NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Institute of Medicine (US) Committee on Military Nutrition Research; Marriott BM, Grumstrup-Scott J, editors. Body Composition and Physical Performance: Applications For the Military Services. Washington (DC): National Academies Press (US); 1990.

Body Composition and Physical Performance: Applications For the Military Services.
Show detailsFrank I. Katch
New Approaches To Body Composition Evaluation
Estimating Excess Body Fat From Changes In Abdominal Girth
A new method has been devised for determining changes in percent body fat (BF) based on the difference between an initial value for abdominal girth (AG) and a calculated ''target'' AG based on a desired level of percent BF (Katch et al., 1989). The new method differs from traditional approaches, such as fatfold and girth-generated regression analysis based on densitometry (Jackson and Pollock, 1978; Katch and McArdle, 1973; Pollock et al., 1976; Wilmore and Behnke, 1968, 1969, 1970), bioelectrical impedance (Deurenberg et al., 1990; Segal et al., 1985), and other indirect appraisal procedures (Borkan et al., 1985; Lukaski, 1987) that first estimate percent BF, and then the individual attempts to achieve a desired change in body mass or composition. With these different methodologies, especially the fatfold technique, accuracy is often attenuated due to statistical factors related to validity (Katch and Katch, 1980), particularly the choice of measurement sites. Nevertheless, the use of regression equations for a one-time assessment has provided important quantitative information. However, using the same regression equation for pre- and post-testing will produce unacceptable estimates of changes in percent BF (Barrows and Snook, 1987). In contrast, the new approach is based on a different strategy than the previous methods. With the new approach, the question is asked: How much does the abdominal girth need to be reduced to achieve a desired percent BF?
Development of the new method had its roots in clinical experience with subjects who altered their body composition dramatically during different regimens of exercise and caloric restriction. But it was Behnke (1963, 1969) who first detailed quantitatively that two abdominal girths (natural waist and umbilicus) showed the greatest absolute changes with body mass loss in relation to 11 other trunk and extremity sites. There is also good experimental evidence that increases in total BF result in proportional increases in abdominal fat (Kvist et al., 1986). Thus, it seemed logical that changes in percent BF could coincide with reductions in excess AG, or that reducing excess AG should coincide with proportionate reductions in percent BF. The proposed method is based on the difference between an initial value for AG and a target AG that corresponds to a "desired" percent BF.
Calculation Of Excess Abdominal Girth
Excess AG is measured with a calibrated anthropometric cloth tape at the natural waist or the abdomen at the level of the umbilicus. A two-step procedure is required to calculate excess AG for an individual. Step 1 requires the development of constants based on the source data for subsequent application in Step 2.
• Step 1. The calculation of excess AG is based on large-scale anthropometric surveys in the military. For men, these included soldiers (White and Churchill, 1971) (Table 8-1) and U.S. Army aviators (Churchill et al., 1971) (Table 8-2), and for women, U.S. Air Force women (Clauser et al., 1972) (Table 8-3). From these data, a target AG was computed as the product of F (kg½ of body mass per meters [m] of stature) and a constant Q. This constant was calculated as the ratio of AG at a predetermined value for percent BF to F (Q = AG/F). From Table 8-1, for example, Q = 12.36 at the fiftieth percentile (Q = 78.9/6.381).
Table 8-1
Source Data for 6,682 U.S. Army Soldiers.
Table 8-2
Source Data for 1,482 U.S. Army Aviators.
Table 8-3
Source Data for 1,357 U.S. Air Force Enlisted Women.
With the data sets from the military, it was necessary to estimate body composition because such criterion measures were not included in the surveys, or they were limited to a small subsample of the data (Clauser et al., 1972). For the soldiers and aviators, fat-free mass (FFM) was computed by the anthropometric method of Wilmore and Behnke (1968), and for U.S. Air Force women, percent BF was derived from a surface area equation that included triceps, subscapular, and supra-iliac fatfolds (Katch et al., 1979) and the variable F (square root of the quantity body mass in kg divided by stature in meters) (Behnke and Wilmore, 1974).
For example, for a group of men (or an individual) with an AG of 89.7 cm, body mass of 85.5 kg, and stature of 1.876 m, the quotient F is (85.5/ 1.876)× = 6.751. This value is then multiplied by the Q value at the desired percent BF (13.23 at the ninety-fifth percentile that corresponds to a desired percent BF of 19.1 percent; Table 8-1) to yield a target AG of 84.4 cm.
