In the course of an epidemiological study covering an entire community, Tecumseh, Michigan, the familial aggregation of various levels of four variables associated with coronary heart disease-serum cholesterol, systolic blood pressure, relative weight, and blood glucose after a standard load-was assessed quantitatively by determining the relative frequencies of high, intermediate, and low levels of the variables for several familial combinations. High, intermediate, and low levels for each variable were defined in terms of quintiles (highest, middle three, lowest, respectively) of the distribution of the variable specific for age and sex.
Parents and their children and siblings were compared. There was a striking resemblance between children and their fathers and mothers for all the four variables under study, as well as for height which was used as a frame of reference. The degree of resemblance between parents and children was approximately the same at the three levels-"low," "intermediate," and "high"-of the variables, pointing to a multifactorial mode of determination.
In an attempt to assess the relative role of environmental and genetic factors with respect to the determination of the variables under study, degrees of aggregation by age were compared. An appreciable degree of resemblance was found among siblings below the age of 16 who may be presumed to share the same environment to a great extent. Aggregation was less marked for siblings between the ages of 16 and 40, a time when less of the same environment is shared, but it rose again after age 40. It is postulated that the observed rise after age 40 is due to an increase in the penetrance of the genes involved, at middle age and beyond. In addition, this pattern of aggregation suggests that environmental factors may play a stronger role prior to middle age. A similar subdivision by age, was used to study relationships between parents and children; although no corresponding differences between age groups were found, this apparent inconsistency with the sibship data could be explained, in part, by differences in age composition of the groups involved.
The practical implication of the foregoing pattern of familial aggregation lies in the possibility of specifying for an individual the risk of developing high values of a variable predisposing to coronary heart disease in the future on the basis of the values shown by his parents and siblings.
