UNDERSTANDING AND UTILIZING BASIC QI TOOLS Leslie M. Beitsch

Charlotte, NC

December 5, 2012

Definition of Quality Improvement In Public Health

“Quality improvement in public health is the use of a deliberate and defined improvement process, such as Plan-Do-Check-Act, which is focused on activities that are responsive to community needs and improving population health. 

It refers to a continuous and ongoing effort to achieve measurable improvements in the efficiency, effectiveness, performance, accountability, outcomes, and other indicators of quality in services or processes which achieve equity and improve the health of the community.”

Defining Quality Improvement in Public Health; Journal of Public Health Management & Practice: January/February 2010 - Volume 16 - Issue 1 - p 5–7, Riley, William J. PhD; Moran, John W. PhD, MBA, CQIA, CQM, CMC; Corso, Liza C. MPA; Beitsch, Leslie M. MD, JD; Bialek, Ronald MPP; Cofsky, Abbey -

Topic Big ‘QI’ – organization-wide Little ‘qi’ – program/unit

Improvement

Quality Improvement Planning

Evaluation of Quality

Processes

Quality Improvement Goals

Individual ‘qi’

Contrasting Big “QI”, Little “qi”, and Individual “qi”

System focus

Tied to the Strategic Plan

Responsiveness to a community need

Cut across all programsand activities

Strategic Plan

Specific project focus

Program/unit level

Performance of a processover time

Delivery of a service

Individual program/unit level plans

Daily work level focus

Tied to yearly individualperformance

Performance of daily work

Daily work

Individual performance plans

MACRO

MESO

MICRO

INDIVIDUAL

Turning Point

Baldrige

QFD

Lean/SS

Daily Management

P

DC

A

P

DC

AP

DC

A

S

DC

A

Big ‘QI’Little ‘qi’

Individual ‘qi’

QI Teams

Rapid Cycle Advance Tools of

QI

Basic Tools of QI

Continuous Quality Improvement System in Public Health

MAPP

Organization QI = Culture

General Approach On How To Use The Basic Tools Of Quality Improvement

Issue ToConsider

Flow ChartExisting Process

Brainstorm& ConsolidateData

Cause & EffectDiagram – Greatest Concern

Use 5 Whys ToDrill Down ToRoot Causes

Gather DataOn Pain Points

Translate DataInto Information

• Pie Charts• Pareto Charts• Histograms• Scatter Plots, etc.

Flow ChartNew Process

Monitor New Process & Hold

The Gains

• Run Charts• Control Charts

Data ManagementStrategy

“As Is” State to “Should Be” State

“As Is” StateBrainstormingForce and Effect

Analyze Information andDevelop Solutions

Solution andEffect Diagram

Source: The Public Health Quality Improvement Handbook, R. Bialek, G. Duffy, J. Moran, Editors, Quality Press, © 2009, p.160

“AIM”

They Are Not the Same

Quality Assurance• Reactive• Works on problems after they

occur• Regulatory usually by State or

Federal Law• Led by management• Periodic look-back• Responds to a mandate or

crisis or fixed schedule• Meets a standard (Pass/Fail)

Quality Improvement• Proactive • Works on processes • Seeks to improve (culture

shift)• Led by staff• Continuous• Proactively selects a process

to improve• Exceeds expectations

What Is Quality?• Today the most progressive view of quality is

that it is defined entirely by the customer or end user and is based upon that person's evaluation of his or her entire customer experience.

• The customer experience is the aggregate of all the Touch Points that customers have with the organization’s product and services, and is by definition a combination of these.

Continuous Improvement

Plan

DoCheck/Study

Act The continuous improvement phase of a process is how youmake a change in direction.The change usually is because the process output is deterioratingor customer needs have changed

Plan

1. Identify and Prioritize Opportunities

2. Develop AIMStatement

3. Describe the CurrentProcess

4. Collect Data on Current Process

5. Identify All PossibleCauses

6. Identify PotentialImprovements

7. Develop Improvement Theory

8. Develop Action Plan

1. Implement theImprovement

Do

2. Collect and DocumentThe data

3. Document Problems,Observations, and Lessons

Learned

Check/Study

1. Reflect on the Analysis

Act

2. Document Problems,Observation, and Lessons learned

Adopt

Adapt

Abandon

Standardize

Do

Plan

ABCs of the PDCA Cycle, G. Gorenflo and J. Moran, http://www.phf.org/pmqi/resources.htm

Maintenance and Standardization

Standardize

DoCheck/Study

Act The Maintenance and Standardization phase of a process is how we hold the gains. If our process is producingthe desired results we standardizewhat we are doing.

