Association, Genetic

Analytical epidemiologic studies investigate potential associations between exposures (e.g., risk factors) and outcomes (e.g., as identified in quantitative data on traits or diseases) of interest. Genetic association studies are a special type of analytical epidemiologic studies in which the exposure of interest is represented by a genetic factor. Genetic factors are usually represented by common forms of genetic variation, such as single nucleotide polymorphisms (i.e., genetic loci for which different DNA variants can be present among different individuals) or deletions or insertions (i.e., losses or additions of a sequence of DNA in a chromosome).

Identification of genetic associations can have important implications in the prevention, diagnosis, and treatment of diseases, including complex diseases such as cancer and cardiovascular illnesses. For example, identification of specific polymorphisms related to the development of disease may lead to prevention programs targeted to specific subgroups of individuals at high risk. Similarly, genetic variation that affects response to specific medications may be taken into account in selecting the most effective and least toxic pharmacologic treatments for individual patients.

Genetic associations can be investigated within several epidemiologic study designs, including the classic case-control study design. In a case-control study of genetic risk, individuals with and without a specific disease are compared according to their exposure to a specific genetic risk factor, usually whether they have or do not have a particular genetic variant. However, the genetic nature of the exposure influences the methodological approach used in these studies in several respects. One reason for this influence is that even in the simplest case of diallelic loci (i.e., genetic loci at which only two possible alternative alleles are present in the population), double copies of most polymorphisms are present in humans (with the exception of loci on the sex chromosomes), and they are located in proximity to other polymorphisms; therefore, genetic risk factors can be linked to disease in several different ways. For example, the disease of interest can be associated with

In addition, because at the population level alleles can have different frequencies in different groups (e.g., ethnic groups) and can be assorted at different loci in a nonrandom fashion (i.e., they can be in linkage disequilibrium), the validity of associations between polymorphisms and disease in epidemiologic studies can be confounded by associations related to [Page 51]population stratification across ethnic groups and/or by the existence of linkage disequilibrium between these polymorphisms and other nearby polymorphic loci. Despite these limitations, genetic association studies are usually less vulnerable to confounding and bias than classic epidemiologic studies. For example, in genetic case-control studies, the risk of recall bias is minimized and the temporal sequence between exposure and outcome is resolved because

...