The basic premise of a cancer-cluster litigation is that there is more than an expected number of cancer cases of a certain type in an area and that the cancers — or at least those of the plaintiffs in the litigation — were caused by a common exposure, such as the release of a chemical carcinogen from a nearby industrial facility.

Statisticians have a field day with such allegations, as it can be shown that, even in the absence of any unusual environmental exposure, there is an 87 percent probability that a census tract will have a statistically significant elevated cancer rate for at least one of 80 types of cancers.[FOOTNOTE 1] Even those scientists who emphasize the role of environmental exposures as a causative element in the incidence of cancer concede that the “data requirements for traditional epidemiological studies are rarely met” in the analysis of disease clusters.[FOOTNOTE 2]