Familial aggregation of environmental risk factors and familial aggregation of disease

Almost all human diseases have been shown to aggregate familially to some degree. This familiality is generally taken as evidence for the existence of a genetic etiologic mechanism or environmental factors common to family members, or a combination of both. It has been argued that for a disease with strong familiar aggregation, environmental risk factors alone are unlikely to account for such strong aggregation, unless the presumed environmental risk factors are associated with enormous risk. This paper revisits this issue through the use of a novel statistical model. Ascertainment bias aside, the author demonstrates that familial aggregation could be explained by multiple interactive risk factors, each of which may confer a low disease risk and thus contribute only a minuscule portion to disease familiarity. For example, two correlated risk factors (r = 0.5), each with a relative risk of 5 and acting multiplicatively, could give rise to a sibling relative risk of 1.96. Therefore, it may not be sufficient to argue for a genetic component for a disease based solely on the notion that no high risk environmental factors have been found. In view of this, there is a need to examine carefully the roles of multiple environmental risk factors in disease familiarity.