The central purpose of the Outbreak Analysis Unit will be to understand the likely pattern of epidemic spread following the emergence of novel communicable pathogens into the human population. In the initial phase of a disease outbreak, when the number of infective people is low, the impact of individuals’ circumstances, characteristics and behaviours in the vicinity of the focus of the outbreak can have a significant influence on the resulting outcome of the epidemic. An example would be the ‘superspreader’ phenomenon seen in the SARS outbreak, and recently identified as important for many other infections.

In these circumstances it has been found that stochastic individual-based approaches and network methods can be particularly useful. Of particular relevance will be the development of techniques that can rapidly characterise the socio-spatial structure of a population in which a novel pathogen has begun to circulate, and in turn use that knowledge to underpin epidemic models. These models represent individual hosts and their inter-relationships as the substrate upon which a given pathogen epidemiology plays itself out.

Specifically we will develop methods to characterise human social network structures at multiple scales. This is essential in order to be able to apply the appropriate level of modelling depending on the scale of the outbreak. As well as identifying the predominant patterns of social interaction the effect of transport networks will also be included, and there will be a strong interaction with the “characterising human contact/travel patterns” research theme. Models of communicable disease spread will be developed and overlaid on these network structures.

Being able to accurately capture the specific pattern of early phase outbreaks using network and/or individual-based methods will assist in giving information on the pathogen virulence and its propensity (or otherwise) for strain evolution within human hosts. The ability to undertake quantitative modelling at the very earliest phases of an outbreak is essential to developing intervention strategies that may be targeted at likely superspreaders or aim to disrupt highly connected network components.