Familial resemblance in fatness indicators

Family resemblance in fatness has been studied in 481 individuals from 114 families of French descent living in the greater Quebec city area. Subjects were measured for the following fatness indicators: triceps, biceps, subscapular, suprailiac, abdominal and medial calf skinfolds. The sum of these six skinfolds as well as a prediction of percent body fat (Durnin and Rahaman 1967) were also considered. Data were standardized for appropriate age and sex classes yielding SS scores. The influence of relevant lifestyle variables (energy intake, energy expenditure and socioeconomic status) were statistically removed from SS yeilding residual scores (RS) that were then submitted to familial analyses. Analyses of variance indicate that there is a larger between family variation than within (P less than or equal to 0.01) for SS when considering either the whole nuclear family or sibships; in these instances, the intra-class correlation ranges from 0.15 to 0.26. There was, however, no significant resemblance among spouses for the SS fatness indicators. Similar values were essentially found for RS fatness indicators. Furthermore, husband-wife inter-class correlations were not significant with the exception of subscapular and calf skinfold RS. Covariations between biological relatives are however significant (0.16 less than or equal to r less than or equal to 0.24, P less than or equal to 0.01) for SS and remain essentially unchanged after statistical control over the lifestyle variables (0.16 less than or equal to r less than or equal to 0.40, P less than or equal to 0.01). The findings that spouses do not covary significantly in fatness, while biological relatives of traditional nuclear families exhibit a significant degree of resemblance even after statistical control over daily energy intake, daily energy expenditure and socioeconomic status provide evidence supporting the hypothesis that there is a substantial genetic effect in human fatness.