Table of Contents
Cover
Title Page
Copyright
List of Tables, Figures, and Exhibits
Preface
How to Use This Book
A Note on Methods
Acknowledgments
From Peter Muennig
From Mark Bounthavong
About the Authors
Peter Muennig
Mark Bounthavong
Chapter 1: Introduction to Cost-Effectiveness
Learning Objectives
Overview
Why Cost-Effectiveness Is Useful
Elements of Cost-Effectiveness Analysis
The Average and Incremental Cost-Effectiveness Ratio
Why Conduct Cost-Effectiveness Analysis?
The Reference Case Analysis
Cost-Effectiveness Analysis and Policy
Summary
Further Readings
References
Chapter 2: Principles of Cost-Effectiveness Analysis
Learning Objectives
Overview
The Perspective of a Cost-Effectiveness Analysis
Capturing Costs
Capturing Quality
Interpreting the Cost-Effectiveness Ratio
Types of Economic Analysis
Summary
Further Readings
References
Chapter 3: Developing a Research Project
Learning Objectives
Overview
Eight Steps to a Perfect Research Project
Developing a Research Question
Designing Your Analysis
Summary
Further Readings
References
Chapter 4: Working with Costs
Learning Objectives
Overview
Opportunity Costs
Identifying Costs
Micro-Costing and Gross Costing
Getting Cost Data
Using Diagnosis Codes
Adjusting Costs
Costs Associated with Pain and Suffering
Assessing the “Relevancy” of Cost Data
Other Cost Considerations
Summary
Further Readings
References
Chapter 5: Probabilities and Decision Analysis Models
Learning Objectives
Overview
The Idea Behind Decision Analysis
Probabilities
Decision Analysis Models
Summary
Further Reading
References
Chapter 6: Calculating Life Expectancy
Learning Objectives
Overview
Hand-Calculating Years Gained
Calculating Life-Years Lost Using Markov Models
Summary
Further Readings
References
Chapter 7: Working with Health-Related Quality-of-Life Measures
Learning Objectives
Overview
Framework
Deriving HRQL Scores
Other Considerations and Reminders
Using Disability-Adjusted Life-Years
Summary
Further Readings
References
Chapter 8: Calculating QALYs
Learning Objectives
Overview
Using the Summation Method
Using the Life Table Method
More On Discounting
Issues with QALYs
Calculating Incremental Cost-Effectiveness
Disability-Adjusted Life-Year (DALY)
Summary
Further Readings
References
Chapter 9: Conducting a Sensitivity Analysis
Learning Objectives
Overview
Performing a Sensitivity Analysis
Sensitivity Analysis
One-Way Sensitivity Analysis
Two-Way Sensitivity Analysis
Tornado Diagram
Multiway Sensitivity Analysis
Summary
Further Readings
References
Chapter 10: Preparing Your Study for Publication
Learning Objectives
Overview
Content and Structure of Cost-Effectiveness Articles
Title and Abstract
Introduction
Methods
Results
Discussion
Other
Technical Appendix
Figures and Illustrations
Summary
Further Reading
References
Chapter 11: Basic Concepts in Epidemiology and Application to Cost-Effectiveness Analysis
Learning Objectives
Overview
Review of Incidence
Understanding Error
Managing Error in Cost-Effectiveness Analysis
Calculating Weighted Means
Evaluating Study Limitations
Summary
Further Readings
References
Chapter 12: Finding the Data You Need
Learning Objectives
Overview
Finding Data in the Medical Literature
Grading Published Data
Using Electronic Datasets
Data Extraction Tools
Using Unpublished Data
Using Expert Opinion
Organizing Your Data
Summary
Further Readings
References
Chapter 13: A Worked Example
Learning Objectives
Overview
Helping Students Learn
Laboratory Exercise One: Building a Markov Tree
Laboratory Exercise Two: Comparing Interventions
Laboratory Exercise Three: The Cost-Effectiveness of Health Insurance
Laboratory Exercise Four: Sensitivity Analysis
Summary
Appendix A: Answer Key to Exercises
Chapter 1
Chapter 2
Chapter 3
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Chapter 8
Chapter 9
Chapter 11
Chapter 12
Appendix B: Life Expectancy and Quality-Adjusted Life Expectancy Tables
Appendix C: EQ-5D-5L Health Questionnaire (English version for the United States)
Appendix D: Diagnosis, Charges, Medicare Reimbursement, Average Length of Stay, and Cost-to-Charge Ratio by Diagnosis-Related Groups, 2011
Glossary
Index
End User License Agreement
Pages
ix
x
xi
xii
xiii
xiv
xv
xvi
xvii
xviii
xix
xx
xxiii
xxiv
xxv
xxvi
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
379
383
384
385
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
Guide
Cover
Table of Contents
Preface
Begin Reading
List of Illustrations
Chapter 1: Introduction to Cost-Effectiveness
Figure 1.1 Example of the Effect of a Health Intervention on the Health States of Patients Admitted to the Emergency Room for an Acute Asthma Attack
Figure 1.2 Components of a Cost-Effectiveness Analysis
Chapter 2: Principles of Cost-Effectiveness Analysis
Figure 2.1 Graphical Representation of an HRQL Score
Figure 2.2 Difference in Total QALYs Between Women Treated and Not Treated for Diabetes over 10 Years
Figure 2.3 Graphical Representation of the Impact of High and Low Numerators and Denominators in Calculating Incremental Cost-Effectiveness Ratios
Chapter 3: Developing a Research Project
Figure 3.1 Flowchart Indicating the Clinical Course of Influenza Illness