• Step 2. Excess AG is computed as the measured AG (89.7 cm in the above example) minus the target AG at the fiftieth percentile of 84.4 cm from Step 1. The 5.3 cm difference (89.7 cm minus 84.4 cm) is the excess AG. The objective is straightforward; try to attain the target AG that corresponds to the desired percent BF. In this example, the predetermined, desired level for percent BF was chosen as 19.1 percent.
An important consideration with the new approach is to decide on the target or desired level of percent BF. If different percentile values are used for percent BF, then different Q values must be applied in Step 1.
The Ag Method During Body Mass Loss By Exercise Plus Diet
Table 8-4 shows the application of the AG method to four obese men and four obese women who reduced their body mass by an average of 20.5 kg in experiments that involved 1 hour daily of cycling and walk-jog exercises over a 6-month period coupled with mild dietary restriction (Katch and Katch, 1984).
Table 8-4
Application During Weight Reduction in Obese Men and Women.
The results of the analyses based on densitometry to estimate percent BF and anthropometry to measure the change in AG (ΔAG) were remarkable for this relatively small sample of subjects. For the men and women of the same age, change in body mass (ΔBM) divided by ΔAG was almost identical for the first two tests (1.08 for men and 1.09 for women). These ratios indicated that the basic assumptions of the current analyses were valid because a ratio of 1.00 would signify a precise correspondence between ΔBM and ΔAG. Both groups reduced nearly the same amount in their AG (men, 13.7 percent; women, 14.6 percent); the women, however, reduced their percent BF to a greater extent (11.0 percent BF units) compared to the men (7.9 percent BF units). This difference probably occurred because the women had a higher initial percent BF (45.0 percent by densitometry) compared to the men (29.7 percent). The women also lost 5.5 kg more body mass than did the men.
An important consideration is the extent of agreement between the measured AG and the predicted AG using F x Q at the desired percent BF. For the initial measurements, the correspondence between the measured and target AG would not be congruent because the subjects were all obese. However, as they begin to reduce body mass, percent BF, and AG, the relationship between the target and measured AG should converge. Inspection of the individual data indicated that this did occur during Tests 1 and 2 except for Subjects 3 and 7. Male Subjects 1 and 2 were model subjects to illustrate the continued decline of the measured minus predicted AG as time progressed. For the first two tests, the percent changes in AG for the group, expressed as (AG minus F × Q)/AG, decreased in the predicted pattern (men from 5.6 percent to 3.4 percent; women from 6.9 percent to 4.9 percent). For Subjects 1 and 3 who were measured 4 times, there was a slight increase in the percent changes in AG, probably because there were no further decreases in BM or AG, and they achieved their target AG and desired percent BF. This also was true for Subjects 2 and 6. For the latter, her target and measured AG coincided at just about the desired level for percent BF. For the remaining subjects, there were discrepancies between the target and measured AG. Although the measured AG actually became smaller than the target AG, percent BF remained above the desired levels defined by the gender-specific Q values. Either there were small errors in the AG measurements, or the group Q values need refinement.
The Ag Method During Body Mass Loss By Diet Only
Recent data made available by A. Weltman at the University of Virginia at Charlottesville shows the application of the AG method in 6 obese men and 19 women who participated in a controlled liquid-diet weight loss program. Table 8-5 shows the changes in body composition for the women and men. The salient features include changes in BM and two AGs (umbilicus and waist girths). For women, the value for Q at the fiftieth percentile for the waist girth is 11.19 (Table 8-3). The men reduced BM more than the women (24.2 kg versus 19.3 kg), as well as waist and umbilicus girths, percent BF, and absolute fat mass. Of interest are the nearly similar gender ΔBM/ΔAG. For women, the ΔBM/ΔAG1 is 1.26, and the ΔBM/ΔAG2 is 1.18. For men, the ΔBM/ΔAG1 is 1.20, and the ΔBM/ΔAG2 is 1.16. Such results provide additional corroborative evidence for the close correspondence between mean ΔBM relative to mean AG (ΔBM/ΔAG). However, a different pattern emerges when the ΔBM/ΔAG is computed for individuals.
Table 8-5
Changes in the Body Composition of 19 Obese Women and 6 Obese Men on a Reducing Diet.
Figure 8-1 shows the results of the simple regression analysis (with 90 percent confidence bands) for the ΔBM/ΔAG for 19 women (top) and 6 men (bottom). The important result is that for men and women, the association is strongest between change in waist girth (abd1) and change in BM (r 2 =.74 for abd1, and r 2 = .88 for abd2). While the results for ