Ebook
Evidence Synthesis for Decision Making in HealthcareISBN: 9781118305409
320 pages
April 2012

In the evaluation of healthcare, rigorous methods of quantitative assessment are necessary to establish interventions that are both effective and costeffective. Usually a single study will not fully address these issues and it is desirable to synthesize evidence from multiple sources. This book aims to provide a practical guide to evidence synthesis for the purpose of decision making, starting with a simple single parameter model, where all studies estimate the same quantity (pairwise metaanalysis) and progressing to more complex multiparameter structures (including metaregression, mixed treatment comparisons, Markov models of disease progression, and epidemiology models). A comprehensive, coherent framework is adopted and estimated using Bayesian methods.
Key features:

Preface xi
1 Introduction 1
1.1 The rise of health economics 1
1.2 Decision making under uncertainty 4
1.2.1 Deterministic models 4
1.2.2 Probabilistic decision modelling 6
1.3 Evidencebased medicine 9
1.4 Bayesian statistics 10
1.5 NICE 11
1.6 Structure of the book 12
1.7 Summary key points 13
1.8 Further reading 13
References 14
2 Bayesian methods and WinBUGS 17
2.1 Introduction to Bayesian methods 17
2.1.1 What is a Bayesian approach? 17
2.1.2 Likelihood 18
2.1.3 Bayes’ theorem and Bayesian updating 19
2.1.4 Prior distributions 22
2.1.5 Summarising the posterior distribution 23
2.1.6 Prediction 24
2.1.7 More realistic and complex models 24
2.1.8 MCMC and Gibbs sampling 25
2.2 Introduction to WinBUGS 26
2.2.1 The BUGS language 26
2.2.2 Graphical representation 31
2.2.3 Running WinBUGS 32
2.2.4 Assessing convergence in WinBUGS 33
2.2.5 Statistical inference in WinBUGS 36
2.2.6 Practical aspects of using WinBUGS 39
2.3 Advantages and disadvantages of a Bayesian approach 39
2.4 Summary key points 40
2.5 Further reading 41
2.6 Exercises 41
References 42
3 Introduction to decision models 43
3.1 Introduction 43
3.2 Decision tree models 44
3.3 Model parameters 45
3.3.1 Effects of interventions 45
3.3.2 Quantities relating to the clinical epidemiology of the clinical condition being treated 50
3.3.3 Utilities 52
3.3.4 Resource use and costs 52
3.4 Deterministic decision tree 52
3.5 Stochastic decision tree 56
3.5.1 Presenting the results of stochastic economic decision models 60
3.6 Sources of evidence 66
3.7 Principles of synthesis for decision models (motivation for the rest of the book) 70
3.8 Summary key points 70
3.9 Further reading 71
3.10 Exercises 71
References 72
4 Metaanalysis using Bayesian methods 76
4.1 Introduction 76
4.2 Fixed Effect model 78
4.3 Random Effects model 81
4.3.1 The predictive distribution 83
4.3.2 Prior specification for τ 84
4.3.3 ‘Exact’ Random Effects model for Odds Ratios based on a Binomial likelihood 84
4.3.4 Shrunken study level estimates 86
4.4 Publication bias 87
4.5 Study validity 88
4.6 Summary key points 88
4.7 Further reading 88
4.8 Exercises 89
References 92
5 Exploring between study heterogeneity 94
5.1 Introduction 94
5.2 Random effects metaregression models 95
5.2.1 Generic random effect metaregression model 95
5.2.2 Random effects metaregression model for Odds Ratio (OR) outcomes using a Binomial likelihood 98
5.2.3 Autocorrelation and centring covariates 100
5.3 Limitations of metaregression 104
5.4 Baseline risk 105
5.4.1 Model for including baseline risk in a metaregression on the (log) OR scale 107
5.4.2 Final comments on including baseline risk as a covariate 109
5.5 Summary key points 110
5.6 Further reading 110
5.7 Exercises 110
References 113
6 Model critique and evidence consistency in random effects metaanalysis 115
6.1 Introduction 115
6.2 The Random Effects model revisited 117
6.3 Assessing model fit 121