Integrated CycleThe SDCA and PDCA cycles areseparate but rather integrated.Once we have made a successfulchange we standardize and holdthe gain.When the process is not performingcorrectly we go from SDCA to PDCAand once we have the process performing correctly we standardize again.This switching back and forth betweenSDCA and PDCA provides us with the opportunity to keep our processcustomer focused.

P

DC

A

P

DC

A

P

DC

A

Knowledge & Experience

Pro

ject

Diff

icul

ty

PDCA should be repeatedly implemented in spirals of increasing knowledge of the system that converge on the ultimate goal, each cycle closer than the previous.

Hold the Gains

Rapid Cycle*

Rapid Cycle PDCA

Rapid Cycle PDCA, G. Duffy, J. Moran, and W. Riley Published: Texas Quality August 2009 Newsletter

The Basic Tools of QI• Flow Chart• Cause and Effect Diagrams• Pareto Chart• Check Sheet• Histogram• Scatter Diagram• Control Chart

People Centric

Data Centric

Enter Building

Enter Building

Greeter

Patient Flow

Possibly not

Yes

Clerical

Screener

Okay to

vaccinate?

Triage

RN

Okay to

vaccinate?

Need medic

al attenti

on?

NoExit Leave Building

EMT

Is patient able

to leave

on own?

No

Yes

No

Yes

Yes

EMT transports patient to medical facility

Swim Lanes

Spaghetti Diagram: Health Department Administrative Office Flow

Prep printed documents for mailing

Stuff Envelopes

With Pamphlets,

etc

Generic Computer

Work Station

Management

4 Supervisors

Prin

ters

Envelope Storage

different types and sizes of envelopes

Inserts for Letters

Retruned items from mail area –

incorrect labeling

Type up form for Fed Ex

Mailing

Labeling for Fed Ex

1 2 3 4 5

6 7 8

9

10

Mail Sort for location

Take Mail to Post Office

Pick-up

11 12

1st Floor

Document Wasted MovementMiles walked per day

40 Feet

20 F

eet

10 F

eet

40 Feet

Cause and Effect Diagram

PoorHIV Testing

ClientTest Location

Don’t see benefit

Counseling

Not Client Centered

Inconvenient

Staff

Not Respectful

Fearful

Not Offered Poor Experience

Too Public

Don’t Want Test

• 80% of process defects arise from 20% of the process issues.

• 80% of delays in schedule arise from 20% of the possible causes of the delays. • 80% of client complaints arise from 20% of your services.

Location Checklist

Frequency Polygon & Histogram – Grouped Data

10

20

30

40

50

Abs

olut

e F

requ

ency

0.41 0.494 0.497 0.500 0.503 0.506 0.509

60

Obese Children

Age in Years

BM

I –

kg/m

²

2 6 10 14 18

12

16

20

24

28

32

Scatter Plot

Run Chart

Time

Mea

sure

men

t

Median Line

x

x

x

xx

x

x

x

5 10 15 20

0.0

2.5

5.0

7.5

10.0

12.5

15.0

Mov

ing

Ra

nge

Obse rvation

MR Chart

LCL = 0.0

CL = 4.0

UCL = 13.2

0

5

10

15

20

25

Wa

it T

ime

(M

inute

s) X Chart

LCL = 1.6

CL = 12.4

UCL = 23.2

Jones County WIC Lobby Wait TimeXMR Chart

Control Chart

Flow Charting

“If you can't describe what you are doing as a process, you don't know what you're doing.”

W. Edwards Deming

Process/Activities:

Begins With:

Ends With:

Inputs:

Suppliers:

Outputs:

Customers:

Constraints:

Immunization High Level S I P O C+CM Collection Form

Measures

Patient Enters Clinic

Too much confusing signageChange mind after counselingCostSit and Wait/ Language barrier

Patient Leaves Clinic

• Fill out Form - Pay• Stand in line - Leave• Register• Sit & Wait• Counseling• Sit & Wait• Immunization

Wait time# PatientsCust Sat. Scores

• Referrals• Schools• WIC• Home Visits

• Pharma Cos• Schools• MD Offices• Dept. Nurses

• Vaccine Record• Immunized Pt.• Protected Community

• Community• Patient• Family• Schools• Dept. nurses

Flow Charting• Flow charting is the first step we take in understanding a

process

• Organized combination of shapes, lines, and text

• Flow charts provide a visual illustration, a picture of the steps the process undergoes to complete it's assigned task 

• From this graphic picture we can see a process and the elements comprising it

• Shows how interactions occur

• Makes the invisible visible

Flow Chart Benefits• Creates a common vision

• Establishes the “AS IS” baseline – Current State– It will take multiple passes to get to the “As Is” State

• Baseline to measure improvements

• Identifies wasteful steps – activities/waits

• Uncovers variations - realize everyone is not doing it the same way – there will be disagreements

