Proximity of the home to a cooling tower and risk of non-outbreak Legionnaires' disease.

OBJECTIVE To study the source of non-outbreak legionnaires' disease, particularly the role of cooling towers, by comparing the locations of patients' homes in relation to the location of cooling towers. DESIGN Retrospective, descriptive study of a case series of patients with legionnaires' disease ill between 1978 and 1986 and, for comparison, a case series of patients with lung cancer. A prospectively developed register and interview based survey provided data on the location of cooling towers. SETTING The city of Glasgow. PATIENTS 134 patients aged 14-84 with legionnaires' disease during 1978-86 and 10,159 patients with lung cancer during the same period. MAIN OUTCOME MEASURES The locations of patients' homes and cooling towers as defined by postcodes, which provided map grid references accurate to 10 m; numbers of expected and observed cases of legionnaires' disease in census enumeration districts; and distance of enumeration districts from the nearest cooling tower as defined by five distance categories. RESULTS Most cooling towers were in or near the city centre or close to the River Clyde, as were the places of residence of patients with community acquired, non-travel, non-outbreak legionnaires' disease (n = 107). There was an inverse association between the distance of residence from any cooling tower and the risk of infection, the population living within 0.5 km of any tower having a relative risk of infection over three times that of people living more than 1 km away. There was no such association with respect to travel related legionnaires' disease, and for lung cancer the association was weak (relative risk less than or equal to 1.2 in any distance group). CONCLUSION In Glasgow cooling towers have been a source of infection in two outbreaks of legionnaires' disease and, apparently, a source of non-outbreak infection also. Better maintenance of cooling towers should help prevent non-outbreak cases. This method of inquiry should be applied elsewhere to study the source of this and other environmentally acquired disease.