[Investigation of regional clusters of environmentally induced cancers]

The environment is source of carcinogen effects, which cannot be monitored as precisely as it would be required. Due to this fact, it is worth to screen for areas with higher than expected number of cancers that is for clusters. The significance of cluster suspicion is highly variable and the investigations for clusters could need significant resources. Therefore step-wise protocols are recommended, which evaluate before proceedings the possibility of exclusion of cluster existence, or of requirement for further epidemiological investigations. Sometimes, the results establish actions to reorganise the environmental control.

The relationship between cancer incidence and dangerous waste disposal sites was investigated in Tolna county (Hungary) and the usefulness of cluster studies was demonstrated by the results.

The incidence data based on histological investigations and the location of 7 dangerous waste disposal sites were analysed by geographical information system.

The incidences were not elevated around 6 sites. The cancer risk seemed to be high by site in settlement S., because of high standardised incidence ratio (SIH=1.41) and empirical Bayes adjusted SIH (SIHEB=1.38). The risk increase proved to be significant in z-test and mid-p test by 10% and 15% type I error. Since the risks showed nonhomogeneous spatial distribution in the county and the number of high-risk settlements was 2.3 to 6.6, the cluster in S. cannot be rejected as false positive observation. The chromium contaminated wastes have been stored in S. for several decades at river-side. Assuming that the exposure was spred by the river and the villages in the 5-km vicinity of the river were exposed, the SIHs were aggregated for every 15-km intervals. The distance from S. was inversely related to the aggregated SIHs.

The sites proved to be noncarcinogenic sources apart from the site S. for which the results suggested the high-risk status. The environmental pollution by site in S. could explain the increased incidence. Consequently, additional studies are indicated in S. to improve the reliability of cluster evaluation. The study also demonstrated that the cluster investigation can be inserted into public health practise to improve the efficiency of cancer control.