Figure 3.2 Flowchart Indicating the Course of Influenza Infection Among Subjects Who Receive a Vaccination
Figure 3.3 Probability of Seeing a Doctor Among Subjects Who Receive Vaccination Versus Those Who Receive Supportive Care
Figure 3.4 Vaccination Strategy Represented with All Probabilities Filled In
Figure 3.5 Vaccination Decision Node
Chapter 4: Working with Costs
Figure 4.1 Partial Flowchart of the Course of the Flu
Figure 4.2 Costs Associated with the “Does Not See Doctor” Box
Chapter 5: Probabilities and Decision Analysis Models
Figure 5.1 Decision Tree for Whether to Pursue Public Health School or Write a Novel
Figure 5.2 Decision Tree with the Potential for Not Finding a Job After Public Health School
Figure 5.3 Example of Two Mutually Exclusive Events
Figure 5.4 Probabilities of Outcomes for Patients Receiving the Vaccine Intervention
Figure 5.5 Course of Events During an Influenza Season Among Those Receiving Supportive Care Alone
Figure 5.6 Course of Events During an Influenza Season Among Those Receiving a Vaccination
Figure 5.7 Supportive Care Versus Vaccination Decision (Figure 5.5 and 5.6) Represented as a Decision Analysis Tree
Figure 5.8 Event Pathway for Vaccination Versus Supportive Care Represented as a Decision Analysis Tree
Figure 5.9 The Expected Probability for Each Terminal Node in the Vaccination and Supportive Care Decision Tree
Figure 5.10 The Total Cost for Each Terminal Node in the Vaccination and Supportive Care Decision Tree
Figure 5.11 Expected Costs for Each Terminal Node in the Vaccination and Supportive Care Decision Tree
Figure 5.12 Total Cost and Probability for All Terminal Nodes in the Vaccination and Supportive Care Decision Tree
Figure 5.13 Expected Cost and Outcomes for Each Chance Node in the Vaccinated and Supportive Care Decision Tree
Figure 5.14 Calculation for the Expected Cost and Probability for a Patient Who Receives Supportive Care, Becomes Ill, and Sees Doctor
Figure 5.15 Expected Costs and Outcomes for Different Chance Nodes in the Vaccinated and Supportive Care Decision Tree
Chapter 6: Calculating Life Expectancy
Figure 6.1 Markov Model for Influenza Mortality in 15-Year-Olds
Figure 6.2 Basic Concept of a Markov Model
Figure 6.3 Complete Decision Analysis Tree for Calculating Life Expectancy Using TreeAge Pro
Figure 6.4 A Rolled-Back Model Using the Probability of Death for the General U.S. Population
Chapter 7: Working with Health-Related Quality-of-Life Measures
Figure 7.1 Trade-off Between the Status Quo Health State and a Gamble
Figure 7.2 EQ-5D-5L Form Filled Out by a Patient
Figure 7.3 Diabetes Markov Model Depicting Three Health States: Mild, Moderate, and Severe Diabetes
Chapter 8: Calculating QALYs
Figure 8.1 Year-to-Year Progress of Treated and Untreated Subjects with Leishmaniasis
Figure 8.2 Basic Markov Model Used to Calculate Life Expectancy
Figure 8.3 Markov Models Designed to Calculate the Life Expectancy of Subjects Receiving the Filmore and Reinkenshein Procedures
Figure 8.4 Difference in HRQL Among Subjects Who Received the Filmore or the Reinkenshein Procedure
Figure 8.5 Filmore Versus Reinkenshein Model Rolled Back to Reveal Gains in Quality-Adjusted Life Expectancy Associated with Each Strategy
Figure 8.6 Filmore Versus Reinkenshein Model with Costs Added
Figure 8.7 Filmore Versus Reinkenshein Model with Discounting Added to the HRQL Values
Figure 8.8 Filmore Versus Reinkenshein Model Rolled Back
Chapter 9: Conducting a Sensitivity Analysis
Figure 9.1 Sensitivity Analysis Focusing on Structure for the Vaccine Event Pathway
Figure 9.2 Sensitivity Analysis Focusing on the Parameter Change (Remains Well) for the Vaccine Event Pathway
Figure 9.3 Terminal Branch of the Filmore Arm Represented in Figure 8.3
Figure 9.4 Incremental Effectiveness of the Reinkenshein Procedure Relative to the Filmore Procedure over a Range of Risk Ratios
Figure 9.5 One-Way Sensitivity Analysis Examining How the Cost of Providing the Influenza Vaccine Influences Intervention
Figure 9.6 Two-Way Sensitivity Analysis Comparing Changes in the Efficacy of the Influenza Vaccine and the Incidence of Influenza-Like Illness
Figure 9.7 Tornado Diagram Example
Figure 9.8 Microsimulation of Individual Patients Through a Decision Path
Figure 9.9 Chance of Incurring Any Given Value of a Normally Distributed Variable
Figure 9.10 Diabetes Model in Which Values of Each Variable Are Normally Distributed
Figure 9.11 The Triangular Distribution
Figure 9.12 Other Distributions Used in Monte Carlo Simulations
Figure 9.13 Cost-Effectiveness Plane with a Single Simulation
Figure 9.14 Hypothetical Results of 100 Simulations on the Cost-Effectiveness Plane
Figure 9.15 Cost-Effectiveness Acceptability Curve
Figure 9.16 Cost-Effectiveness Acceptability Curve for Exercises 2 and 3
Chapter 11: Basic Concepts in Epidemiology and Application to Cost-Effectiveness Analysis
Figure 11.1 Nonrandom Error
Figure 11.2 Random Error
Figure 11.3 Graphical Representation of the 100 Cholesterol Values
Figure 11.4 Probability Distribution of the 100 Cholesterol Values
Figure 11.5 Hypothetical Probability Distribution of 1,000 Cholesterol Values
