Multi-Criteria Decision Analysis to Support Healthcare Decisions (eBook)

Dr Kevin Marsh is Executive Director responsible for Patient Preference work at Evidera. Prior to joining Evidera’s London office in April 2012, Dr Marsh completed his PhD at the University of Bath, specializing in economic valuation techniques. After a year at Oxford University, he joined the Matrix Knowledge Group in London, where he built their economics practice. Dr Marsh’s research interests include the use of economic and decision analysis to inform health decisions, including pipeline optimization, authorization, reimbursement, and prescription decisions. His expertise include decision modeling, multi-criteria decision analysis (MCDA), and a range of economic valuation methods, such as stated preference value approaches. He recently co-chaired the ISPOR Taskforce on the use of MCDA in Health Care Decision-Making, and he is a co-convener of the Campbell and Cochrane Economic Methods Group.Dr Mireille Goetghebeur is Global Scientist at LASER Analytica, Adjunct Professor at the School of Public Health, University of Montreal and President of the EVIDEM Collaboration. She holds a PhD in basic sciences and an Engineering Diploma from France. Pioneering pragmatic decisionmaking approaches building on ethics, HTA and MCDA, her motivation is to stimulate meaningful evidence generation, holistic evaluation of healthcare interventions and accountable, reasonable and deliberative decisionmaking processes to tackle the ethical dilemmas of our time. She contributes to university programs in several countries and is involved in initiatives with organizations such as the World Health Organization, government agencies, and academic centers around the world. She collaborates with stakeholders across the decision continuum to optimize patient and population health and to develop effective, equitable and sustainable healthcare systems locally and globally.Praveen Thokala joined the School of Health and Related Research (ScHARR) at the University of Sheffield in March 2010 after completing an MASc from the University of Toronto and a PhD from the University of Southampton. His research interests include health economic modelling, multi-criteria decision analysis, discrete event simulation modelling and optimisation. He recently co-chaired the ISPOR Taskforce on the use of MCDA in Health Care Decision-Making.Dr. Rob Baltussen is an economist specialized in international health economics, with a PhD in costing and cost-effectiveness analysis. He has extensive field work experience in around 25 countries in Africa and Asia, and worked as a senior health economist at the World Health Organization in Geneva. At WHO, he was responsible for the development of WHO guidelines on cost-effectiveness analysis. At present, he is Head of Research at NICHE – the Nijmegen International Center for Health Systems Research and Education, located at the Radboud University Nijmegen Medical Center in the Netherlands. NICHE focuses on international health systems research, more specifically on improving access to health care for poor people, cost and cost-effectiveness analysis of SRH and HIV/AIDS, multi-criteria priority setting, and primary health care. Rob Baltussen is editor-in-chief of the journal Cost-Effectiveness and Resource Allocation and has published more than 100 international papers in Medline-indexed journals.

