Understanding the epidemiology of directly transmitted pathogens such as influenza requires an understanding of the nature, frequency and social structuring of contacts between individuals which could lead to disease transmission. At a local (i.e. household/neighbourhood) scale, the contact patterns of populations determine the proportion infected during an epidemic and their demographic characteristics. At a larger scale (regional/national/international), contacts between travellers and a local population determine the rate at which a new outbreak spreads geographically. Understanding the contact rates at both scales is the priority of this research cluster.

Diary-based methods are revealing, for the first time, the underlying structure and dynamics of local epidemiological contact networks. These methods can compliment movement-based measures of contact (such as travel statistics) allowing estimation of the number and intensity of contacts in different social settings (work, home, school, etc), greatly facilitating accurate model development and parameterisation. They can also be used to track changes in contact patterns that might be expected to occur during an outbreak. Similar methods have also been used, in combination with bed occupancy and bacterial swabbing data, to model the spread of pathogens within the hospital environment – an important area for modelling, as hospitals often play a key role in amplifying the spread of new or re-emerging pathogens, such as SARS and smallpox.

It is a truism to say that human travel determines how pandemics spread globally, but unravelling the precise relationship between human travel patterns at a range of scales (travel to school, work, long distance national and international journeys) requires careful collation of detailed travel data, its statistical characterisation and mathematical description. Furthermore, validation that incorporating detailed human travel patterns into epidemiological models improves their representation of geographic disease spread is still lacking in many cases. This research cluster will continue to collate a range of travel data (air and rail traffic, survey data, census travel to work datasets, mobile phone and diary data) and investigate how travel behaviour can be minimally parameterised (e.g. using kernel models) while still capturing observed patterns of disease spread.