Timothy V. Sullivan

Continuing bone expansion and increasing bone loss over a two‐decade period in men and women from a total community sample

As shown in 744 adult men and women aged 30–49 at entry and followed for 21.4 ± 0.9 years there is continuing subperiosteal expansion in both sexes as well as continuing and increasing endosteal surface resorption. In this longitudinal study, bone loss (as shown by medullary cavity expansion) begins by the 5th decade and increases thereafter. The smaller gains at the outer bone surface are essentially independent of the larger losses at the inner (endosteal) surface and neither functionally nor causally related. Though bone loss and net bone loss is nearly as great in men as in women, absolutely speaking, two‐decade bone loss constitutes a larger percentage of the initially smaller bone mass in the female. In both sexes subperiosteal apposition (delta TA) and endosteal resorption (delta MA) are bone‐size dependent though in diametrically opposite directions. These trends in two‐decade bone change are not affected by smoking behavior, alcoholic beverage usage, antihypertensive usage, or early menopausal age. Similarly, the long‐term bone changes prove to be independent of energy and mineral intakes and to long‐term changes in calcium, phosphorus, magnesium, and vitamin D intake. Though dietary intakes do not predict long‐term bone changes, the amount of tissue bone present at entry is highly correlated (> 0.93) with tissue bone 21.4 years later in men and women alike. Accordingly, only a small amount of intraindividual cortical variance in the later years still remains to be explained by life‐style, dietary, medication, and error variables.

Longitudinal Growth in Height, Weight, and Bone Age of Guatemalan Ladino and Indian Schoolchildren

Three longitudinal samples of Guatemalan schoolchildren are compared for amounts and rates of growth in height, weight, and bone age. The samples include children of two ethnic backgrounds: Ladinos, Spanish‐speaking people of, generally, Western cultural orientation; and Indians, people of Mayan cultural descent. The Indians are of very low socioeconomic status (SES) and attend a public school in a rural village. The Ladinos come from two SES groups living in Guatemala City, one of high SES attending a private school and the other of low SES attending a public school. Graphical and statistical analyses show that for all samples of boys and girls there are generally, significant differences between samples (high SES>low SES>Indian) for amounts of growth in height, weight, and bone age. Boys show significant differences in rates of growth between samples, with the high SES sample growing more rapidly than the two low SES samples. Girls show significant differences in the rate of growth in height, but not in the rate of growth in weight or bone age. For Both boys and girls, rates of growth in height and weight differ more between samples than does rate of Skeltal development. These results demonstrate that (1) SES‐related deficits in growth are cumulative during childhood and early adolescence, that (2) rates of growth for boys are, generally, more sensitive to the influence of SES than are the growth rates of girls, and that (3) childhood growth deficits of low SES children of low SES children are likely to carry over into adulthood.

The education of one spouse and the fatness of the other spouse

As shown in 702 wives with 9–12 years of education and 612 husbands similarly educated, the summed skinfolds of one spouse are influenced by the educational level of the other spouse, considerably so for the husbands. Women with 9–12 years of education married to men of lower educational attainment are higher in the sum of four skinfolds while women of similar years of schooling married to men of college education and beyond are leaner (P=0.001). Possible explanations for the effect of the education of one spouse on the fatness level of the other spouse include selective mating in the direction set by the husbands socioeconomic milieu and fatness “drift” on the part of the wives, again in the direction of the husbands socioeconomic status (SES). While these findings do not lend themselves to a simple biological explanation, they do reiterate the effects of socioeconomic variables on fatness level within populations and even within families.

Fatness dependence of skinfold ratios and its implications to fat patterning

As shown in 1639 adult white men and 1851 adult white women aged 20–49, skinfold/skinfold ratios involving the triceps, subscapular, iliac and abdominal skinfolds are all fatness‐dependent, exhibiting a nonlinear relationship with the summed skinfold thickness (2 sf) as variously calculated. When adjusted or partialled for the level of fatness, these ratios no longer exhibit useful correlations with systolic and diastolic blood pressure or total serum cholesterol. Other approaches to relative skinfold thicknesses such as Z‐score ratios or residuals from the skinfold/skinfold regressions may help to distinguish individual differences in fat placement or distribution from differences in the level of fatness itself.

Fatness and obesity among the parents of lean probands

As shown in 1368 proband‐parent pairings involving lean probands and their parents, the fathers and mothers of lean probands tend to be of reduced fatness level themselves (–0.25 Z scores for the sum of two skinfolds) and infrequently obese (10.2% of cases as compared with 16% expectancy). However, these descriptive values are curvilinearily related to the age of the proband, maximizing when the proband is an adolescent. For adolescents, fathers and mothers of lean probands average 0.36 Z scores below the total sample and are least often obese (<8%). It may be concluded that adolescents comprise the prototypical sample for family‐line studies of fatness and leanness, either because “familial” obesity is best expressed in this age group or because parent‐child resemblances in fatness level are a temporal function of the living‐together (or cohabitational) effect.