Bayesian Methods in Health Economics (Chapman & Hall/CRC Biostatistics Series Book 53)

Health economics is concerned with the study of the cost-effectiveness of health care interventions. This book provides an overview of Bayesian methods for the analysis of health economic data. After an introduction to the basic economic concepts and methods of evaluation, it presents Bayesian statistics using accessible mathematics. The next chapters describe the theory and practice of cost-effectiveness analysis from a statistical viewpoint, and Bayesian computation, notably MCMC. The final chapter presents three detailed case studies covering cost-effectiveness analyses using individual data from clinical trials, evidence synthesis and hierarchical models and Markov models. The text uses WinBUGS and JAGS with datasets and code available online. Content: Machine generated contents note: 1.Introduction to health economic evaluation -- 1.1.Introduction -- 1.2.Health economic evaluation -- 1.2.1.Clinical trials versus decision-analytical models -- 1.3.Cost components -- 1.3.1.Perspective and what costs include -- 1.3.2.Sources and types of cost data -- 1.4.Outcomes -- 1.4.1.Condition specific outcomes -- 1.4.2.Generic outcomes -- 1.4.3.Valuing outcomes -- 1.5.Discounting -- 1.6.Types of economic evaluations -- 1.6.1.Cost-minimisation analysis -- 1.6.2.Cost-benefit analysis -- 1.6.3.Cost-effectiveness analysis -- 1.6.4.Cost-utility analysis -- 1.7.Comparing health interventions -- 1.7.1.The cost-effectiveness plane -- 2.Introduction to Bayesian inference -- 2.1.Introduction -- 2.2.Subjective probability and Bayes theorem -- 2.2.1.Probability as a measure of uncertainty against a standard -- 2.2.2.Fundamental rules of probability -- 2.2.3.Coherence -- 2.2.4.Bayes theorem -- 2.3.Bayesian (parametric) modelling -- 2.3.1.Exchangeability and predictive inference -- 2.3.2.Inference on the posterior distribution -- 2.4.Choosing prior distributions and Bayesian computation -- 2.4.1.Vague priors -- 2.4.2.Conjugate priors -- 2.4.3.Monte Carlo estimation -- 2.4.4.Nonconjugate priors -- 2.4.5.Markov Chain Monte Carlo methods -- 2.4.6.MCMC convergence -- 2.4.7.MCMC autocorrelation -- 3.Statistical cost-effectiveness analysis -- 3.1.Introduction -- 3.2.Decision theory and expected utility -- 3.2.1.Problem -- 3.2.2.Decision criterion: Maximisation of the expected utility -- 3.3.Decision-making in health economics -- 3.3.1.Statistical framework -- 3.3.2.Decision process -- 3.3.3.Choosing a utility function: The net benefit -- 3.3.4.Uncertainty in the decision process -- 3.4.Probabilistic sensitivity analysis to parameter uncertainty -- 3.5.Reporting the results of probabilistic sensitivity analysis -- 3.5.1.Cost-effectiveness acceptability curves -- 3.5.2.The value of information -- 3.5.3.The value of partial information -- 3.6.Probabilistic sensitivity analysis to structural uncertainty -- 3.7.Advanced issues in cost-effectiveness analysis -- 3.7.1.Including a risk aversion parameter in the net benefit -- 3.7.2.Expected value of information for mixed strategies -- 4.Bayesian analysis in practice -- 4.1.Introduction -- 4.2.Software configuration -- 4.3.An example of analysis in JAGS/BUGS -- 4.3.1.Model specification -- 4.3.2.Pre-processing in R -- 4.3.3.Launching JAGS from R -- 4.3.4.Checking convergence and post-processing in R -- 4.4.Logical nodes -- 4.5.For loops and node transformations -- 4.5.1.Blocking to improve convergence -- 4.6.Predictive distributions -- 4.6.1.Predictive distributions as missing values -- 4.7.Modelling the cost-effectiveness of a new chemotherapy drug in R/JAGS -- 4.7.1.Programming the analysis of the EVPPI -- 4.7.2.Programming probabilistic sensitivity analysis to structural uncertainty -- 5.Health economic evaluation in practice -- 5.1.Introduction -- 5.2.Cost-effectiveness analysis alongside clinical trials -- 5.2.1.Example: RCT of acupuncture for chronic headache in primary care -- 5.2.2.Model description -- 5.2.3.JAGS implementation -- 5.2.4.Cost-effectiveness analysis -- 5.2.5.Alternative specifications of the model -- 5.3.Evidence synthesis and hierarchical models -- 5.3.1.Example: Neuraminidase inhibitors to reduce influenza in healthy adults -- 5.3.2.Model description -- 5.3.3.JAGS implementation -- 5.3.4.Cost-effectiveness analysis -- 5.4.Markov models -- 5.4.1.Example: Markov model for the treatment of asthma -- 5.4.2.Model description -- 5.4.3.JAGS implementation -- 5.4.4.Cost-effectiveness analysis -- 5.4.5.Adding memory to Markov models -- 5.4.6.Indirect estimation of the transition probabilities. Health economics is a relatively new discipline, characterised by the integration of different expertise and perspectives. Clearly, the clinical aspect is fundamental and the clinical background plays a basic role in the definition of any health economic evaluation. However, in its modern incarnation, health economics is effectively identified by the integration of economic models and increasingly advanced statistical techniques, particularly under the Bayesian approach