Figure 11.6 The Normal Curve
Figure 11.7 Example of a Triangular Distribution
Figure 11.8 Retrospective Study Designs: Case-Control and Cohort Designs
Figure 11.9 Prospective Study Design
Figure 11.10 Randomized Controlled Trial Study Design
Figure 11.11 Network of Studies Comparing Drugs Directly and Indirectly
Chapter 12: Finding the Data You Need
Figure 12.1 Pyramid Analogy for the Different Levels of Evidence Criteria
Figure 12.2 An Example of a Jadad Score Grading Form
Chapter 13: A Worked Example
Figure 13.1 Starting a New Project in TreeAge Pro
Figure 13.2 Configuring the Model
Figure 13.3 Changing the Calculation Method to Cost-Effectiveness
Figure 13.4 Selecting the Payoffs for a Cost-Effectiveness Analysis
Figure 13.5 Selecting the Number of Payoffs
Figure 13.6 Changing Payoffs Names Under the “Custom Names” Option
Figure 13.7 Numerical Formatting for the Cost-Effectiveness Analysis
Figure 13.8 Saving a New Document in TreeAge Pro as a Package Using the *.trex Extension
Figure 13.9 Adding Two Chance Nodes to the Existing Decision Node
Figure 13.10 Deleting a Branch Using the Table Icons
Figure 13.11 Selecting a Markov Node Using “Change Type” from the Menu
Figure 13.12 Markov Node for a Decision Analysis Model
Figure 13.13 Creating Branches on a Markov Tree
Figure 13.14 Labeling Branches in a Markov Tree
Figure 13.15 Assigning an Initial Probability of 100 Percent to the Alive State in the Markov Model
Figure 13.16 Changing the Node from Chance Node to a Terminal Node
Figure 13.17 Edit Jump State Dialog Box Appears When the Chance Node Is Changed to a Terminal Node
Figure 13.18 Markov Tree After Changing the Remaining Chance Nodes into Terminal Nodes
Figure 13.19 Selecting the Markov Info View in Order to Enter Values
Figure 13.20 Selecting the Markov Info View Using the Icon on the Button Toolbar
Figure 13.21 Markov Info View Dialog Box
Figure 13.22 Setting the Initial and Incremental Rewards
Figure 13.23 Preference Box for the Markov Model
Figure 13.24 Selecting the “Variables/Markov Info” Option from the Tree Preferences Dialog Box
Figure 13.25 Termination Conditions for the Markov Node
Figure 13.26 Setting the Termination Conditions for the Markov Model
Figure 13.27 Creating a New Variable Using the New Variable Dialog Box
Figure 13.28 Assigning a Value to the Variable “age”
Figure 13.29 Defining the Variable “age”
Figure 13.30 Setting the Probability of Die as tdead2000[age], Which Is Derived from an n × 2 Matrix Table
Figure 13.31 Layout for Entering Values for a User-Defined Table
Figure 13.32 An Excel Table with the Probabilities of Death Associated with Each Age
Figure 13.33 Entering Values for tdead2000 Table
Figure 13.34 Selecting the Numeric Formatting Preferences from the Edit Menu
Figure 13.35 Changing the Payoff Units for Cost and Effectiveness
Figure 13.36 Overall Illustration of the Markov Model After Parameter Inputs and Unit Changes
Figure 13.37 Results of the Rollback Analysis for the Markov Model
Figure 13.38 Transition State Diagram of a Markov Model with Alive and Dead States
Figure 13.39 Results of Rollback Function
Figure 13.40 U.S. Life Table for Ages 0 to 100 and Over
Figure 13.41 Changing the Initial Cycle's Worth of Reward with the Half-Cycle Correction
Figure 13.42 Rollback Results Using the Half-Cycle Correction on the Initial Effectiveness
Figure 13.43 Opening the Define Initial Reward Window
Figure 13.44 Entering the Half-Cycle Correction into the Markov Information Dialog Box
Figure 13.45 Defining the Termination Condition to an Age That Is 105 Years or Older
Figure 13.46 Rollback Analysis with Half-Cycle Correction and Life Cycle Greater Than 105 Years
Figure 13.47 Selecting and Copying the Subtree
Figure 13.48 Adding a Second Subtree to the Decision Node
Figure 13.49 Multiplying Mortality Probability by a Factor of 1.25 and Capping It at 1
Figure 13.50 Changing Calculation Method from Cost-Effectiveness to Simple Analysis
Figure 13.51 Results of Rollback Between Filmore and Rinkenshein Procedures
Figure 13.52 Modifying the Payoff for a Cost-Effectiveness Analysis
Figure 13.53 Entering a Willingness to Pay of $40,000 per Life-Year Gained
Figure 13.54 Defining the Reward Set in the Initial Stage of the Markov Model
Figure 13.55 Rollback Results Comparing the Filmore and Rinkenshein Procedures
Figure 13.56 Selecting the Rankings from the Analysis Tab
Figure 13.57 Rankings Output Comparing the Cost-Effectiveness of Rinkenshein and Filmore Procedures
Figure 13.58 Insurance Versus No Insurance Competing Alternatives Model
Figure 13.59 Adding Tables Under the Tables View Option
Figure 13.60 Adding a New Table Under the Add/Change Box
Figure 13.61 The New Table, cInsurance, Is Not Listed in the Tables Window
Figure 13.62 Including Values for a User-Defined Table
Figure 13.63 Completed Tables for Costs and HRQL for Those Who Are Insured and Not Insured
Figure 13.64 Changing the Effectiveness Units to “QALYs”
Figure 13.65 Creating a New Variable Called “HR” and Defining Its Value
Figure 13.66 Defining the Value for “age” Under the Variables Properties Window
Figure 13.67 Formula for the Four New Variables
