Oberwolfach Reports


Full-Text PDF (1347 KB) | Introduction as PDF | Metadata | Table of Contents | OWR summary
Volume 1, Issue 4, 2004, pp. 2591–2652
DOI: 10.4171/OWR/2004/49

Published online: 2005-09-30

Design and Analysis of Infectious Disease Studies

Niels Becker[1], Klaus Dietz[2] and Niels Keiding[3]

(1) Australian National University, Canberra, Australia
(2) Universität Tübingen, Germany
(3) University of Copenhagen, Denmark

This workshop gathered 44 participants from 19 countries and represented a correspondingly multifaceted program concerning various diseases, public health issues and methodological innovations. The presentations and discussions highlighted again the crucial role that mathematical models and statistical analyses play in understanding the transmission of infectious diseases and in the development of strategies for their control. Mathematical transmission models and analyses are needed to assess potential control strategies (including determination of optimal strategies), to develop statistical analyses that allow for the dependence in data generated by transmission of infection and to keep track of aspects of the infection dynamics that one cannot observe in practice. Modeling approaches, data analyses and parameter inferences were applied to diseases like malaria, yellow fever, pneumococcal and meningococcal infections, hepatitis C and smallpox. Due to their current relevance, emerging infectious diseases and SARS attracted special attention and generated much discussion. These were complemented by talks on the public health issues of bioterrorism and interventions like vaccination and contact tracing and general aspects of control and eradicability of infections. While the research focus is motivated by real world applications, the need to accommodate complex population structures makes this area one requiring diverse and innovative mathematics and statistics. For example, the models draw on differential calculus, graph theory and multi-type stochastic processes with novel specifications of 'type'. The statistical analyses based on such models often require modern computer intensive methods with novel features, such as the use of random graphs for the transmission chains as latent variables. Furthermore, the methodological spectrum in this workshop comprised Bayesian approaches, computer simulations and the modelling of spatial structures and households. An evening discussion session on reproduction numbers focussed on the problems of its applicability, in particular with respect to infections in which density-dependent processes operate. The variety of topics made it necessary to organize a session for short talks and an evening with presentations where speakers presented current work on their individual laptops. This provided a very stimulating platform for collaborations and discussions because lecturers could interactively present the work on his/her specific computer environment to a specifically interested audience. This mode of presentation, which turned out to be a novelty at the Oberwolfach institute, could represent an appropriate alternative to poster presentations in future workshops. The workshop closed with a discussion session about future activities and open problems. On Thursday evening after a first class concert with piano recitals of Bach, Mozart and Chopin and Lieder of Schubert and Schumann an old Oberwolfach tradition was revived when participants from nearly all represented countries contributed to a most enjoyable variety program of poems and songs.

No keywords available for this article.

Becker Niels, Dietz Klaus, Keiding Niels: Design and Analysis of Infectious Disease Studies. Oberwolfach Rep. 1 (2004), 2591-2652. doi: 10.4171/OWR/2004/49