7/11/2017 Run Charts 1 7/11/2017 The Importance of Data within - - PDF document

7/11/2017 1 Run Charts

7/11/2017 2

3

The Importance of Data within the BTS

Each team will have theories on why they have problems within their sites, the collection of data and its analysis assists these teams with discovering answers to their hypotheses. If we can’t measure something we have no way of improving it. Measuring your data points over time will help you identify areas within your project that need more focus You can then identify specific areas of need without spending extended time and effort with the whole project

4

What is a Run Chart

Run charts are used to understand the trends and shifts in a process or variation over time, to identify decline or improvement in a process over time

7/11/2017 3

5

Why a Run Chart?

Process change happens over time Determining if a change has happened, and if that change lasts

• vertime is important to process improvement

A run chart will help study observed data for trends or patterns over a specific period of time and focus attention on vital changes in the process The run chart is useful for tracking information and predicting trends

• r patterns

Courtesy: IHI

Static vs Dynamic View of Data

7/11/2017 4

7 Key Resource: Moen et al Quality Improvement through Planned Experimentation 1999

Static vs. Dynamic Interpretation

Cycle time (min)

Change implemented Change implemented

8  Chad Glenn, Innovation Group

Static vs Dynamic View

Static View

Every process displays variation

• Common Cause/ Statistical Variation –

Stable, consistent patterns of variation, “chance”, constant causes

• Special Cause Variation – assignable,

pattern changes over time, non-random

UCL LCL

Dynamic View

7/11/2017 5

9

How many data points?

• Typically you should have between 15 – 20 data points

before constructing a chart

• 15 – 20 patients
• 15 – 20 days
• 15 – 20 weeks
• 15 – 20 months
• 15 - 20 quarters…?

Courtesy: IHI 10

Elements of a Run Chart

Four simple run rules are used to determine if special cause variation is present

Courtesy: IHI

Data Point

7/11/2017 6

11

Rules to Identify non-random patterns (Special Cause Variation) in the data displayed on a Run Chart

 Rule #1: A shift in the process, or too many data points in a run (6 or more consecutive points above or below the median)  Rule #2: A trend (5 or more consecutive points all increasing or decreasing)  Rule #3: Too many or too few runs (use a table to determine this

• ne)

 Rule #4: An “astronomical” data point

Courtesy: IHI 12

Rule 1: Shift

Courtesy: IHI

7/11/2017 7

Rule 2: Trend

Courtesy: IHI

Rule 3: Number of Runs

Courtesy: IHI

7/11/2017 8

Rule 4: Astronomical Data Point

Courtesy: IHI 16  Chad Glenn, Innovation Group

Making sense of the data is a process

Data Interpretation Analysis

Input Transformation Output

Key Resource: Donald Wheeler Understanding Variation/ Managing Chaos 1993 “This shift in data points seems to indicate that the underlying process has undergone some kind of change…”

7/11/2017 9

17  Chad Glenn, Innovation Group

Important Uses of Run Charts for Improvement Activities

• Display data over time to make process performance

visible

• Determine whether a change led to an improvement
• Determine if gains are held after a change has been

implemented

Courtesy: IHI