Figure 13.68 Adding the Half-Cycle Correction Factor for the Initial and Final Stages
Figure 13.69 Markov Model Comparing Insured and Uninsured Strategies with the Updated Variables and Tables
Figure 13.70 Rankings Output Comparing Insurance to No Insurance
Figure 13.71 Tree Properties with the Addition of a Discount Rate
Figure 13.72 All the Variables That Are Used in the Current Markov Model
Figure 13.73 Changing the Start Age at the Decision Node
Figure 13.74 Changing the Termination Conditions for the Insurance and No Insurance Arms of the Markov Model
Figure 13.75 One-Way Sensitivity Analysis Setup Window
Figure 13.76 Defining Low and High Values for the HR Parameter
Figure 13.77 Cost-Effectiveness Sensitivity Analysis Output Window
Figure 13.78 Results of the One-Way Sensitivity Analysis
Figure 13.79 Creating a New Distribution Variable
Figure 13.80 Add/Change Distribution Window
Figure 13.81 Assigning a Distribution to the Hazard Ratio Variable
Figure 13.82 Creating Another Distribution Called “dist_cInsur_err” Using a Triangular Distribution
Figure 13.83 Distributions for Three Variables (HR, cInsur, and cNoInsur) in the Model
Figure 13.84 Tree Properties with the Inclusion of Distributions
Figure 13.85 Selecting the Probabilistic Sensitivity Analysis
Figure 13.86 Monte Carlo Simulation Options Window
Figure 13.87 Selecting Incremental CE Ratio Output from the Monte Carlo Simulation Results Window
Figure 13.88 Distributions of ICERs Comparing Insurance Versus No Insurance Varying the Hazard Ratio and Cost Error Terms
Figure 13.89 Selecting the ICER Scatterplot Comparing Insurance to No Insurance
Figure 13.90 ICER Scatterplot Comparing Insurance to No Insurance
Figure 13.91 Proportion of ICER Scatterplots Below the Willingness-to-Pay Threshold of $40,000 per QALY Gained
Figure 13.92 Cost-Effectiveness Acceptability Curve Parameter Window
Figure 13.93 Cost-Effectiveness Acceptability Curves for Insurance and No Insurance
Appendix A: Answer Key to Exercises
Figure A.1 The Rollback Results for the Expected Costs and Outcomes for Vaccinated and Not Vaccinated Strategies
Figure A.2 Markov Model from Chapter 6, Exercise 1
Figure A.3 A Tree Diagram with Rollback Results from TreeAge Pro
Figure A.4 Cost-Effectiveness Acceptability Curve Showing Where the 75 Percent Probability of Cost-Effectiveness Is in Relation to the Willingness-to-Pay Axis
Figure A.5 Using the CDC's Wonder Mortality Database and Selecting Breast Cancer
Figure A.6 Selecting the ICD-10 Code for Breast Cancer
Figure A.7 Results of the Breast Cancer Query Grouped by Gender for 2011
List of Tables
Chapter 1: Introduction to Cost-Effectiveness
Table 1.1 Hypothetical League Table for a Village in Malawi with a $58,000 Health Budget
Chapter 2: Principles of Cost-Effectiveness Analysis
Table 2.1 Costs Included in a Cost-Effectiveness Analysis of Free Contraception, Conducted from Three Perspectives
Table 2.2 Hypothetical Differences in Health-Related Quality of Life over 10 Years for Diabetic Women and Women in Perfect Health
Table 2.3 Decision Matrix for Various Cost-Effectiveness Scenarios
Chapter 4: Working with Costs
Table 4.1 Comparison of Pharmaceutical Benchmark Prices
Table 4.2 Partial List of Costs for Treatment of Influenza Infection
Table 4.3 Common Codes Used to Group Diseases
Table 4.4 MEDPAR Cost Data by DRG for 2011
Table 4.5 Medical Component of the Consumer Price Index 2004–2014, Annual Percentage Change over Previous Year
Table 4.6 Hypothetical and Discounted Costs of a Cohort of 1,000 Elderly Persons over 10 Years
Chapter 5: Probabilities and Decision Analysis Models
Table 5.1 Probabilities and Costs for Vaccinated and Not Vaccinated (Supportive Care) Strategies
Table 5.2 Calculation of Expected Cost for Each Event Pathway for Vaccination and Supportive Care Strategies
Chapter 6: Calculating Life Expectancy
Table 6.1 Number of Deaths due to Influenza Virus Infection, by Age Group
Table 6.2 Deaths, Mean Age of Death due to Influenza Virus Infection, and Life Expectancy for Persons Aged 15 to 65
Table 6.3 Calculating Total Years of Life Lost due to Influenza Virus Infection in the United States
Table 6.4 Total Deaths, Deaths due to Influenza Virus Infection, and Total Survivors in a Cohort of 1 Million 15-Year-Olds
Table 6.5 Total Person-Years Lived by the Cohort of 1 Million 15-Year-Olds
Table 6.6 Person-Years Lived Among the Cohort of 15-Year-Olds, Including and Excluding Deaths due to Influenza Virus Infection
Table 6.7 Age-Specific Mortality Rates, Survivors, and Number of Deaths in the Cohort of 1 Million 15-Year-Old Subjects
Table 6.8 Progression of a Cohort of 10 Women with Breast Cancer over a Six-Year Period
Chapter 7: Working with Health-Related Quality-of-Life Measures
Table 7.1 Example of How an HRQL Score for Influenza Illness May Be Derived Using the EQ-5D
Table 7.2 EQ-5D Preference Score Variation Among Age Categories
Chapter 8: Calculating QALYs
Table 8.1 Total Person-Years Lived by the Cohort of 1 Million 15-Year-Olds
Table 8.2 Sum of Person-Years Across Age Groups for the Cohort of 1 Million 15-Year-Olds
Table 8.3 Calculating Life Expectancy at a Given Age
Table 8.4 Abridged Life Table for 2011