• Shows where improvements could be made and potential impacts

• Training tool – for new employees

Flow Chart People BenefitsPeople involved in constructing a flow chart begin

to:

– Better understand the process

– Understand the process in the same terms

– Realize how the process and all the people involved, including them, fit into the overall process or business

– Identify areas for improving the process

Olmsted County , MN – Performance Appraisal Process

Flow Charting Construction

• Clearly define the process boundaries to be studied

• Define the first and last steps – start and end points

• Get the right people in the room

• Decide on the level of detail– Complete the big picture first – macro view– Fill in the details – micro view

• Gather information of how the process flows: • Experience• Observation• Conversation• Interviews• Research

• Clearly define each step in the process– Be accurate and honest

Flow Charting Steps• Use the simplest symbols possible – Post-Its

• Make sure every loop has an escape

• There is usually only one output arrow out of a process box. Otherwise, it may require a decision diamond.

• Trial process flow – walk through people involved in the process to get their comments

• Make changes if necessary

• Identify time lags and non-value-adding steps.

Hands Are Dirty

Hands Are Clean

Macro to Micro

Flow Charting Moments• Aha!

• Surprise

• Bafflement

• Duh!

• Embarrassment

Flow Chart Symbols

Activity:Operation/Inspection

Decision

Start/EndBookends

Document

Wait/Delay

Storage

Data Base

Transport

Input

Output

Flow Lines

A Connector

Forms

CommentCollector

Input/OutputData

ManualOperation

Preparation

ManualInput

Display

Unfamiliar/Research

Constructing a Flow Chart• Asking questions is the key to flow charting a

process. 

• Such as for this process:– Who is the customer(s)?– Who is the supplier(s) ?– What is the first thing that happens? – What is the next thing that happens?– Where does the input(s) to the process come from? – How does the input(s) get to the process? – Where does the output(s) of this operation go? – Is their anything else that must be done at this

point? – What is the baseline measurement of this process?

Analyzing A Flow Chart

Examine each:– Activity symbol – value/cost? – Decision point – necessary/redundant?– Choke Points – bottlenecks?– Rework loop – time/cost?– Handoff – is it seamless?– Document or data point – useful?– Wait or delay symbol – why?/reduce/eliminate– Transport Symbol – time/cost/location?– Data Input Symbol – right format/timely?– Document/Form Symbol – needed/cost/value?

Flow Charting Basics

Flow

DataVoice of the Process - VOP

Customer - VOC• Internal• External

Flow Charting Exercise

Flow Charting Exercise

Alarm Goes OffStart You are sound asleep and your alarm just went off at 6 AM this morning

Open your door andleave End

You open your door and leave forthis workshop

Your assignment: Flow chart all the activities you have to do to get to the point where you open the door and leave for this workshop.

* Identify two steps that are not needed and should be eliminated.

* Identify two steps that should be added that would add value to the process.

Only once on any flow chart

Alarm Goes off

Put on CoatShower

Dress

Eat & ReadPaper

Multi Tasking

Get Up

Open Door & Leave

Cause and Effect Diagrams

Moving from Treating Symptoms

To

Treating Causes

Problem Solving – What we usually see is the tip of iceberg – “The Symptom”

The Symptom

The Root Causes

Invisible

Hidden

Problem Solving• When confronted with a problem most people like to

tackle the obvious symptom and fix it

• This often results in more problems  • Using a systematic approach to analyze the problem and

find the root cause is more efficient and effective

• Symptom – sign or indication

• Cause – whatever makes something happen

Cause and Effect Diagrams - Construction

• Write the issue as a problem statement on the right hand side of the page and draw a box around it with an arrow running to it. 

• This issue is now the effect

Effect

Cause and Effect Diagrams - Construction

• Generate ideas as to what are the main causes of the effect

• Label these as the main branch headers

• Organizes group knowledge about causes of a problem and displays the information graphically

Effect

HeaderHeader

HeaderHeader

Cause and Effect Diagrams - Construction

• Typical Main Header are:– 4 M’s – Manpower, Materials, Methods, Machinery

– People– Policies– Materials– Equipment– Life style– Environment– Etc.

Cause and Effect Diagrams - Construction

• For each main cause category brainstorm ideas as to what are the related sub-causes that might effect our issue

  • Use the 5 Why techniques when

a cause is identified

• Keep repeating the question until no other causes can be identified

• List the sub-cause using arrows

Effect

HeaderHeader

HeaderHeader

why

why

why

why

Selecting Items to Investigate• When the Cause and Effect Diagram is finished it is time

to decide what few areas should be focused on to develop solutions to solve the effect.

