1286
Sent
R.C. Sreinkamp, N. L. Cowen, W. E. Sint, T, W. SarGcenr and H. E. WALSH
dictary intake and physical activity estimate. The interview required approximately
30 min and was conducted immediately after the anthropometric measurements were
completed. Since selection of subjects was made at the time of the anthropometric
measurements, these individuals were previously unaware they would be interviewed.
The nutritionist asked the individual to describe his physical activity chronologically
and by amountof elapsed time for the immediately preceding 24-hour period with the
reference pomt as time of awakening. Concurrently, description of food eaten was
recorded for each time of eating.
Food was described by kind, preparation and amount. Reference was made to food
models, measuring cups, spoons, rulers and glasses of known volume. Whenever
possible, composite ingredients for mixed dishes such as casseroles and sandwiches
were noted. Weights or prices (to estimate weight) of individually packaged foads,
candy bars, soft drinks, nuts, popcorn, beer, or other, were obtained when available.
Brand namesof special diet items were noted.
In addition to the detailed food and physical activity record, each individual was
questioned concerning adherence to any special diet, use of vitamin supplements, of
other food supplements, of 900 calorie preparations and of salt. Body weight change
within the previous year wasalso recorded.
Calculations of diets to obtain intakes of eleven food nutrients, alcohol and food
groups were performed by the mechanical method of the U.S. Department of Agriculture [41]. Total 24-hour food consumption as well as the amounts consumed during
2-hour intervals after arising were calculated. Master cards were available for nutritive
values of 512 different foods [42]. Since the 516 diet recordslisted over 500 items or
composites of individual foods not available in the master deck, additional master
cards necessarily were calculated. Whenever feasible, substitutions were made for
specific food items using nutrient comparable foods; i.e., chicken for turkey, cornstarch pudding for tapioca pudding, papaya for mangoandsimilar substitutions.
Physical activity was accounted for by time spent sleeping or performing each of
tour arbitrary classifications of expended energy. These were adapted from PASSMORE
and Durnin [43] and were very light (<2.5 Cal/min), light (2.54.9 Cal/min),
moderate (5.0-7.4Cal/min) and strenuous (2 7.5Cal/min). Since energy expenditure
for comparable physical activity is greater with higher body weight, an attempt was
made to define the body weight above which the arbitrary categories of caloric
expenditure would be greater. Application of the regression equation for energy
expenditure in slow walking [43], C=0.047 W+1.02 in which C is energy expenditure
in Calories per minute and W is body weight in kilograms, was made for various
weights. At body weights of 260-310 1b, activities grouped as light (2.5-4.9
Cal/min) were reclassified as moderate (5.0-7 .4 Cal/min).
Somatotyping technique
Somatotyping was carried out by another of the authors (A.T.H.), according to the
technique originated [35] and modified by SHELDON [44], at the Cowell Memorial
Hospital Constitutional Research Laboratory, University of California, Berkeley,
under the direction of HENRY B. Bruyn, M.D.
This procedure determinesthe position of the physical constitution of an individual
within a three dimensional taxonomy whichincludes the normalvariations of human
physique. A somatotypeis a point in space where three determining parameterscross,
Measures of Body Fat and Related Factors in Normal Adults---1
1287
or simply, where three metric distributions cross. These are height, trunk index and
somatotype ponderalindex; the latter two are defined below. Tables* based upon
these parameters, determined by SHELDON for more than 5000 subjects, present the
distribution of the specific somatotype. The latter can be determined from these tables
in the same manner by which various numberfunctions are obtained from mathematical tables.
Height, weight and standardized medical black and white photographs were taken.
The maximum weight experienced by the subject at any age was recorded. From these
data the somatotype ponderal index and trunk index were determined. The somatotype
ponderal index is the determined height divided by the cube root of the maximum
experienced weight. Trunk index is the ratio of the thoracic area to the abdominal
area as measured with a planimeter on the dorsal photograph.
By the use of the tables, the somatotype, described in terms of three components,
was determined from the subject’s height, trunk index and the somatotype ponderal
index. The somatotype components—endomorphy, mesomorphy and ectomorphy—
are measured on a 13-point scale, 1 to 7, with intervals of 4 unit, ie., 3, 34, 4. Data
accumulated on several thousand subjects [44] have established standards for each
componentas follows:
Component
Males
Range
Females
Mean
Endomorphy
Mesomorphy
344
4-41
5
3
Ectomorphy
3-34
(3
SUMMARY
A valid clinically applicable anthropometric method to estimate body fat and lean
body mass in adults would aid in the epidemiological study of diseases associatedwith
overweight. Such a method has been developed from anthropometric measurements
of 2053 healthy Caucasian men and women and Negro men between the ages of 25 and
44 years and from laboratory determinations of body density and total body water for
8 per cent of all subjects. This is the subject of the following report [45].
Successful employment of this method depends upon exact duplication of the
measurement techniques. The anthropometric measurements have been detailed for
reproducibility by interested clinicians.
Methods used to study additional parameters associated with body composition
are given: diet, physical activity and somatotype. Potassium-40 and cesium-137
techniques have been described.
Subsequent papers will present results and interrelationships of the several studied
factors.
*These unpublished tables are six in number for each sex; oneset for each of the following age groups
by sex: (1) 18 years and less; (2) 18-23 years; (3) 24-28 years; (4) 29-33 years; (5) 34-38 years;
and (6) 39-43 years.