The Health Care Data Guide: Learning From Data For Improvement

The Health Care Data Guide is designed to help students and professionals build a skill set specific to using data for improvement of health care processes and systems. Even experienced data users will find valuable resources among the tools and cases that enrich The Health Care Data Guide. Practical and step-by-step, this book spotlights statistical process control (SPC) and develops a philosophy, a strategy, and a set of methods for ongoing improvement to yield better outcomes.

Provost and Murray reveal how to put SPC into practice for a wide range of applications including evaluating current process performance, searching for ideas for and determining evidence of improvement, and tracking and documenting sustainability of improvement. A comprehensive overview of graphical methods in SPC includes Shewhart charts, run charts, frequency plots, Pareto analysis, and scatter diagrams. Other topics include stratification and rational sub-grouping of data and methods to help predict performance of processes.

Illustrative examples and case studies encourage users to evaluate their knowledge and skills interactively and provide opportunity to develop additional skills and confidence in displaying and interpreting data.

Companion Web site: www.josseybass.com/go/provost

Lloyd P. Provost is a cofounder of Associates in Process Improvement, the developers of the Model for Improvement roadmap and the Quality as a Business Strategy template for focusing organizations on improvement. Lloyd is a senior fellow at the Institute for Healthcare Improvement, where he supports the use of data for learning in programs.

Sandra K. Murray is a principal in Corporate Transformation Concepts, an independent consulting firm. She is faculty for the Institute for Healthcare Improvement's year-long Improvement Advisor Professional Development Program and their Breakthrough Series College. Sandra has taught numerous programs through the National Association for Healthcare Quality. Her active cohort of client organizations encompasses the spectrum of health care delivery.

Preface.

Who Is This Book For?

Why Are We Writing This Book?

What Is the Book About?

How Is the Book Organized?

Acknowledgments.

Part I Using Data for Improvement.

1 Improvement Methodology.

Fundamental Questions for Improvement: The Model for Improvement.

What Are We Trying to Accomplish?

How Will We Know That a Change Is an Improvement?

What Changes Can We Make That Will Result in Improvement?

The PDSA Cycle for Improvement.

Tools and Methods to Support the Model for Improvement.

Designing PDSA Test Cycles.

Analysis of Data from PDSA Cycles.

Case Study: Improving Care of Diabetic Patients.

2 Obtaining Data for Improvement.

What does the Concept of Data Mean?

How Are Data Used?

Types of Data.

The Importance of Operational Definitions.

Form for Developing Improvement Project Measures.

Data for Different Types of Studies.

Use of Sampling.

What About Sample Size?

Appropriate Scope for a PDSA Cycle.

Stratification of Data.

What About Risk or Case-Mix Adjustment?

Transforming Data.

Analysis and Presentation of Data.

Using a Family of Measures.

3 Understanding Variation Using Run Charts.

What Is a Run Chart?

Use of a Run Chart.

Constructing a Run Chart.

Examples of Run Charts for Improvement Projects.

Probability-Based Tests to Aid in Interpreting Run Charts.

Special Issues in Using Run Charts.

Stratification with Run Charts.

The Multi-Vari Run Chart.

Using the Cumulative Sum Statistic with Run Charts.

4 Learning From Variation in Data.

The Concept of Variation.

Depicting Variation.

Introduction to Shewhart Charts.

Interpretation of a Shewhart Chart.

Establishing and Revising Limits for Shewhart Charts.

When Do We Revise Limits?

Stratification with Shewhart Charts.

Rational Subgrouping.

Shewhart Charts with Targets, Goals, or Other Specifications.

Special Cause: Is It Good or Bad?

Other Tools for Learning from Variation.

Frequency Plots.

Pareto Charts.

Scatter Plot.

5 Understanding Variation Using Shewhart Charts.

Selecting the Type of Shewhart Chart.

Chart Selection Guide.

Shewhart Charts for Continuous Data.

I Charts.

Examples of Shewhart Charts for Individual Measurements.

Rational Ordering with an Individual Chart.

