Population and familial relative risks of disease associated with environmental factors in the presence of gene-environment interaction.

With increasing availability of polymorphic DNA markers, family studies can be used to identify linked susceptibility genes for chronic diseases. Nevertheless, in the presence of gene-environment interaction, the ability of family studies to detect associations between exposures and diseases has not been adequately compared with population studies. The authors derive and compare exposure relative risk measures obtained from population studies with those obtained from studying siblings of affected individuals. For multiplicative joint effects on disease risk between the gene and the exposure, it can be shown that exposure relative risks in siblings are identical to those obtained from the population. For less than multiplicative joint effects (e.g., additive) sibling relative risks are less than population relative risks. However, in the presence of more than multiplicative joint effects, it can be shown that sibling relative risks could be much higher than population relative risks, more so for dominant susceptibility genes than recessive ones, and depend on the model of gene-environment interaction. Under these conditions of strong gene-environment interaction, family studies can be useful in searching for exposure-disease associations, in addition to detecting underlying genetic mechanisms.