Health-related decision making, whether at population or individual level, needs an understanding of how diseases spread and how interventions will impact on them. This often requires us to identify and combine many sources of information so that recommendations and subsequent decisions are based on a relevant and sound evidence base. For example, when the objective is to control an infectious disease in the community, we need to know how many people are affected by the disease (prevalence) and in which age groups; how fast it is currently spreading incidence); how it is transmitted (transmission and infectivity); and the geographical locations where the disease is more prevalent. This information feeds into relevant recommendations, such as on vaccination or treatment strategies, which also require an understanding of which subgroups of the population should be vaccinated, or for which patients a particular treatment is cost-effective. Evidence on each of these aspects typically comes from several studies, may be incomplete and biased, refers to populations different from those of interest, and/or is sparse. Robust statistical methods are then needed to integrate such multiplicity of evidence in a coherent manner to make them useful inputs to decision making.
Research in the “Evidence Synthesis to inform Health” theme aims to develop methods for the design, estimation and assessment of models for combining diverse sources of information in order to answer questions about the optimum management of patients and health resources. As well as developing methods that are general and can be used in a wide range of situations, we apply these methods to important topical questions about infectious diseases, addictions, dementia and health technologies.
Other Research Themes:
- DART: Design and Analysis of Randomised Trials
- SGX: Statistical Genomics
- COLD: Methods for the Analysis of Complex Observational and Longitudinal Data