Cover Page



Third Edition

Peter Muennig

Mark Bounthavong




Title Page

List of Tables, Figures, and Exhibits


  1. Table 1.1 Hypothetical League Table for a Village in Malawi with a $58,000 Health Budget
  2. Table 2.1 Costs Included in a Cost-Effectiveness Analysis of Free Contraception, Conducted from Three Perspectives
  3. Table 2.2 Hypothetical Differences in Health-Related Quality of Life over 10 Years for Diabetic Women and Women in Perfect Health
  4. Table 2.3 Decision Matrix for Various Cost-Effectiveness Scenarios
  5. Table 4.1 Comparison of Pharmaceutical Benchmark Prices
  6. Table 4.2 Partial List of Costs for Treatment of Influenza Infection
  7. Table 4.3 Common Codes Used to Group Diseases
  8. Table 4.4 MEDPAR Cost Data by DRG for 2011
  9. Table 4.5 Medical Component of the Consumer Price Index 2004–2014, Annual Percentage Change over Previous Year
  10. Table 4.6 Hypothetical and Discounted Costs of a Cohort of 1,000 Elderly Persons over 10 Years
  11. Table 5.1 Probabilities and Costs for Vaccinated and Not Vaccinated (Supportive Care) Strategies
  12. Table 5.2 Calculation of Expected Cost for Each Event Pathway for Vaccination and Supportive Care Strategies
  13. Table 6.1 Number of Deaths due to Influenza Virus Infection, by Age Group
  14. Table 6.2 Deaths, Mean Age of Death due to Influenza Virus Infection, and Life Expectancy for Persons Aged 15 to 65
  15. Table 6.3 Calculating Total Years of Life Lost due to Influenza Virus Infection in the United States
  16. Table 6.4 Total Deaths, Deaths due to Influenza Virus Infection, and Total Survivors in a Cohort of 1 Million 15-Year-Olds
  17. Table 6.5 Total Person-Years Lived by the Cohort of 1 Million 15-Year-Olds
  18. Table 6.6 Person-Years Lived Among the Cohort of 15-Year-Olds, Including and Excluding Deaths due to Influenza Virus Infection
  19. Table 6.7 Age-Specific Mortality Rates, Survivors, and Number of Deaths in the Cohort of 1 Million 15-Year-Old Subjects
  20. Table 6.8 Progression of a Cohort of 10 Women with Breast Cancer over a Six-Year Period
  21. Table 7.1 Example of How an HRQL Score for Influenza Illness May Be Derived Using the EQ-5D
  22. Table 7.2 EQ-5D Preference Score Variation Among Age Categories
  23. Table 8.1 Total Person-Years Lived by the Cohort of 1 Million 15-Year-Olds
  24. Table 8.2 Sum of Person-Years Across Age Groups for the Cohort of 1 Million 15-Year-Olds
  25. Table 8.3 Calculating Life Expectancy at a Given Age
  26. Table 8.4 Abridged Life Table for 2011
  27. Table 8.5 A Quality-Adjusted Life Table
  28. Table 10.1 Simple Summary of Costs Used in a Cost-Effectiveness Model
  29. Table 10.2 Cost-Effectiveness Table
  30. Table 10.3 Example of a Cost-Effectiveness Table
  31. Table 11.1 Calculating the Incidence Rate of Developing Cancer due to Exposure to Radiation in a Two-Year Observation Period, 2011–2012
  32. Table 11.2 Calculating the Age-Adjusted Mortality Rate Using a Hypothetical U.S. Population
  33. Table 11.3 Frequency Distribution of Hypothetical Cholesterol Values Obtained from 100 Subjects
  34. Table 12.1 Datasets Useful for Finding Frequently Needed Cost-Effectiveness Parameters
  35. Table 13.1 Results of the Base-Case Analysis
  36. Table 13.2 Age-Indexed Table for Use in the New Tree
  37. Table 13.3 Base-Case Results for the Markov Model
  38. Table 13.4 Base-Case Results After Applying a 3 Percent Discount Rate
  39. Table 13.5 Results After Terminating Calculations at Age 65 or Older
  40. Table A.1 Comparison Between Vaccinated and Not Vaccinated Strategies
  41. Table A.2 Markov Model Using a Vaccine Effectiveness of 75 Percent
