In studies of past, present, and likely future disease distributions, the ‘added value’ provided by the geographer lies in three main areas: detecting spatial concentrations of disease; isolating the processes (environmental, social, demographic, and pathogenic) which cause these disease hotspots; and in enhancing our understanding of the space–time dynamics of disease spread. This is as true of war-related epidemics as of any others. Within geography, there is a long-standing tradition of mapping disease. In this early history, the incidence maps of yellow fever produced in 1798 are often given pride of place (Robinson, 1982). These were, however, pre-dated by maps of topics as diverse as hospital capacities and the distribution of dressing-stations on a battlefield, through to maps of pestilential swamps and other hostile medical environments. But, so far as most epidemiological reports were concerned, such maps were usually incidental. The breakthrough in disease mapping occurred in the middle of the nineteenth century with the cholera map produced by Dr John Snow to accompany the second edition of his prize-winning essay On the Mode of Communication of Cholera (1855a). What set Snow’s work apart was not the cartography (dot maps, which were a well-established cartographic device, to show the geographical distribution of individual cholera deaths), but his inductive reasoning from the map. By showing what he termed the ‘topography of the outbreak’, Snow was able to draw inferences about the central source of infection. The use of mapping as an important device for suggesting hypotheses of medical interest may be traced through to the present day. For war and disease, the classic example is the Seuchen Atlas. This atlas of epidemic disease (Zeiss, 1942–5; Anderson, 1947) was conceived by the German army as an adjunct to war, enhancing its ability to mount military campaigns. The atlas was produced as separate sheets over the years 1942–5. Its distribution was confined to military institutes and to those German university institutes involved in training medical students. The scope of the atlas was not global but confined largely to those areas where the Army High Command expected to be fighting.
Household members do not contact each other at random: implications for infectious disease modelling