Evidence synthesis in health economic evaluations and public health policy.
Health policies, such as the introduction of new treatments, are commonly evaluated using decision-analytic models. These models combine all relevant evidence from multiple sources, commonly extrapolate it to the long term, and involve a lot of uncertain assumptions. My work involves statistical methods to ensure that the decisions based on these models accurately reflect the available evidence and the extent of uncertainty. Methods for model comparison, flexible modelling and model calibration are used, typically within a Bayesian framework, which allows direct data, indirect data and expert belief to be combined.
Multi-state models for longitudinal data
Multi-state and time-to-event models in health care and disease progression. I maintain the msm R package for continuous-time Markov and hidden Markov modelling.
Developing and maintaining several R packages, including msm, flexsurv for survival analysis, denstrip for illustrating distributions, ecoreg for ecological inference. Contributing to the OpenBUGS and BRugs software for Bayesian analysis.
Calibration of complex models through Bayesian evidence synthesis: a demonstration and tutorial.
Medical Decision Making,
The BUGS Book: A Practical Introduction to Bayesian Analysis.
A framework for addressing structural uncertainty in decision models.
Medical Decision Making 31: (4), 662-674
Multi-State Models for Panel Data: The msm Package for R.
Journal of Statistical Software 38: (8)
Structural and parameter uncertainty in Bayesian cost-effectiveness models.
Journal of the Royal Statistical Society, Series C 59: (2), 233-253
Survival models in health economic evaluations: balancing fit and parsimony to improve prediction.
International Journal of Biostatistics 6: (1), Article 34