6.3.1 Deviance 121
6.3.2 Residual deviance 122
6.4 Model comparison 124
6.4.1 Effective number of parameters, pD 125
6.4.2 Deviance Information Criteria 126
6.5 Exploring inconsistency 127
6.5.1 Crossvalidation 128
6.5.2 Mixed predictive checks 131
6.6 Summary key points 134
6.7 Further reading 134
6.8 Exercises 134
References 137
7 Evidence synthesis in a decision modelling framework 138
7.1 Introduction 138
7.2 Evaluation of decision models: Onestage vs twostage approach 139
7.3 Sensitivity analyses (of model inputs and model specifications) 147
7.4 Summary key points 147
7.5 Further reading 147
7.6 Exercises 147
References 149
8 Multiparameter evidence synthesis 151
8.1 Introduction 151
8.2 Prior and posterior simulation in a probabilistic model: Maple Syrup Urine Disease (MSUD) 152
8.3 A model for prenatal HIV testing 155
8.4 Model criticism in multiparameter models 161
8.5 Evidencebased policy 163
8.6 Summary key points 164
8.7 Further reading 165
8.8 Exercises 166
References 167
9 Mixed and indirect treatment comparisons 169
9.1 Why go beyond ‘direct’ headtohead trials? 169
9.2 A fixed treatment effects model for MTC 172
9.2.1 Absolute treatment effects 176
9.2.2 Relative treatment efficacy and ranking 176
9.3 Random Effects MTC models 178
9.4 Model choice and consistency of MTC evidence 179
9.4.1 Techniques for presenting and understanding the results of MTC 180
9.5 Multiarm trials 181
9.6 Assumptions made in mixed treatment comparisons 182
9.7 Embedding an MTC within a costeffectiveness analysis 183
9.8 Extension to continuous, rate and other outcomes 185
9.9 Summary key points 187
9.10 Further reading 187
9.11 Exercises 189
References 190
10 Markov models 193
10.1 Introduction 193
10.2 Continuous and discrete time Markov models 195
10.3 Decision analysis with Markov models 196
10.3.1 Evaluating Markov models 197
10.4 Estimating transition parameters from a single study 199
10.4.1 Likelihood 202
10.4.2 Priors and posteriors for multinomial probabilities 202
10.5 Propagating uncertainty in Markov parameters into a decision model 206
10.6 Estimating transition parameters from a synthesis of several studies 209
10.6.1 Challenges for metaanalysis of evidence on Markov transition parameters 209
10.6.2 The relationship between probabilities and rates 211
10.6.3 Modelling study effects 213
10.6.4 Synthesis of studies reporting aggregate data 215
10.6.5 Incorporating studies that provide event history data 217
10.6.6 Reporting results from a Random Effects model 219
10.6.7 Incorporating treatment effects 220
10.7 Summary key points 224
10.8 Further reading 224
10.9 Exercises 224
References 225
11 Generalised evidence synthesis 227
11.1 Introduction 227
11.2 Deriving a prior distribution from observational evidence 230
11.3 Bias allowance model for the observational data 233
11.4 Hierarchical models for evidence from different study designs 238
11.5 Discussion 244
11.6 Summary key points 244
11.7 Further reading 245
11.8 Exercises 246
References 248
12 Expected value of information for research prioritisation and study design 251
12.1 Introduction 251
12.2 Expected value of perfect information 256
12.3 Expected value of partial perfect information 259
12.3.1 Computation 261
12.3.2 Notes on EVPPI 264
12.4 Expected value of sample information 264
12.4.1 Computation 265
12.5 Expected net benefit of sampling 266
12.6 Summary key points 267
12.7 Further reading 268
12.8 Exercises 268
References 268
Appendix 1 Abbreviations 270
Appendix 2 Common distributions 272
A2.1 The Normal distribution 272
A2.2 The Binomial distribution 273
A2.3 The Multinomial distribution 273
A2.4 The Uniform distribution 274
A2.5 The Exponential distribution 274
A2.6 The Gamma distribution 275
A2.7 The Beta distribution 276
A2.8 The Dirichlet distribution 277
Index 278