  42. Table B.1 Abridged Life Table for the Total Population, United States, 2011
  43. Table B.2 Abridged Quality-Adjusted Life Table for the Total Population, United States, 2011


  1. 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
  2. Figure 1.2 Components of a Cost-Effectiveness Analysis
  3. Figure 2.1 Graphical Representation of an HRQL Score
  4. Figure 2.2 Difference in Total QALYs Between Women Treated and Not Treated for Diabetes over 10 Years
  5. Figure 2.3 Graphical Representation of the Impact of High and Low Numerators and Denominators in Calculating Incremental Cost-Effectiveness Ratios
  6. Figure 3.1 Flowchart Indicating the Clinical Course of Influenza Illness
  7. Figure 3.2 Flowchart Indicating the Course of Influenza Infection Among Subjects Who Receive a Vaccination
  8. Figure 3.3 Probability of Seeing a Doctor Among Subjects Who Receive Vaccination Versus Those Who Receive Supportive Care
  9. Figure 3.4 Vaccination Strategy Represented with All Probabilities Filled In
  10. Figure 3.5 Vaccination Decision Node
  11. Figure 4.1 Partial Flowchart of the Course of the Flu
  12. Figure 4.2 Costs Associated with the “Does Not See Doctor” Box
  13. Figure 5.1 Decision Tree for Whether to Pursue Public Health School or Write a Novel
  14. Figure 5.2 Decision Tree with the Potential for Not Finding a Job After Public Health School
  15. Figure 5.3 Example of Two Mutually Exclusive Events
  16. Figure 5.4 Probabilities of Outcomes for Patients Receiving the Vaccine Intervention
  17. Figure 5.5 Course of Events During an Influenza Season Among Those Receiving Supportive Care Alone
  18. Figure 5.6 Course of Events During an Influenza Season Among Those Receiving a Vaccination
  19. Figure 5.7 Supportive Care Versus Vaccination Decision (Figure 5.5 and 5.6) Represented as a Decision Analysis Tree
  20. Figure 5.8 Event Pathway for Vaccination Versus Supportive Care Represented as a Decision Analysis Tree
  21. Figure 5.9 The Expected Probability for Each Terminal Node in the Vaccination and Supportive Care Decision Tree
  22. Figure 5.10 The Total Cost for Each Terminal Node in the Vaccination and Supportive Care Decision Tree
  23. Figure 5.11 Expected Costs for Each Terminal Node in the Vaccination and Supportive Care Decision Tree
  24. Figure 5.12 Total Cost and Probability for All Terminal Nodes in the Vaccination and Supportive Care Decision Tree
  25. Figure 5.13 Expected Cost and Outcomes for Each Chance Node in the Vaccinated and Supportive Care Decision Tree
  26. Figure 5.14 Calculation for the Expected Cost and Probability for a Patient Who Receives Supportive Care, Becomes Ill, and Sees Doctor
  27. Figure 5.15 Expected Costs and Outcomes for Different Chance Nodes in the Vaccinated and Supportive Care Decision Tree
  28. Figure 6.1 Markov Model for Influenza Mortality in 15-Year-Olds
  29. Figure 6.2 Basic Concept of a Markov Model
  30. Figure 6.3 Complete Decision Analysis Tree for Calculating Life Expectancy Using TreeAge Pro
  31. Figure 6.4 A Rolled-Back Model Using the Probability of Death for the General U.S. Population
  32. Figure 7.1 Trade-off Between the Status Quo Health State and a Gamble
  33. Figure 7.2 EQ-5D-5L Form Filled Out by a Patient
  34. Figure 7.3 Diabetes Markov Model Depicting Three Health States: Mild, Moderate, and Severe Diabetes
  35. Figure 8.1 Year-to-Year Progress of Treated and Untreated Subjects with Leishmaniasis
  36. Figure 8.2 Basic Markov Model Used to Calculate Life Expectancy
  37. Figure 8.3 Markov Models Designed to Calculate the Life Expectancy of Subjects Receiving the Filmore and Reinkenshein Procedures