Table 8.5 A Quality-Adjusted Life Table
Chapter 10: Preparing Your Study for Publication
Table 10.1 Simple Summary of Costs Used in a Cost-Effectiveness Model
Table 10.2 Cost-Effectiveness Table
Table 10.3 Example of a Cost-Effectiveness Table
Chapter 11: Basic Concepts in Epidemiology and Application to Cost-Effectiveness Analysis
Table 11.1 Calculating the Incidence Rate of Developing Cancer due to Exposure to Radiation in a Two-Year Observation Period, 2011–2012
Table 11.2 Calculating the Age-Adjusted Mortality Rate Using a Hypothetical U.S. Population
Table 11.3 Frequency Distribution of Hypothetical Cholesterol Values Obtained from 100 Subjects
Chapter 12: Finding the Data You Need
Table 12.1 Datasets Useful for Finding Frequently Needed Cost-Effectiveness Parameters
Chapter 13: A Worked Example
Table 13.1 Results of the Base-Case Analysis
Table 13.2 Age-Indexed Table for Use in the New Tree
Table 13.3 Base-Case Results for the Markov Model
Table 13.4 Base-Case Results After Applying a 3 Percent Discount Rate
Table 13.5 Results After Terminating Calculations at Age 65 or Older
Appendix A: Answer Key to Exercises
Table A.1 Comparison Between Vaccinated and Not Vaccinated Strategies
Table A.2 Markov Model Using a Vaccine Effectiveness of 75 Percent
Appendix B: Life Expectancy and Quality-Adjusted Life Expectancy Tables
Table B.1 Abridged Life Table for the Total Population, United States, 2011
Table B.2 Abridged Quality-Adjusted Life Table for the Total Population, United States, 2011
COST-EFFECTIVENESS ANALYSIS IN HEALTH
A PRACTICAL APPROACH
Third Edition
Peter Muennig
Mark Bounthavong
Copyright © 2016 by John Wiley & Sons, Inc. All rights reserved.
Published by Jossey-Bass
A Wiley Brand
One Montgomery Street, Suite 1000, San Francisco, CA 94104-4594–www.josseybass.com
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400, fax 978-646-8600, or on the Web at www.copyright.com. Requests to the publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, 201-748-6011, fax 201-748-6008, or online at www.wiley.com/go/permissions.
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Readers should be aware that Internet Web sites offered as citations and/or sources for further information may have changed or disappeared between the time this was written and when it is read.
Jossey-Bass books and products are available through most bookstores. To contact Jossey-Bass directly call our Customer Care Department within the U.S. at 800-956-7739, outside the U.S. at 317-572-3986, or fax 317-572-4002.
Wiley publishes in a variety of print and electronic formats and by print-on-demand. Some material included with standard print versions of this book may not be included in e-books or in print-on-demand. If this book refers to media such as a CD or DVD that is not included in the version you purchased, you may download this material at http://booksupport.wiley.com . For more information about Wiley products, visit www.wiley.com .
Library of Congress Cataloging-in-Publication Data
Muennig, Peter, author.
Cost-effectiveness analysis in health : a practical approach / Peter Muennig, Mark Bounthavong.-Third edition.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-1-119-01126-2 (paper), 978-1-119-01127-9 (epdf), 978-1-119-01128-6 (epub)
I. Bounthavong, Mark, 1976-, author. II. Title.
[DNLM: 1. Cost-Benefit Analysis-methods. 2. Health Care Costs. 3. Quality of Life. W 74.1]
RA410.5
338.4′33621–dc23
2015036511
Cover design: Wiley
Cover image: © ilolab/Shutterstock
List of Tables, Figures, and Exhibits
Tables
Table 1.1 Hypothetical League Table for a Village in Malawi with a $58,000 Health Budget
Table 2.1 Costs Included in a Cost-Effectiveness Analysis of Free Contraception, Conducted from Three Perspectives
Table 2.2 Hypothetical Differences in Health-Related Quality of Life over 10 Years for Diabetic Women and Women in Perfect Health
Table 2.3 Decision Matrix for Various Cost-Effectiveness Scenarios
Table 4.1 Comparison of Pharmaceutical Benchmark Prices
Table 4.2 Partial List of Costs for Treatment of Influenza Infection
Table 4.3 Common Codes Used to Group Diseases
Table 4.4 MEDPAR Cost Data by DRG for 2011
Table 4.5 Medical Component of the Consumer Price Index 2004–2014, Annual Percentage Change over Previous Year
Table 4.6 Hypothetical and Discounted Costs of a Cohort of 1,000 Elderly Persons over 10 Years
Table 5.1 Probabilities and Costs for Vaccinated and Not Vaccinated (Supportive Care) Strategies
Table 5.2 Calculation of Expected Cost for Each Event Pathway for Vaccination and Supportive Care Strategies
Table 6.1 Number of Deaths due to Influenza Virus Infection, by Age Group
Table 6.2 Deaths, Mean Age of Death due to Influenza Virus Infection, and Life Expectancy for Persons Aged 15 to 65
Table 6.3 Calculating Total Years of Life Lost due to Influenza Virus Infection in the United States
Table 6.4 Total Deaths, Deaths due to Influenza Virus Infection, and Total Survivors in a Cohort of 1 Million 15-Year-Olds
Table 6.5 Total Person-Years Lived by the Cohort of 1 Million 15-Year-Olds
Table 6.6 Person-Years Lived Among the Cohort of 15-Year-Olds, Including and Excluding Deaths due to Influenza Virus Infection
