With increased availability of multi-generation cancer data, most studies to date on family cancer risks have been measuring the incidence of various cancers within families and found higher family
risks for breast, prostate, ovarian and uterine cancer. L ess is known about the underlying causes of increased family cancer risks familiality). Cancer may arise because of both pre-birth factors
(genetics and prenatal environment) and post-birth factors (non-genetic endowments and behavioral and environmental factors), and intra-family correlation of cancer can therefore be caused by shared genes and shared environments. Empirical studies that try to separate hereditary and environmental factors as distinctive causes for cancer are relatively scarce. To our kn owledge the existing work draws on cancer registry data from Sweden, Finland and Denmark. This work by epidemiologists and bio-scientists exploits differences in correlations between phenotypes (observed cancer risks) of different kin pairs (including mon o zygotic
twins, zygotic twins, full siblings and half siblings). For most cancer sites these studies find that although genetic factors are important, environmental factors dominate. Medical researchers are predominantly interested in the genetic compone nt of cancer risk and treat the environmental component as residual. Moreover, existing studies that study the relationship between demographics and cancer do not attempt to separate hereditary and nvironmental components, and (twin) studies that do separ ate these components do not consider socio-economic factors in the
analysis but instead present uncorrected concordance rates and variance decompositions. The main contribution of this project therefore is to measure the extent environmental factors that drive various cancer risks can be attributed to socio-economic background and conditions experienced during
infancy, early childhood and adolescence.