[Childhood cancer epidemiology and available evidence on case clustering]

This paper introduces a Monograph on childhood cancer clusters addressed to Italian public health workers, paediatricians and paediatric oncologists, and the general public. Two issues have been underlined in most definitions of cancer clusters reported in the literature over the last 30 years. The first is the need for an unbiased methodological approach for the detection and the interpretation of clustering and of individual clusters. The other is the responsibility for scientists and public health workers to unravel suspicious events and to interact with the population in circumstances which may turn out to be false alarms. The relative weight given to epidemiological methods and to risk perception has varied between definitions of clusters given in different periods. In the field of childhood cancer clusters, epidemiological research has produced very little fresh knowledge. However, it is recognized that this is not a sufficient reason for refusing to pay attention to reports of perceived clusters. Models of spatio-temporal interactions according to Birch et al. (2000) are taken in consideration: the literature offers a very limited number of circumstances fitting such models. In Italy, over the years, only two childhood cancer clusters have been reported in the indexed literature, both of them regarded cases of acute lymphatic leukaemia and occurred in Sardinia in the Eighties and in the area of Roma in the Nineties. A possible reason for the paucity of reports (compared, for instance, to the UK scenario) is the limited availability of health statistics.