Contents 5
Chapter 1: Introduction 7
1.1 Introduction 7
1.2 Process of Developing the Book 9
1.3 Outline of the Book 9
1.4 Future Direction 10
References 11
Part I: Foundations of MCDA in Healthcare 13
Chapter 2: Theoretical Foundations of MCDA 14
2.1 Introduction 15
2.2 Principles of MCDA and Decision-Making 16
2.3 Problem Structuring 17
2.4 Model Building 20
2.4.1 Value Measurement 21
2.4.1.1 Multi-attribute Value Theory 21
2.4.1.2 Multi-Attribute Utility Theory (MAUT) 23
2.4.1.3 Analytic Hierarchy Process 24
2.4.2 Outranking 26
2.4.3 Goal Programming 28
2.5 Concluding Remarks 31
References 31
Chapter 3: Identifying Value(s): A Reflection on the Ethical Aspects of MCDA in Healthcare Decisionmaking 34
3.1 Introduction 35
3.2 Who Should Decide? Legitimacy of Decisions and Representativeness of MCDA Users 37
3.3 How to Decide? 39
3.3.1 Step 1: Defining the Decision Problem 39
3.3.2 Step 2: Selecting and Structuring Criteria 39
3.3.2.1 Patient 41
3.3.2.2 Population 41
3.3.2.3 Healthcare Systems 42
3.3.2.4 Knowledge and Context 43
3.3.3 Step 3 of MCDA: Weighting Criteria 44
3.3.4 Step 4 of MCDA: Providing Evidence to Measure Performance 45
3.3.5 Step 5: Scoring the Criteria to Evaluate Performance of the Intervention 45
3.3.6 Step 6: Aggregating Data for Ranking, Investing, and Disinvesting 46
3.3.7 Step 7: Dealing with Uncertainty 46
3.3.8 Step 8: Reporting Results, Deliberation, Decision, Communication, and Implementation 47
3.4 Conclusion 47
References 49
Chapter 4: Incorporating Preferences and Priorities into MCDA: Selecting an Appropriate Scoring and Weighting Technique 52
4.1 Introduction 52
4.2 Overview of Weighting and Scoring Techniques 54
4.2.1 Direct Rating 55
4.2.2 Keeney-Raiffa MCDA 55
4.2.2.1 Construction of Partial Value Functions 56
4.2.2.2 Swing Weighting 57
4.2.3 Pairwise Comparison Using Ordinal Scales (Analytic Hierarchy Process) 57
4.2.4 Discrete Choice Experiment 59
4.3 Which Scoring and Weighting Techniques Are Most Appropriate? 61
4.3.1 ‘Validity’ of Scores and Weights 62
4.3.1.1 Do Scores Display Interval Properties? 62
4.3.1.2 Do Weights Reflect Scaling Constant or Trade-Offs? 64
4.3.2 Cognitive Burden on Stakeholders 64
4.3.3 Interpreting the Outputs from MCDA 65
4.3.4 Practical Challenges 66
4.4 Discussion 67
References 69
Chapter 5: Dealing with Uncertainty in the Analysis and Reporting of MCDA 72
5.1 Introduction 72
5.1.1 Problem Structuring 73
5.1.2 Uncertainty in Problem Structuring 75
5.2 Uncertainty in Scoring 77
5.2.1 Performance Estimates 77
5.2.2 From Performance to Value 77
5.3 Uncertainty in Weighting 79
5.4 Aggregation Methods 81
5.5 Sensitivity Analysis 82
5.6 Summary and Conclusions 88
References 88
Part II: Applications and Case Studies 91
Chapter 6: Supporting the Project Portfolio Selection Decision of Research and Development Investments by Means of Multi-Criteria Resource Allocation Modelling 92
6.1 Introduction 93
6.2 Case Study and Method 95
6.2.1 Case Study 95
6.2.2 Resource Allocation Modelling on Robotic Innovations with MACBETH 95
6.3 Results 98
6.4 Lessons Learned and Discussion 103
References 105
Chapter 7: Benefit–Risk Assessment 107
7.1 Introduction 107
7.1.1 The Purpose and Timing of Benefit–Risk Assessments 107
7.1.2 History of Benefit–Risk Assessment and Methodologies 108
7.2 Overview of the Development of an MCDA-Based Benefit–Risk Model 109
7.2.1 Introduction 109
7.2.2 Establishing the Decision Context 110
7.2.3 Identification of the Alternatives to Be Appraised 110