Effect of the Distribution of the Measurements Example of Individual Chart for Deviations from a Target.

X-Bar and S Chart.

Shewhart Charts for Attribute Data.

Subgroup Size For Attribute Charts.

The P Chart for Classification Data.

Examples of P Charts.

Creation of Funnel Limits for P chart.

C and U Charts for Counts of Nonconformities.

C Charts.

U Charts.

Creation of Funnel Limits for a U chart.

Process Capability.

Process Capability from an I Chart.

Capability of a Process from X-bar and S Chart (or R chart).

Capability of a Process from Attribute Control Charts.

Capability from a P Chart.

Capability from a C or U Chart.

Appendix 5.1: Calculating Shewhart Limits.

Part II Advanced Theory and Methods with Data.

6 Shewhart Chart Savvy: Dealing with Some Issues.

Designing Effective Shewhart Charts.

Typical Problems with Software for Calculating Shewhart Charts.

Some Cautions when Using I Charts.

7 More Shewhart-Type Charts.

Other Shewhart-Type Charts.

NP Chart.

X-Bar and Range Chart.

Median Chart.

Shewhart Charts for Rare Events.

G Chart for Opportunities Between Rare Events.

T Chart for Time Between Rare Events.

Some Alternatives to Shewhart-Type Charts.

Moving Average Chart.

Cumulative Sum (CUSUM) Chart.

Exponentially Weighted Moving Average (EWMA) Chart.

Standardized Shewhart Charts.

Multivariate Shewhart-Type Charts.

8 Special Uses for Shewhart Charts.

Shewhart Charts with a Changing Center Line.

Shewhart Charts with Sloping Center Line.

Shewhart Charts with Seasonal Effects.

Transformation of Data with Shewhart Charts.

Shewhart Charts for Autocorrelated Data.

Shewhart Charts for Attribute Data with Large Subgroup Sizes (Over-Dispersion).

Prime Charts (P&' and U').

Comparison Charts.

Confidence Intervals and Confidence Limits.

Shewhart Charts for Case-Mix Adjustment.

9 Drilling Down into Aggregate Data for Improvement.

What are Aggregate Data?

What is the Challenge Presented by Aggregate Data?

Introduction to the Drill Down Pathway.

An Illustration.

Part III Applications of Shewhart Charts in Health Care.

10 Using Shewhart Charts for Individual Patient Care.

Examples of Shewhart Charts for Individual Patients.

Example 1: Temperature Readings for a Hospitalized Patient.

Example 2: Bone Density for a Patient Diagnosed with Osteoporosis.

Example 3: PSA Screening for Prostate Cancer.

Example 4: Shewhart Charts for Continuous Monitoring of Patients.

Example 5: Asthma Patient Use of Shewhart Charts.

Example 6: Monitoring Weight.

Example 7: Monitoring Blood Sugar Control for Patients with Diabetes.

Example 8: Monitoring Patient Measures in the Hospital.

Example 9: Using Shewhart Charts in Pain Management.

11 Learning from Patient Feedback to Improve Care

Patient Surveys.

Summarizing Patient Feedback Data.

Presentation of Patient Satisfaction Data.

Using Patient Feedback for Improvement.

Using Patient Satisfaction Data in Planning for Improvement.

Special Issues with Patient Feedback Data.

12 Using Shewhart Charts in Health Care Leadership.

A Health Care Organization's Vector of Measures.

Developing a Vector of Measures.

Displaying and Learning from a Vector of Measures.

Administrative Issues with a Vector of Measures.

Some Examples of Other Vectors of Measures.

Summary.

Part IV Case Studies.

13 Case Studies Using Shewhart Charts.

Case Study A Improving Access to a Specialty Care Clinic.

Case Study B Radiology Improvement Projects.

Case Study C Reducing Post CABG Infections.

Case Study D Drilling Down Into Percent of C-Sections.

Case Study E Accidental Puncture/Laceration Rate.

Case Study F Reducing Readmission Rates.

Case Study G Variation in Financial Data.

Shewhart Chart Selection Guide.

Notes.

Index.