Comparison of cancer incidence among similar geographic areas: a better source of etiologic hypotheses?

Difference in incidence rates of cancer of specific sites among different populations has been a powerful tool for the generation of etiologic hypotheses. The use of strictly comparable data from cancer registries, on populations from Italy, France, Switzerland and Spain, similar for geographic location, race and ancestry, shows unexpected differences of rates of single cancer sites, being significantly higher or lower than those of the same subgroup of registries. The study of risk factors associated with such differences might be more profitable than those carried out on the basis of comparisons between distantly located populations.