  38. Figure 8.4 Difference in HRQL Among Subjects Who Received the Filmore or the Reinkenshein Procedure
  39. Figure 8.5 Filmore Versus Reinkenshein Model Rolled Back to Reveal Gains in Quality-Adjusted Life Expectancy Associated with Each Strategy
  40. Figure 8.6 Filmore Versus Reinkenshein Model with Costs Added
  41. Figure 8.7 Filmore Versus Reinkenshein Model with Discounting Added to the HRQL Values
  42. Figure 8.8 Filmore Versus Reinkenshein Model Rolled Back
  43. Figure 9.1 Sensitivity Analysis Focusing on Structure for the Vaccine Event Pathway
  44. Figure 9.2 Sensitivity Analysis Focusing on the Parameter Change (Remains Well) for the Vaccine Event Pathway
  45. Figure 9.3 Terminal Branch of the Filmore Arm Represented in Figure 8.3
  46. Figure 9.4 Incremental Effectiveness of the Reinkenshein Procedure Relative to the Filmore Procedure over a Range of Risk Ratios
  47. Figure 9.5 One-Way Sensitivity Analysis Examining How the Cost of Providing the Influenza Vaccine Influences Intervention
  48. Figure 9.6 Two-Way Sensitivity Analysis Comparing Changes in the Efficacy of the Influenza Vaccine and the Incidence of Influenza-Like Illness
  49. Figure 9.7 Tornado Diagram Example
  50. Figure 9.8 Microsimulation of Individual Patients Through a Decision Path
  51. Figure 9.9 Chance of Incurring Any Given Value of a Normally Distributed Variable
  52. Figure 9.10 Diabetes Model in Which Values of Each Variable Are Normally Distributed
  53. Figure 9.11 The Triangular Distribution
  54. Figure 9.12 Other Distributions Used in Monte Carlo Simulations
  55. Figure 9.13 Cost-Effectiveness Plane with a Single Simulation
  56. Figure 9.14 Hypothetical Results of 100 Simulations on the Cost-Effectiveness Plane
  57. Figure 9.15 Cost-Effectiveness Acceptability Curve
  58. Figure 9.16 Cost-Effectiveness Acceptability Curve for Exercises 2 and 3
  59. Figure 11.1 Nonrandom Error
  60. Figure 11.2 Random Error
  61. Figure 11.3 Graphical Representation of the 100 Cholesterol Values
  62. Figure 11.4 Probability Distribution of the 100 Cholesterol Values
  63. Figure 11.5 Hypothetical Probability Distribution of 1,000 Cholesterol Values
  64. Figure 11.6 The Normal Curve
  65. Figure 11.7 Example of a Triangular Distribution
  66. Figure 11.8 Retrospective Study Designs: Case-Control and Cohort Designs
  67. Figure 11.9 Prospective Study Design
  68. Figure 11.10 Randomized Controlled Trial Study Design
  69. Figure 11.11 Network of Studies Comparing Drugs Directly and Indirectly
  70. Figure 12.1 Pyramid Analogy for the Different Levels of Evidence Criteria
  71. Figure 12.2 An Example of a Jadad Score Grading Form
  72. Figure 13.1 Starting a New Project in TreeAge Pro
  73. Figure 13.2 Configuring the Model
  74. Figure 13.3 Changing the Calculation Method to Cost-Effectiveness
  75. Figure 13.4 Selecting the Payoffs for a Cost-Effectiveness Analysis
  76. Figure 13.5 Selecting the Number of Payoffs
  77. Figure 13.6 Changing Payoffs Names Under the “Custom Names” Option
  78. Figure 13.7 Numerical Formatting for the Cost-Effectiveness Analysis
  79. Figure 13.8 Saving a New Document in TreeAge Pro as a Package Using the *.trex Extension
  80. Figure 13.9 Adding Two Chance Nodes to the Existing Decision Node
  81. Figure 13.10 Deleting a Branch Using the Table Icons
  82. Figure 13.11 Selecting a Markov Node Using “Change Type” from the Menu
  83. Figure 13.12 Markov Node for a Decision Analysis Model
  84. Figure 13.13 Creating Branches on a Markov Tree
  85. Figure 13.14 Labeling Branches in a Markov Tree
  86. Figure 13.15 Assigning an Initial Probability of 100 Percent to the Alive State in the Markov Model
