Bayesian Evidence Synthesis for Estimating Infectious Disease Burden

• Activity Number: 619
• Type: Invited
• Date/Time: Thursday, August 13, 2015 : 8:30 AM to 10:20 AM
• Sponsor: IMS
• Author(s): Anne Presanis*
• Companies: MRC Biostatistics Unit
• Keywords: Bayesian ; Evidence Synthesis ; Public Health ; Infectious Disease ; Estimation ; Graphical models

Abstract

Estimation of infectious disease burden characteristics such as prevalence, incidence or severity, is challenging, as they are typically not easy to measure directly. However, quantification of different aspects of the burden is vital for public health, to inform both planning of health-care services and policies to prevent transmission. Estimation of the burden may, nevertheless, be feasible from a network of information from multiple and varied sources, combining data using evidence synthesis methods. The available data sources relevant to the disease under study may include, for example, surveillance systems, observational studies, registries and community surveys. Such data may be incomplete and/or biased, therefore only indirectly informing the quantities to be estimated. To synthesise such diverse and challenging data usually implies the formulation of complex probabilistic models, often in a Bayesian framework. In the context of such complexity, critical model assessment is essential. Concepts such as inconsistency and identifiability in evidence synthesis will be illustrated through examples such as the estimation of influenza severity and of HIV prevalence and incidence.