Table 6.7 Age-Specific Mortality Rates, Survivors, and Number of Deaths in the Cohort of 1 Million 15-Year-Old Subjects
Table 6.8 Progression of a Cohort of 10 Women with Breast Cancer over a Six-Year Period
Table 7.1 Example of How an HRQL Score for Influenza Illness May Be Derived Using the EQ-5D
Table 7.2 EQ-5D Preference Score Variation Among Age Categories
Table 8.1 Total Person-Years Lived by the Cohort of 1 Million 15-Year-Olds
Table 8.2 Sum of Person-Years Across Age Groups for the Cohort of 1 Million 15-Year-Olds
Table 8.3 Calculating Life Expectancy at a Given Age
Table 8.4 Abridged Life Table for 2011
Table 8.5 A Quality-Adjusted Life Table
Table 10.1 Simple Summary of Costs Used in a Cost-Effectiveness Model
Table 10.2 Cost-Effectiveness Table
Table 10.3 Example of a Cost-Effectiveness Table
Table 11.1 Calculating the Incidence Rate of Developing Cancer due to Exposure to Radiation in a Two-Year Observation Period, 2011–2012
Table 11.2 Calculating the Age-Adjusted Mortality Rate Using a Hypothetical U.S. Population
Table 11.3 Frequency Distribution of Hypothetical Cholesterol Values Obtained from 100 Subjects
Table 12.1 Datasets Useful for Finding Frequently Needed Cost-Effectiveness Parameters
Table 13.1 Results of the Base-Case Analysis
Table 13.2 Age-Indexed Table for Use in the New Tree
Table 13.3 Base-Case Results for the Markov Model
Table 13.4 Base-Case Results After Applying a 3 Percent Discount Rate
Table 13.5 Results After Terminating Calculations at Age 65 or Older
Table A.1 Comparison Between Vaccinated and Not Vaccinated Strategies
Table A.2 Markov Model Using a Vaccine Effectiveness of 75 Percent
Table B.1 Abridged Life Table for the Total Population, United States, 2011
Table B.2 Abridged Quality-Adjusted Life Table for the Total Population, United States, 2011
Figures
Figure 1.1 Example of the Effect of a Health Intervention on the Health States of Patients Admitted to the Emergency Room for an Acute Asthma Attack
Figure 1.2 Components of a Cost-Effectiveness Analysis
Figure 2.1 Graphical Representation of an HRQL Score
Figure 2.2 Difference in Total QALYs Between Women Treated and Not Treated for Diabetes over 10 Years
Figure 2.3 Graphical Representation of the Impact of High and Low Numerators and Denominators in Calculating Incremental Cost-Effectiveness Ratios
Figure 3.1 Flowchart Indicating the Clinical Course of Influenza Illness
Figure 3.2 Flowchart Indicating the Course of Influenza Infection Among Subjects Who Receive a Vaccination
Figure 3.3 Probability of Seeing a Doctor Among Subjects Who Receive Vaccination Versus Those Who Receive Supportive Care
Figure 3.4 Vaccination Strategy Represented with All Probabilities Filled In
Figure 3.5 Vaccination Decision Node
Figure 4.1 Partial Flowchart of the Course of the Flu
Figure 4.2 Costs Associated with the “Does Not See Doctor” Box
Figure 5.1 Decision Tree for Whether to Pursue Public Health School or Write a Novel
Figure 5.2 Decision Tree with the Potential for Not Finding a Job After Public Health School
Figure 5.3 Example of Two Mutually Exclusive Events
Figure 5.4 Probabilities of Outcomes for Patients Receiving the Vaccine Intervention
Figure 5.5 Course of Events During an Influenza Season Among Those Receiving Supportive Care Alone
Figure 5.6 Course of Events During an Influenza Season Among Those Receiving a Vaccination
Figure 5.7 Supportive Care Versus Vaccination Decision (Figure 5.5 and 5.6) Represented as a Decision Analysis Tree
Figure 5.8 Event Pathway for Vaccination Versus Supportive Care Represented as a Decision Analysis Tree
Figure 5.9 The Expected Probability for Each Terminal Node in the Vaccination and Supportive Care Decision Tree
Figure 5.10 The Total Cost for Each Terminal Node in the Vaccination and Supportive Care Decision Tree
Figure 5.11 Expected Costs for Each Terminal Node in the Vaccination and Supportive Care Decision Tree
Figure 5.12 Total Cost and Probability for All Terminal Nodes in the Vaccination and Supportive Care Decision Tree
Figure 5.13 Expected Cost and Outcomes for Each Chance Node in the Vaccinated and Supportive Care Decision Tree
Figure 5.14 Calculation for the Expected Cost and Probability for a Patient Who Receives Supportive Care, Becomes Ill, and Sees Doctor
Figure 5.15 Expected Costs and Outcomes for Different Chance Nodes in the Vaccinated and Supportive Care Decision Tree
Figure 6.1 Markov Model for Influenza Mortality in 15-Year-Olds
Figure 6.2 Basic Concept of a Markov Model
Figure 6.3 Complete Decision Analysis Tree for Calculating Life Expectancy Using TreeAge Pro
Figure 6.4 A Rolled-Back Model Using the Probability of Death for the General U.S. Population
Figure 7.1 Trade-off Between the Status Quo Health State and a Gamble
Figure 7.2 EQ-5D-5L Form Filled Out by a Patient
Figure 7.3 Diabetes Markov Model Depicting Three Health States: Mild, Moderate, and Severe Diabetes
Figure 8.1 Year-to-Year Progress of Treated and Untreated Subjects with Leishmaniasis
Figure 8.2 Basic Markov Model Used to Calculate Life Expectancy