  87. Figure 13.16 Changing the Node from Chance Node to a Terminal Node
  88. Figure 13.17 Edit Jump State Dialog Box Appears When the Chance Node Is Changed to a Terminal Node
  89. Figure 13.18 Markov Tree After Changing the Remaining Chance Nodes into Terminal Nodes
  90. Figure 13.19 Selecting the Markov Info View in Order to Enter Values
  91. Figure 13.20 Selecting the Markov Info View Using the Icon on the Button Toolbar
  92. Figure 13.21 Markov Info View Dialog Box
  93. Figure 13.22 Setting the Initial and Incremental Rewards
  94. Figure 13.23 Preference Box for the Markov Model
  95. Figure 13.24 Selecting the “Variables/Markov Info” Option from the Tree Preferences Dialog Box
  96. Figure 13.25 Termination Conditions for the Markov Node
  97. Figure 13.26 Setting the Termination Conditions for the Markov Model
  98. Figure 13.27 Creating a New Variable Using the New Variable Dialog Box
  99. Figure 13.28 Assigning a Value to the Variable “age”
  100. Figure 13.29 Defining the Variable “age”
  101. Figure 13.30 Setting the Probability of Die as tdead2000[age], Which Is Derived from an n × 2 Matrix Table
  102. Figure 13.31 Layout for Entering Values for a User-Defined Table
  103. Figure 13.32 An Excel Table with the Probabilities of Death Associated with Each Age
  104. Figure 13.33 Entering Values for tdead2000 Table
  105. Figure 13.34 Selecting the Numeric Formatting Preferences from the Edit Menu
  106. Figure 13.35 Changing the Payoff Units for Cost and Effectiveness
  107. Figure 13.36 Overall Illustration of the Markov Model After Parameter Inputs and Unit Changes
  108. Figure 13.37 Results of the Rollback Analysis for the Markov Model
  109. Figure 13.38 Transition State Diagram of a Markov Model with Alive and Dead States
  110. Figure 13.39 Results of Rollback Function
  111. Figure 13.40 U.S. Life Table for Ages 0 to 100 and Over
  112. Figure 13.41 Changing the Initial Cycle's Worth of Reward with the Half-Cycle Correction
  113. Figure 13.42 Rollback Results Using the Half-Cycle Correction on the Initial Effectiveness
  114. Figure 13.43 Opening the Define Initial Reward Window
  115. Figure 13.44 Entering the Half-Cycle Correction into the Markov Information Dialog Box
  116. Figure 13.45 Defining the Termination Condition to an Age That Is 105 Years or Older
  117. Figure 13.46 Rollback Analysis with Half-Cycle Correction and Life Cycle Greater Than 105 Years
  118. Figure 13.47 Selecting and Copying the Subtree
  119. Figure 13.48 Adding a Second Subtree to the Decision Node
  120. Figure 13.49 Multiplying Mortality Probability by a Factor of 1.25 and Capping It at 1
  121. Figure 13.50 Changing Calculation Method from Cost-Effectiveness to Simple Analysis
  122. Figure 13.51 Results of Rollback Between Filmore and Rinkenshein Procedures
  123. Figure 13.52 Modifying the Payoff for a Cost-Effectiveness Analysis
  124. Figure 13.53 Entering a Willingness to Pay of $40,000 per Life-Year Gained
  125. Figure 13.54 Defining the Reward Set in the Initial Stage of the Markov Model
  126. Figure 13.55 Rollback Results Comparing the Filmore and Rinkenshein Procedures
  127. Figure 13.56 Selecting the Rankings from the Analysis Tab
  128. Figure 13.57 Rankings Output Comparing the Cost-Effectiveness of Rinkenshein and Filmore Procedures
  129. Figure 13.58 Insurance Versus No Insurance Competing Alternatives Model
  130. Figure 13.59 Adding Tables Under the Tables View Option
  131. Figure 13.60 Adding a New Table Under the Add/Change Box
  132. Figure 13.61 The New Table, cInsurance, Is Not Listed in the Tables Window
  133. Figure 13.62 Including Values for a User-Defined Table
