Legionnaires' disease outbreak investigation toolbox

Download Page

2.3 Demographic Data

Demographic data are a useful addition to Legionnaires' disease outbreak investigation as it can be utilised to calculate attack rates and relative risk for a given population (often quoted as the number of cases per 100,000 population). Attack rates calculated using GIS(as discussed in section 1.2.2) can provide a relative measure of disease prevalence. For example, a specific number of cases reported in a relatively sparsely populated area could be more indicative of an outbreak than an identical number of cases reported in a densely populated area. Similarly as Legionnaires' disease tends to impact the elderly rather than the young, calculating age-specific attack rates can provide additional insight into a Legionnaires' disease outbreak.

Demographic data are available in different forms, scales and resolutions. Raster based demographic datasets divide an area into a gridded structure and assign a population count or density to each grid cell. Population data are also available attached to administrative boundaries such as Nomenclature of Units for Territorial Statistics (NUTS) and Local Administrative Units (LAU's). It is important to understand that demographic data can change with space and time (i.e. commuting and tourism) and mechanisms for understanding population changes in a locality are important for informing risk of exposure.

Considerations for cross-border outbreak: In a cross border outbreak scenario it is important to realise that countries collect demographic data in different ways and as such are not necessarily directly comparable with demographic data of neighbouring nations. Data may have been collected using different methodologies, reporting different denominator population counts, covering different time periods and be presented at different resolutions. Various datasets exist that attempt to combine demographic data across Europe. For example, EuroGeoGraphics EuroBoundaryMap, merges equivalent statistical geographies into a seamless European dataset, and there are grid-based pan-European population initiatives that exist such as the European Forum for Geostatistics "Representing Census data in a European population grid initiative ( http://www.earsc.eu/news/geostat-representing-census-data-in-a-european-population-grid) and also the European Commission Joint Research Centre and European Statistics Agencies hybrid 1km resolution Geostat Population map 2010 (http://www.efgs.info/grid-maps). The suitability of such datasets, particularly with reference to scale and precision, should be carefully considered.