Figure 8.3 Markov Models Designed to Calculate the Life Expectancy of Subjects Receiving the Filmore and Reinkenshein Procedures
Figure 8.4 Difference in HRQL Among Subjects Who Received the Filmore or the Reinkenshein Procedure
Figure 8.5 Filmore Versus Reinkenshein Model Rolled Back to Reveal Gains in Quality-Adjusted Life Expectancy Associated with Each Strategy
Figure 8.6 Filmore Versus Reinkenshein Model with Costs Added
Figure 8.7 Filmore Versus Reinkenshein Model with Discounting Added to the HRQL Values
Figure 8.8 Filmore Versus Reinkenshein Model Rolled Back
Figure 9.1 Sensitivity Analysis Focusing on Structure for the Vaccine Event Pathway
Figure 9.2 Sensitivity Analysis Focusing on the Parameter Change (Remains Well) for the Vaccine Event Pathway
Figure 9.3 Terminal Branch of the Filmore Arm Represented in Figure 8.3
Figure 9.4 Incremental Effectiveness of the Reinkenshein Procedure Relative to the Filmore Procedure over a Range of Risk Ratios
Figure 9.5 One-Way Sensitivity Analysis Examining How the Cost of Providing the Influenza Vaccine Influences Intervention
Figure 9.6 Two-Way Sensitivity Analysis Comparing Changes in the Efficacy of the Influenza Vaccine and the Incidence of Influenza-Like Illness
Figure 9.7 Tornado Diagram Example
Figure 9.8 Microsimulation of Individual Patients Through a Decision Path
Figure 9.9 Chance of Incurring Any Given Value of a Normally Distributed Variable
Figure 9.10 Diabetes Model in Which Values of Each Variable Are Normally Distributed
Figure 9.11 The Triangular Distribution
Figure 9.12 Other Distributions Used in Monte Carlo Simulations
Figure 9.13 Cost-Effectiveness Plane with a Single Simulation
Figure 9.14 Hypothetical Results of 100 Simulations on the Cost-Effectiveness Plane
Figure 9.15 Cost-Effectiveness Acceptability Curve
Figure 9.16 Cost-Effectiveness Acceptability Curve for Exercises 2 and 3
Figure 11.1 Nonrandom Error
Figure 11.2 Random Error
Figure 11.3 Graphical Representation of the 100 Cholesterol Values
Figure 11.4 Probability Distribution of the 100 Cholesterol Values
Figure 11.5 Hypothetical Probability Distribution of 1,000 Cholesterol Values
Figure 11.6 The Normal Curve
Figure 11.7 Example of a Triangular Distribution
Figure 11.8 Retrospective Study Designs: Case-Control and Cohort Designs
Figure 11.9 Prospective Study Design
Figure 11.10 Randomized Controlled Trial Study Design
Figure 11.11 Network of Studies Comparing Drugs Directly and Indirectly
Figure 12.1 Pyramid Analogy for the Different Levels of Evidence Criteria
Figure 12.2 An Example of a Jadad Score Grading Form
Figure 13.1 Starting a New Project in TreeAge Pro
Figure 13.2 Configuring the Model
Figure 13.3 Changing the Calculation Method to Cost-Effectiveness
Figure 13.4 Selecting the Payoffs for a Cost-Effectiveness Analysis
Figure 13.5 Selecting the Number of Payoffs
Figure 13.6 Changing Payoffs Names Under the “Custom Names” Option
Figure 13.7 Numerical Formatting for the Cost-Effectiveness Analysis
Figure 13.8 Saving a New Document in TreeAge Pro as a Package Using the *.trex Extension
Figure 13.9 Adding Two Chance Nodes to the Existing Decision Node
Figure 13.10 Deleting a Branch Using the Table Icons
Figure 13.11 Selecting a Markov Node Using “Change Type” from the Menu
Figure 13.12 Markov Node for a Decision Analysis Model
Figure 13.13 Creating Branches on a Markov Tree
Figure 13.14 Labeling Branches in a Markov Tree
Figure 13.15 Assigning an Initial Probability of 100 Percent to the Alive State in the Markov Model
Figure 13.16 Changing the Node from Chance Node to a Terminal Node
Figure 13.17 Edit Jump State Dialog Box Appears When the Chance Node Is Changed to a Terminal Node
Figure 13.18 Markov Tree After Changing the Remaining Chance Nodes into Terminal Nodes
Figure 13.19 Selecting the Markov Info View in Order to Enter Values
Figure 13.20 Selecting the Markov Info View Using the Icon on the Button Toolbar
Figure 13.21 Markov Info View Dialog Box
Figure 13.22 Setting the Initial and Incremental Rewards
Figure 13.23 Preference Box for the Markov Model
Figure 13.24 Selecting the “Variables/Markov Info” Option from the Tree Preferences Dialog Box
Figure 13.25 Termination Conditions for the Markov Node
Figure 13.26 Setting the Termination Conditions for the Markov Model
Figure 13.27 Creating a New Variable Using the New Variable Dialog Box
Figure 13.28 Assigning a Value to the Variable “age”
Figure 13.29 Defining the Variable “age”
Figure 13.30 Setting the Probability of Die as tdead2000[age], Which Is Derived from an n × 2 Matrix Table
Figure 13.31 Layout for Entering Values for a User-Defined Table
Figure 13.32 An Excel Table with the Probabilities of Death Associated with Each Age
Figure 13.33 Entering Values for tdead2000 Table
Figure 13.34 Selecting the Numeric Formatting Preferences from the Edit Menu
Figure 13.35 Changing the Payoff Units for Cost and Effectiveness
Figure 13.36 Overall Illustration of the Markov Model After Parameter Inputs and Unit Changes
Figure 13.37 Results of the Rollback Analysis for the Markov Model
Figure 13.38 Transition State Diagram of a Markov Model with Alive and Dead States
Figure 13.39 Results of Rollback Function