  134. Figure 13.63 Completed Tables for Costs and HRQL for Those Who Are Insured and Not Insured
  135. Figure 13.64 Changing the Effectiveness Units to “QALYs”
  136. Figure 13.65 Creating a New Variable Called “HR” and Defining Its Value
  137. Figure 13.66 Defining the Value for “age” Under the Variables Properties Window
  138. Figure 13.67 Formula for the Four New Variables
  139. Figure 13.68 Adding the Half-Cycle Correction Factor for the Initial and Final Stages
  140. Figure 13.69 Markov Model Comparing Insured and Uninsured Strategies with the Updated Variables and Tables
  141. Figure 13.70 Rankings Output Comparing Insurance to No Insurance
  142. Figure 13.71 Tree Properties with the Addition of a Discount Rate
  143. Figure 13.72 All the Variables That Are Used in the Current Markov Model
  144. Figure 13.73 Changing the Start Age at the Decision Node
  145. Figure 13.74 Changing the Termination Conditions for the Insurance and No Insurance Arms of the Markov Model
  146. Figure 13.75 One-Way Sensitivity Analysis Setup Window
  147. Figure 13.76 Defining Low and High Values for the HR Parameter
  148. Figure 13.77 Cost-Effectiveness Sensitivity Analysis Output Window
  149. Figure 13.78 Results of the One-Way Sensitivity Analyses
  150. Figure 13.79 Creating a New Distribution Variable
  151. Figure 13.80 Add/Change Distribution Window
  152. Figure 13.81 Assigning a Distribution to the Hazard Ratio Variable
  153. Figure 13.82 Creating Another Distribution Called “dist_cInsur_err” Using a Triangular Distribution
  154. Figure 13.83 Distributions for Three Variables (HR, cInsur, and cNoInsur) in the Model
  155. Figure 13.84 Tree Properties with the Inclusion of Distributions
  156. Figure 13.85 Selecting the Probabilistic Sensitivity Analysis
  157. Figure 13.86 Monte Carlo Simulation Options Window
  158. Figure 13.87 Selecting Incremental CE Ratio Output from the Monte Carlo Simulation Results Window
  159. Figure 13.88 Distributions of ICERs Comparing Insurance Versus No Insurance Varying the Hazard Ratio and Cost Error Terms
  160. Figure 13.89 Selecting the ICER Scatterplot Comparing Insurance to No Insurance
  161. Figure 13.90 ICER Scatterplot Comparing Insurance to No Insurance
  162. Figure 13.91 Proportion of ICER Scatterplots Below the Willingness-to-Pay Threshold of $40,000 per QALY Gained
  163. Figure 13.92 Cost-Effectiveness Acceptability Curve Parameter Window
  164. Figure 13.93 Cost-Effectiveness Acceptability Curves for Insurance and No Insurance
  165. Figure A.1 The Rollback Results for the Expected Costs and Outcomes for Vaccinated and Not Vaccinated Strategies
  166. Figure A.2 Markov Model from Chapter 6, Exercise 1
  167. Figure A.3 A Tree Diagram with Rollback Results from TreeAge Pro
  168. 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
  169. Figure A.5 Using the CDC's Wonder Mortality Database and Selecting Breast Cancer
  170. Figure A.6 Selecting the ICD-10 Code for Breast Cancer
  171. Figure A.7 Results of the Breast Cancer Query Grouped by Gender for 2011


  1. 2.1 How the Recommendations for Numerator and Denominator Values in the Incremental Cost-Effectiveness Ratio have Changed with the New Recommendations in 2017
  2. 4.2 Can an Economic Crisis Reduce the Value of Human Life?
  3. 7.1 Exhibit 7.1 EQ-5D-5L Health Domains
  4. 10.1 Exhibit The Complete CHEERS Checklist
  5. 10.2 The Impact Inventory Checklist
  6. 12.1 Example of a Table Created by Cross-Tabulation (Contingency Table)
  7. 12.2 Major U.S. Health Datasets Available to the Public
  8. 12.3 Major Sources of International Health Data
  9. 12.4 Some Data Sources for Which Data Extraction Tools Are Available