Evidence Synthesis for Decision Making in Healthcare -- Contents -- Preface -- Chapter 1 Introduction -- 1.1 The rise of health economics -- 1.2 Decision making under uncertainty -- 1.2.1 Deterministic models -- 1.2.2 Probabilistic decision modelling -- 1.3 Evidence-based medicine -- 1.4 Bayesian statistics -- 1.5 NICE -- 1.6 Structure of the book -- 1.7 Summary key points -- 1.8 Further reading -- References -- Chapter 2 Bayesian methods and WinBUGS -- 2.1 Introduction to Bayesian methods -- 2.1.1 What is a Bayesian approach? -- 2.1.2 Likelihood -- 2.1.3 Bayes' theorem and Bayesian updating -- 2.1.4 Prior distributions -- 2.1.5 Summarising the posterior distribution -- 2.1.6 Prediction -- 2.1.7 More realistic and complex models -- 2.1.8 MCMC and Gibbs sampling -- 2.2 Introduction to WinBUGS -- 2.2.1 The BUGS language -- 2.2.2 Graphical representation -- 2.2.3 Running WinBUGS -- 2.2.4 Assessing convergence in WinBUGS -- 2.2.5 Statistical inference in WinBUGS -- 2.2.6 Practical aspects of using WinBUGS -- 2.3 Advantages and disadvantages of a Bayesian approach -- 2.4 Summary key points -- 2.5 Further reading -- 2.6 Exercises -- References -- Chapter 3 Introduction to decision models -- 3.1 Introduction -- 3.2 Decision tree models -- 3.3 Model parameters -- 3.3.1 Effects of interventions -- 3.3.2 Quantities relating to the clinical epidemiology of the clinical condition being treated -- 3.3.3 Utilities -- 3.3.4 Resource use and costs -- 3.4 Deterministic decision tree -- 3.5 Stochastic decision tree -- 3.5.1 Presenting the results of stochastic economic decision models -- 3.6 Sources of evidence -- 3.7 Principles of synthesis for decision models (motivation for the rest of the book) -- 3.8 Summary key points -- 3.9 Further reading -- 3.10 Exercises -- References.

Chapter 4 Meta-analysis using Bayesian methods -- 4.1 Introduction -- 4.2 Fixed Effect model -- 4.3 Random Effects model -- 4.3.1 The predictive distribution -- 4.3.2 Prior specification for τ -- 4.3.3 `Exact' Random Effects model for Odds Ratios based on a Binomial likelihood -- 4.3.4 Shrunken study level estimates -- 4.4 Publication bias -- 4.5 Study validity -- 4.6 Summary key points -- 4.7 Further reading -- 4.8 Exercises -- References -- Chapter 5 Exploring between study heterogeneity -- 5.1 Introduction -- 5.2 Random effects meta-regression models -- 5.2.1 Generic random effect meta-regression model -- 5.2.2 Random effects meta-regression model for Odds Ratio (OR) outcomes using a Binomial likelihood -- 5.2.3 Autocorrelation and centring covariates -- 5.3 Limitations of meta-regression -- 5.4 Baseline risk -- 5.4.1 Model for including baseline risk in a meta-regression on the (log) OR scale -- 5.4.2 Final comments on including baseline risk as a covariate -- 5.5 Summary key points -- 5.6 Further reading -- 5.7 Exercises -- References -- Chapter 6 Model critique and evidence consistency in random effects meta-analysis -- 6.1 Introduction -- 6.2 The Random Effects model revisited -- 6.3 Assessing model fit -- 6.3.1 Deviance -- 6.3.2 Residual deviance -- 6.4 Model comparison -- 6.4.1 Effective number of parameters, pD -- 6.4.2 Deviance Information Criteria -- 6.5 Exploring inconsistency -- 6.5.1 Cross-validation -- 6.5.2 Mixed predictive checks -- 6.6 Summary key points -- 6.7 Further reading -- 6.8 Exercises -- References -- Chapter 7 Evidence synthesis in a decision modelling framework -- 7.1 Introduction -- 7.2 Evaluation of decision models: One-stage vs two-stage approach -- 7.3 Sensitivity analyses (of model inputs and model specifications) -- 7.4 Summary key points -- 7.5 Further reading -- 7.6 Exercises -- References.

Chapter 8 Multi-parameter evidence synthesis -- 8.1 Introduction -- 8.2 Prior and posterior simulation in a probabilistic model: Maple Syrup Urine Disease (MSUD) -- 8.3 A model for prenatal HIV testing -- 8.4 Model criticism in multi-parameter models -- 8.5 Evidence-based policy -- 8.6 Summary key points -- 8.7 Further reading -- 8.8 Exercises -- References -- Chapter 9 Mixed and indirect treatment comparisons -- 9.1 Why go beyond `direct' head-to-head trials? -- 9.2 A fixed treatment effects model for MTC -- 9.2.1 Absolute treatment effects -- 9.2.2 Relative treatment efficacy and ranking -- 9.3 Random Effects MTC models -- 9.4 Model choice and consistency of MTC evidence -- 9.4.1 Techniques for presenting and understanding the results of MTC -- 9.5 Multi-arm trials -- 9.6 Assumptions made in mixed treatment comparisons -- 9.7 Embedding an MTC within a cost-effectiveness analysis -- 9.8 Extension to continuous, rate and other outcomes -- 9.9 Summary key points -- 9.10 Further reading -- 9.11 Exercises -- References -- Chapter 10 Markov models -- 10.1 Introduction -- 10.2 Continuous and discrete time Markov models -- 10.3 Decision analysis with Markov models -- 10.3.1 Evaluating Markov models -- 10.4 Estimating transition parameters from a single study -- 10.4.1 Likelihood -- 10.4.2 Priors and posteriors for multinomial probabilities -- 10.5 Propagating uncertainty in Markov parameters into a decision model -- 10.6 Estimating transition parameters from a synthesis of several studies -- 10.6.1 Challenges for meta-analysis of evidence on Markov transition parameters -- 10.6.2 The relationship between probabilities and rates -- 10.6.3 Modelling study effects -- 10.6.4 Synthesis of studies reporting aggregate data -- 10.6.5 Incorporating studies that provide event history data.

10.6.6 Reporting results from a Random Effects model -- 10.6.7 Incorporating treatment effects -- 10.7 Summary key points -- 10.8 Further reading -- 10.9 Exercises -- References -- Chapter 11 Generalised evidence synthesis -- 11.1 Introduction -- 11.2 Deriving a prior distribution from observational evidence -- 11.3 Bias allowance model for the observational data -- 11.4 Hierarchical models for evidence from different study designs -- 11.5 Discussion -- 11.6 Summary key points -- 11.7 Further reading -- 11.8 Exercises -- References -- Chapter 12 Expected value of information for research prioritisation and study design -- 12.1 Introduction -- 12.2 Expected value of perfect information -- 12.3 Expected value of partial perfect information -- 12.3.1 Computation -- 12.3.2 Notes on EVPPI -- 12.4 Expected value of sample information -- 12.4.1 Computation -- 12.5 Expected net benefit of sampling -- 12.6 Summary key points -- 12.7 Further reading -- 12.8 Exercises -- References -- Appendix 1 Abbreviations -- Appendix 2 Common distributions -- A2.1 The Normal distribution -- A2.2 The Binomial distribution -- A2.3 The Multinomial distribution -- A2.4 The Uniform distribution -- A2.5 The Exponential distribution -- A2.6 The Gamma distribution -- A2.7 The Beta distribution -- A2.8 The Dirichlet distribution -- Index.