Figure 13.40 U.S. Life Table for Ages 0 to 100 and Over
Figure 13.41 Changing the Initial Cycle's Worth of Reward with the Half-Cycle Correction
Figure 13.42 Rollback Results Using the Half-Cycle Correction on the Initial Effectiveness
Figure 13.43 Opening the Define Initial Reward Window
Figure 13.44 Entering the Half-Cycle Correction into the Markov Information Dialog Box
Figure 13.45 Defining the Termination Condition to an Age That Is 105 Years or Older
Figure 13.46 Rollback Analysis with Half-Cycle Correction and Life Cycle Greater Than 105 Years
Figure 13.47 Selecting and Copying the Subtree
Figure 13.48 Adding a Second Subtree to the Decision Node
Figure 13.49 Multiplying Mortality Probability by a Factor of 1.25 and Capping It at 1
Figure 13.50 Changing Calculation Method from Cost-Effectiveness to Simple Analysis
Figure 13.51 Results of Rollback Between Filmore and Rinkenshein Procedures
Figure 13.52 Modifying the Payoff for a Cost-Effectiveness Analysis
Figure 13.53 Entering a Willingness to Pay of $40,000 per Life-Year Gained
Figure 13.54 Defining the Reward Set in the Initial Stage of the Markov Model
Figure 13.55 Rollback Results Comparing the Filmore and Rinkenshein Procedures
Figure 13.56 Selecting the Rankings from the Analysis Tab
Figure 13.57 Rankings Output Comparing the Cost-Effectiveness of Rinkenshein and Filmore Procedures
Figure 13.58 Insurance Versus No Insurance Competing Alternatives Model
Figure 13.59 Adding Tables Under the Tables View Option
Figure 13.60 Adding a New Table Under the Add/Change Box
Figure 13.61 The New Table, cInsurance, Is Not Listed in the Tables Window
Figure 13.62 Including Values for a User-Defined Table
Figure 13.63 Completed Tables for Costs and HRQL for Those Who Are Insured and Not Insured
Figure 13.64 Changing the Effectiveness Units to “QALYs”
Figure 13.65 Creating a New Variable Called “HR” and Defining Its Value
Figure 13.66 Defining the Value for “age” Under the Variables Properties Window
Figure 13.67 Formula for the Four New Variables
Figure 13.68 Adding the Half-Cycle Correction Factor for the Initial and Final Stages
Figure 13.69 Markov Model Comparing Insured and Uninsured Strategies with the Updated Variables and Tables
Figure 13.70 Rankings Output Comparing Insurance to No Insurance
Figure 13.71 Tree Properties with the Addition of a Discount Rate
Figure 13.72 All the Variables That Are Used in the Current Markov Model
Figure 13.73 Changing the Start Age at the Decision Node
Figure 13.74 Changing the Termination Conditions for the Insurance and No Insurance Arms of the Markov Model
Figure 13.75 One-Way Sensitivity Analysis Setup Window
Figure 13.76 Defining Low and High Values for the HR Parameter
Figure 13.77 Cost-Effectiveness Sensitivity Analysis Output Window
Figure 13.78 Results of the One-Way Sensitivity Analysis
Figure 13.79 Creating a New Distribution Variable
Figure 13.80 Add/Change Distribution Window
Figure 13.81 Assigning a Distribution to the Hazard Ratio Variable
Figure 13.82 Creating Another Distribution Called “dist_cInsur_err” Using a Triangular Distribution
Figure 13.83 Distributions for Three Variables (HR, cInsur, and cNoInsur) in the Model
Figure 13.84 Tree Properties with the Inclusion of Distributions
Figure 13.85 Selecting the Probabilistic Sensitivity Analysis
Figure 13.86 Monte Carlo Simulation Options Window
Figure 13.87 Selecting Incremental CE Ratio Output from the Monte Carlo Simulation Results Window
Figure 13.88 Distributions of ICERs Comparing Insurance Versus No Insurance Varying the Hazard Ratio and Cost Error Terms
Figure 13.89 Selecting the ICER Scatterplot Comparing Insurance to No Insurance
Figure 13.90 ICER Scatterplot Comparing Insurance to No Insurance
Figure 13.91 Proportion of ICER Scatterplots Below the Willingness-to-Pay Threshold of $40,000 per QALY Gained
Figure 13.92 Cost-Effectiveness Acceptability Curve Parameter Window
Figure 13.93 Cost-Effectiveness Acceptability Curves for Insurance and No Insurance
Figure A.1 The Rollback Results for the Expected Costs and Outcomes for Vaccinated and Not Vaccinated Strategies
Figure A.2 Markov Model from Chapter 6, Exercise 1
Figure A.3 A Tree Diagram with Rollback Results from TreeAge Pro
Figure A.4 Cost-Effectiveness Acceptability Curve Showing Where the 75 Percent Probability of Cost-Effectiveness Is in Relation to the Willingness-to-Pay Axis
Figure A.5 Using the CDC's Wonder Mortality Database and Selecting Breast Cancer
Figure A.6 Selecting the ICD-10 Code for Breast Cancer
Figure A.7 Results of the Breast Cancer Query Grouped by Gender for 2011
Exhibits
2.1 How the Recommendations for Numerator and Denominator Values in the Incremental Cost-Effectiveness Ratio have Changed with the New Recommendations in 2017
4.2 Can an Economic Crisis Reduce the Value of Human Life?
7.1 Exhibit 7.1 EQ-5D-5L Health Domains
10.1 Exhibit The Complete CHEERS Checklist
10.2 The Impact Inventory Checklist
12.1 Example of a Table Created by Cross-Tabulation (Contingency Table)
12.2 Major U.S. Health Datasets Available to the Public
12.3 Major Sources of International Health Data
12.4 Some Data Sources for Which Data Extraction Tools Are Available