean Tools to Drive Results

Lean Tools, such as 5S, allowed us to organize the medical

supply closets in the ED. Labels were created, excess inventory returned and par levels adjusted

to provide a savings of over $32,000.00.


1-37

Result 1: Triage to ER BedResult 1: Triage to ER BedOld Process New Process

Average of 19 MinutesAverage of 43 Minutes

Over 50% reduction in overall cycle time!

Observation

Indi

vidu

al V

alue

2532251971691411138557291

150

100

50

0

-50

_X=19.2

UCL=48.7

LCL=-10.4

October Direct April

Triage to ER Bed


1-38

Result 2: Average Length of StayResult 2: Average Length of Stay

6005004003002001000

Median

Mean

145140135130125120

A nderson-Darling Normality Test

V ariance 7618.62Skewness 1.32885Kurtosis 2.54811N 2587

Minimum 1.00

A -Squared

1st Q uartile 80.00Median 123.003rd Q uartile 183.00Maximum 680.00

95% C onfidence Interv al for Mean

139.31

56.75

146.04

95% C onfidence Interv al for Median

120.00 127.00

95% C onfidence Interv al for StDev

84.97 89.73

P-V alue < 0.005

Mean 142.68StDev 87.28

95% Confidence Intervals

Summary for LOS

Old Process New Process

Variation of 29 MinutesVariation of 87 Minutes

Over 60% reduction in overall variation!

18016014012010080

Median

Mean

140135130125120

A nderson-Darling Normality Test

V ariance 883.18Skewness -0.152343Kurtosis -0.856168N 31

Minimum 76.00

A -Squared

1st Q uartile 108.00Median 126.003rd Q uartile 157.00Maximum 178.00

95% C onfidence Interv al for Mean

116.71

0.56

138.51

95% C onfidence Interv al for Median

119.40 137.52

95% C onfidence Interv al for StDev

23.75 39.72

P-V alue 0.140

Mean 127.61StDev 29.72

95 % Confidence Intervals

Summary for June


1-39

Result 3: Left Without Being Seen (LWBS)Result 3: Left Without Being Seen (LWBS)

Reduction from 2% to 0.84%!

60% Reduction in LWBS

Observation

Indi

vidu

al V

alue

554943373125191371

4

3

2

1

0

_X=0.84

UCL=1.950

LCL=-0.270

Before After

2

2

226

6

1

I Chart of Rate by Label 2


1-40

End NotesEnd Notes

Before Project:

− ER was ranked anywhere from 7 to 10 out of the Multi-State region of 17

− No goals to become the “2nd

to None” ED

− No formal data collection process to prove where the problems were coming from

− We knew we had problems, however lacked a clear roadmap

After Project:

− 2nd to None ED-pushing into national top quartile customer satisfaction ratings

− Moving to become the benchmark (leading) ED

− LWBS = $181,800.00 savings

− Reducing Average Length of Stay to allow for increase of 95 patients per month

− Medical supplies returned and par level adjustments =$32,923.70


1-41

End NotesEnd Notes

Before Project:

− ER was ranked anywhere from 7 to 10 out of the Multi-State region of 17

− No goals to become the “2nd

to None” ED

− No formal data collection process to prove where the problems were coming from

− We knew we had problems, however lacked a clear roadmap

After Project:

− 2nd to None ED-pushing into national top quartile customer satisfaction ratings

− Moving to become the benchmark (leading) ED

− LWBS = $181,800.00 savings

− Reducing Average Length of Stay to allow for increase of 95 patients per month

− Medical supplies returned and par level adjustments =$32,923.70


1-42

Questions & CommentsQuestions & Comments

Microsoft PowerPoint - Presentation - Madrigal SBTI

Documents

onfidence interv al

sigma breakthrough technologies

final yield performance

reducing average length

shrinkage process variation

average length

final

post measurement