7.2.4 Identification of the Criteria 111
7.2.5 Scoring of the Criteria 113
7.2.6 Assigning Weights to the Criteria 114
7.3 Examples of MCDA-Based Benefit–Risk Models 114
7.4 Outstanding Issues of MCDA-Based Benefit–Risk Models 116
7.5 Conclusion: The Place of MCDA-Based Benefit–Risk Models 117
Bibliography 118
Chapter 8: Advancing MCDA and HTA into Coverage Decision-Making 121
8.1 Introduction 122
8.2 The Case Studies 123
8.2.1 Testing MCDA in Colombia 123
8.2.1.1 Methods 124
Selecting Criteria and Assigning Weights 124
Assembling the Evidence for Selected Technologies 125
Appraisal of Interventions and Discussion 125
8.2.1.2 Results 128
8.2.2 Institutional HTA/MCDA Approach in Lombardy 129
8.2.2.1 Historical Perspective and Rationale for Developing and Implementing an MCDA-Based Appraisal Process 130
8.2.2.2 The Appraisal Process 130
8.2.2.3 MCDA Outputs in Lombardy 132
8.2.2.4 Latest Developments in Lombardy and Future in Italy 133
8.2.3 Developing an MCDA Approach for Coverage Decision-­Making in Belgium 134
8.2.3.1 Transparency 136
8.2.3.2 A Belgian MCDA Framework 136
8.2.3.3 Deriving Preferences for Healthcare Reimbursement Criteria from the General Public 137
8.2.3.4 Belgian Weights for Reimbursement Criteria 139
Therapeutic Need 139
Societal Need 140
Added Value 140
Using the MCDA in Decision-Making 140
8.3 Discussion 142
8.4 Conclusions 145
References 145
Chapter 9: Embedding MCDA in Priority-Setting Policy Processes: Experiences from Low- and Middle-Income Countries 149
9.1 MCDA to Support Priority Setting in LMICs 150
9.2 Overview of MCDA Case Studies in LMICs 150
9.2.1 General Characteristics 151
9.2.2 Findings on Policy-Related Issues 151
9.3 The Application of MCDA in a Policy-Making Process: HIV/AIDS Strategic Planning at Provincial Level in Indonesia 161
9.3.1 Implementation of MCDA During the Strategic Planning Process for HIV/AIDS Control in Indonesia 162
9.3.2 Barriers and Enablers for the Application of MCDA in a Policy-Making Context 167
9.4 Recommendations for Future Applications of MCDA 169
9.4.1 Methodological Guidance on Which Stakeholders to Involve and Why 169
9.4.2 Institutionalization of MCDA Including Capacity Building for HTA Research 171
9.4.3 Evaluation of the Impact of MCDA Approaches 172
9.5 Conclusion 173
References 173
Chapter 10: MCDA for Resource Allocation at a Local Level: An Application in the UK 176
10.1 Introduction 177
10.2 Overview of MCDA Methods for Priority Setting at Local Level 178
10.2.1 Option Appraisal, Portsmouth Scorecard and Other Prioritisation Matrices to Guide Deliberations 178
10.2.2 Socio-Technical Allocation of Resources (Star) 180
10.2.3 Programme Budgeting and Marginal Analysis (PBMA) 181
10.2.4 MCDA Using Discrete Choice Experiments (DCEs) 181
10.2.5 The South Yorkshire Multi-criteria Decision Analysis Prioritisation for Local Effectiveness (SYMPLE) Approach 183
10.3 Discussion 183
10.4 Case Study: Using MCDA to Inform Local Government Investment in Public Health 186
10.4.1 Defining the Decision Problem 186
10.4.2 Selecting and Structuring Criteria 188
10.4.3 Weighting Criteria 189
10.4.4 Measuring Performance 189
10.4.5 Scoring Alternatives 190
10.4.6 Calculating Aggregate Scores 191
10.4.7 Dealing with Uncertainty 191
10.4.8 Reporting and Examination of Findings 194
10.4.9 Reflections 194
10.5 Limitations of This Study 196
10.6 Conclusions 197
References 198
Chapter 11: Shared Decision-Making 200
11.1 Introduction 201
11.1.1 Medical Decisions Are Frequently Complex and Involve Trade-Offs Among Alternatives 201
