Given recent genetic advances, it is not surprising that genetics information is increasingly being used to improve health care. Thousands of conditions caused by single genes (Mendelian diseases) have been identified over the last century. However, Mendelian diseases are rare; thus, few individuals directly benefit from gene identification. In contrast, common complex diseases, such as obesity, breast cancer, and depression, directly affect many more individuals. Common complex diseases are caused by multiple genes, environmental factors, and/or interaction of genetic and environmental factors. This article provides a framework for the successful conduct of gene—environment studies. To accomplish this goal, the basic study designs and procedures of implementation for gene—environment interaction are described. Next, examples of gene—environment interaction in obesity epidemiology are reviewed. Last, the authors review reasons why epidemiological studies that incorporate gene—environment interaction have been unable to demonstrate statistically significant interactions and why conflicting results are reported.
