Temporal
and
spatio-temporal
monitoring
of
infectious
diseases
held at the Department of Statistics “G. Parenti” University of
Florence, Italy, Feb 9-10, 2011.
Course contents
Public health authorities have, in an attempt to meet the threats of
infectious diseases, created comprehensive mechanisms for the
collection of data on cases of infectious diseases. The vast amounts of
acquired data demand appropriate statistical methods describing the
dynamics and the development of algorithms for the automated detection
of abnormalities.
This short course covers statistical aspects of how to model and
monitor routine collected surveillance data which, depending on the
temporal and spatial scale, can be seen as realizations of the
following stochastic processes:
discrete space - discrete time process, i.e.
univariate or multivariate count data time series
discrete space - continuous time point processes
continuous space - continuous time point processes
With a strong emphasis on the simplest structure - the univariate count
data time series - the course presents methods for the retrospective
and prospective analysis. An implementational aspect of the methods is
given by applications using the R package 'surveillance'. The structure
of the short course will be as follows:
Motivating examples: Why is there an interest in the modelling
and monitoring of routine collected public health data.
Overview of temporal surveillance and introduction to the R
package 'surveillance'
Specific treatment of Shewhart methods and CUSUM based methods
for univariate surveillance.
Among others this includes: The
Farrington et al. (1996) procedure, Rossi et al. (1999), Höhle &
Paul (2008).
Comparing univariate surveillance methods: empirical
investigations and
theoretical considerations
Endemic-epidemic two component modelling: Three views in time and
space-time
Outlook: Towards multivariate surveillance - how to extend the
known
approaches?
The intended audience of the course are biostatisticians,
epidemiologists and master students of these directions. Prerequisites
are a knowledge of statistics up to a basic understanding of Poisson
regression models, the Poisson process and familiarity with R - a free
software environment
for statistical computing and graphics.
Speaker:
Michael
Höhle, Department for Infectious Disease Epidemiology, Robert Koch
Institute, Germany
Tutorials on the R package surveillance:
Annibale Biggeri, Dolores Catelan,
Emanuela Dreassi, Laura Grisotto, Department of Statistics “G. Parenti”
University of Florence, Italy
Practical information
Location:
Seminar Room 32, Department of
Statistics “G. Parenti”, Viale Morgagni 59, Florence
Dates:
Wednesday 9 Feb 2011, from 9:00 to 16:30
Thursday 10 Feb 2011: from9:00
to 16:30
Registration:
Contact Lucia Castellucci
(l.castellucci <AT> ispo.toscana.it) with your CV
Course material
A good way to get started with the area and the software is to read the
Höhle (2007) article in Computational Statistics and the Höhle, Paul
and Held (2009) work in Preventive Veterinary Medicine. More detailed
references follow below.
Surveillance package and univariate
monitoring
Höhle M, surveillance: An R package for
the
surveillance of infectious diseases. Computational Statistics, 2007,
22(4), pp. 571–582 [preprint]
Höhle M and Mazick A. Aberration
detection in R illustrated by Danish mortality monitoring. Book chapter
in T. Kass-Hout and X. Zhang (Eds.) Biosurveillance: Methods and Case
Studies, CRC Press, 2010; pp. 215-238. [preprint]