Inference of the strength of genotype-disease association from studies comparing offspring with and without parental history of disease.

The association between genetic polymorphisms and a multifactorial disease is generally investigated by case-control studies. However, inference about the genotype-disease association can also be drawn from studies comparing offspring having a parental history of disease with offspring having no parental history. In such studies, differences in genotype frequencies between the two groups of offspring will reflect a different transmission of alleles from affected and unaffected parents. We showed that in offspring studies, the odds ratios (ORs) associated with heterozygous and homozygous genotypes are related by the formula: OR(het) = (OR(hom) + 1)/2. These ORs depend only on the allele frequencies in affected and unaffected parents, and not on the pattern of genotype-disease association. Under simple patterns of association, it is possible to infer from the ORs observed among offspring, the expected ORs for the disease. The decrease of power of offspring studies by comparison with classical case-control studies is evaluated, and an application is given on the association between the apoE polymorphism and coronary heart disease.