11.1.2 What Is a Good Decision? 201
11.1.3 Relationship Between Shared Decision-Making and MCDA 202
11.2 Conjoint Analysis 202
11.2.1 Lupus Nephritis 204
11.2.2 Knee Osteoarthritis 206
11.3 The Analytic Hierarchy Process 208
11.3.1 Colorectal Cancer 211
11.3.2 The Colorectal Cancer Studies 211
11.3.3 Colorectal Cancer Study 1 212
11.3.4 Colorectal Cancer Study 2 213
11.4 Overall Results and Conclusions 214
References 215
Chapter 12: The Role of MCDA in Health Research Priority Setting 217
12.1 Introduction 218
12.2 What Are the Similarities and Differences Between Health Research and Intervention Priority Setting? 219
12.2.1 Similarities 219
12.2.2 Differences 219
12.3 Health Research Priority Setting and MCDA 222
12.3.1 Methodologies to Health Research Priority Setting (And Their Link with MCDA) 222
12.3.1.1 Three Types of Methodologies 222
12.3.1.2 An Implicit Link Between Health Research Priority Setting and MCDA 226
12.3.2 Health Research Priority Exercises and MCDA 229
12.3.3 Implications: Health Research Priority Setting and MCDA 231
12.4 Methodological Development in Health Research Priority Setting: The Way Forward 232
References 234
Chapter 13: MCDA for the Development of Clinical Practice Guidelines and for the Prioritization Clinical Research Questions 238
13.1 Introduction 239
13.2 Case Study 1: MCDA for Structuring Evidence and Identifying Most Important Outcomes 250
13.2.1 MCDA to Structure Evidence and Clinical Practice Questions 251
13.2.2 MCDA to Identify Most Important Outcomes 258
13.3 Case Study 2: MCDA for Clinical Research Prioritization 260
13.3.1 MCDA for Clinical Research Prioritization: Case Study of Hemodialysis 261
13.3.2 MCDA for Clinical Research Prioritization: Case Study of Registry Trial Prioritization for the Kidney, Dialysis, and Transplantation (KDT) Program 265
13.4 Conclusion 268
References 269
Part III: Future Directions 273
Chapter 14: Using MCDA as a Decision Aid in Health Technology Appraisal for Coverage Decisions: Opportunities, Challenges and Unresolved Questions 274
14.1 Introduction 275
14.2 Why Do We Need MCDA in HTA? 276
14.3 Do the Criteria and Weights Need to Be ‘Fixed’ in Order to Provide Consistency Between HTA Decisions? 282
14.4 Whose Criteria? 284
14.5 Whose Preferences Should Be Used in Weighting Criteria? 285
14.6 Can MCDA Incorporate the Concept of Opportunity Cost of New Technologies? 287
14.7 How Can Uncertainty Be Addressed in MCDA Approaches to HTA? 288
14.8 Final Issues to Consider 291
14.9 Conclusions 292
References 293
Chapter 15: Beyond Value Function Methods in MCDA for Health Care 296
15.1 Introduction 296
15.2 Alternative Value Function Methods 297
15.3 Non-value Function Methods for Health-Care Decision-Making 299
15.3.1 Goal and Reference Point Methods 300
15.3.2 Dominance-Based Approaches 301
15.3.3 Outranking Approaches 302
15.4 Concluding Remarks 304
References 305
Chapter 16: Best Practice for MCDA in Healthcare 308
16.1 Introduction 308
16.2 Decision Theory 310
16.3 Decision Analysis 311
16.4 Principles for Establishing the Decision Context 312
16.5 Principles for Identifying Options 314
16.6 Principles for Identifying Objectives and Criteria 314
16.7 Principles for Scoring the Options 316
16.8 Principles for Weighting the Criteria 319
16.9 Principles for Calculating Weighted Scores 321
16.10 Principles for Examining Results 322
16.11 Principles for Sensitivity Analyses and Scenario Analyses 323
16.12 Conclusion 324
References 325