Background The source of infection for most sporadic cases of Legionnaires' disease remains unknown. This study aims quantify the relationship between cases and wet cooling systems (WCS), a potential source of aerosolised legionella bacteria.

Methods The study analysed data on 1163 sporadic, community-acquired cases of Legionnaires' disease in England and Wales with onset between 1996 and 2006, and 11630 postcode controls randomly sampled in proportion to population size and matched on region, age group and sex. The relationship between risk of Legionnaires' disease and distance from a WCS was analysed by conditional logistic regression.

Results Cases and controls had a mean age of 56.3 years; 79.3% were male. Cases lived appreciably closer to WCS than their controls (mean distance of cases=2.11 km, controls=2.58 km; mean difference 0.47 km (95% CI 0.28 to 0.65)). The OR for disease within 1 km of a WCS compared with over 6 km (a distance taken to reflect background rates of Legionnaires' disease) was 1.59 (95% CI 1.26 to 2.01) when adjusted for socio-economic deprivation, and 1.33 (95% CI 1.04 to 1.71) when additionally adjusted for population density. The results suggest that residential proximity to a WCS may account for 19.6% of sporadic community-acquired cases.

Conclusions WCS may be an important source of sporadic, community-acquired cases of Legionnaires' disease, an observation that has important implications for health protection, especially given the likely increase in such systems as a component of strategies to improve energy efficiency in buildings.