• Some are obvious – low hanging fruit » Caution - It could be rotten

• Some require some research using the other QI tools such as:– Pareto Diagrams– Run Charts– Surveys– Histograms– Etc.

Obese Children

Life Style

PoliciesEnvironment

TV ViewingNo Time For Food Prep

No Outdoor PlayUnsafe

Juices

Bottle Pacifier

Less Fruits and Veg.

Less Income Maternal

Choices

Less Vigorous Exercise

Curriculum

No Sidewalks

Unhealthy Food Choices

Few Community Recreational Areas or Programs

Built Environment For Strollers Not Toddling

Less Indoor Mobility

TV Pacifier

UnsafeHousing

Sodas/Snacks

Decreased Breast Feeding

Early Feeding Practices

Genetics

Syndromes

Genes

Pre NatalPractices

Excess Maternal Weight Gain

Over Weight Newborn

Over WeightPre School At School

At Home

Problem (Effect)

5 Why’s Technique

Why?

Why?

Why?

Why?

Why?

http://www.phf.org/resourcestools/Pages/Solve_The_Real_Problem_Using_RCA.aspx

Con

trol

Influ

ence

How To Prioritize

• Sort

• Vote

• Vote with data

• By Branch

• Branch with the most causes

• Must Confirm With Data

Cause and Effect Exercise

Why Are Employees Late For Work?

• It is a question that management has wrestled with for a long time.

• Every time we think we have heard all the excuses someone invents another creative one.

Investigating The Reasons

• Why Are Employee Late For Work?

• Use a Cause and Effect diagram to understand the causes.

• Use the 5 Why Technique to get to Root Cause.

Why EmployeesAre Late For Work?

Cause and Effect Diagram

Why EmployeesAre Late For Work?

Work IssuesTransportationFamily

ProcedureCulture

Lack of SupervisionSick Child

Late for Bus

Car problems

Poor Morale

No valuefor promptness

Wardrobe problems

Alarm not set

Unclear policy at work

Daycare closed

Pet Care

WeatherUnclear expectations

Dislikesco-workers

Uncooperative child

Health problems

Lots of parties

Cause and Effect Diagram

Flat Tire

UncontrollableFactors

Car Crash

Weather

Labor Strike

No Public Transportation

Traffic

Power outage

Boss is always late

Accepted Norm

64

Static and Dynamic Data AnalysisFour Generations

• Static Analysis– First Generation: Tables of Data,

Comparison of Summary Measures• Dynamic Analysis

– Second Generation: Trend Line– Third Generation: Run Chart– Fourth Generation: Control Chart

65

Run Chart

• A running record of process behavior over time.

• Easily understood by all,• Can be used on any type of process

and any type of data.• Requires no statistical calculations, can

detect some special causes.

66Run Charts

Determining a Run

• A run is one or more consecutive data points on the same side of the median.

• Do not count a point if it is on the centerline. (Put a box around it and ignore)

• “Useful Observations”– Subtract any observation that falls on median.

• Draw circle around each run, and count the number of runs.

67 Run Chart2 Tests to Identify Special Cause

• Test 1 – Long run: – If 7 or more in run( When

less than 20 “useful observations”) When 20 or more “useful observations” , then 8 or more data points needed for a run.

• Test 2 – Trend:– An unusually long series of

consecutive increase or decrease.

Total Data Points on Chart

Number Ascending or Descending

5-8 5 or more

9-20 6 or more

21 to 100 7 or more

Determining a Trend

68

Two Types of Variation

• Common Cause – Inherent in every process– Reflects a stable process because

variation is predictable– Is random variation

• Special Cause – A noticeable shift or trend in data over time– Process is unstable or unpredictable– Process is out of statistical control– Not present in every process

Jun-0

8Aug

Oct

DecFeb

AprJu

nAug

Oct

DecFeb

Apr0

200

400

600

800

1000

1200

Port Gamble S'Klallam TribeRun Chart: Number of Clinic Visits Per Month

June 2008 - May 2010

Number of Clinic Visits

Xbar (Median)

Month

Nu

mb

er o

f C

lin

ic V

isit

s

Median = 637 visits per month

Start ofIntervention

1st Staff Training 2nd Staff

Training

69

69

Jun-0

8Aug

Oct

DecFeb

AprJu

nAug

Oct

DecFeb

Apr0

200

400

600

800

1000

1200

Port Gamble S'Klallam TribeRun Chart: Number of Clinic Visits Per Month

June 2008 - May 2010Runs Above or Below Median

Number of Clinic Visits

Xbar (Median)

Month

Nu

mb

er o

f C

lin

ic V

isit

s

Median = 637 visits per month

Median = 637 visits per month

70

Interpretation

• How do you interpret the run chart?• What do you recommend?

71