STATISTICAL MODELS FOR INFECTIOUS DISEASE
A half-day meeting organised by the General Applications Section of the
Royal Statistical Society.
Tuesday 10th June, 2.00pm (Tea 3.30pm)
The Royal Statistical Society, Errol Street, London
(Directions: http://www.rss.org.uk/about/direction.html)
Free meeting, all welcome!
14:00 "Bayesian inference and model selection for spatial epidemics"
PETE NEAL* and GARETH ROBERTS (University of Lancaster)
A stochastic epidemic model is proposed which incorporates heterogeneity
in the spread of a disease through a population. In particular, three
factors are considered; the spatial location of an individual's home and
household and school class to which the individual belongs. The model is
applied to an extremely informative measles data set and the model is
compared with three nested models, each of which incorporates two of the
three factors. A reversible jump MCMC algorithm is then introduced which
assists in selecting the most appropriate model to fit the data.
14:30 "Time series modelling of disease transmission"
BARBEL FINKENSTADT* and ALEXANDER NORTON (University of Warwick)
A class of stochastic compartmental models is suggested that allows to
draw inference from time series of reported cases or, if available, from
spatio-temporal data. We demonstrate its use to predict (1) the spatial
and temporal extinction and recurrence of measles epidemics, and (2) the
epidemic size of influenza.
15:00 "Parameter estimation for within-host transmission of lymphatic
filariasis"
STEVE RILEY* (Imperial College London)
A discrete-time dynamic process is given as a model for a simple
within-host response to filarial infection. Parameters of the model are
estimated using Markov chain Monte Carlo (MCMC) methods and data from a
laboratory model of infection. The implied immune response seems to be
capable of reproducing the characteristic aggregation of worm burden
across a population of independent hosts. The general approach used to
construct the MCMC sampler could be adapted for other infectious disease
systems.
15:30 Tea
16:00 "Bayesian inference for epidemics in structured populations"
PHIL O'NEILL* (University of Nottingham)
Recently, there has been considerable interest in stochastic epidemic
models incorporating structured populations (eg communities of households;
classrooms in a school etc). Performing inference for such models is
often complicated by intractable likelihoods. We consider ways for
overcoming these problems, and illustrate the methods in various
applications.
16:30 "Using Bayesian MCMC to model the spatiotemporal dynamics of
foot-and-mouth disease"
ROB DEARDON* and STEVE BROOKS (University of Cambridge)
The 2001 UK foot-and-mouth epidemic exerted a major toll on the livestock
industry, and highlighted a series of important questions about the
management of infectious diseases in animal populations. In this talk we
will discuss some recent analyses of the DEFRA data and how Bayesian
approaches provide both a more flexible modelling framework and a means of
incorporating the wide variety of expert prior information available.
17:00 Close of meeting
The speakers are starred*.
-------------------------------------------------
Clare Marshall
([log in to unmask])
|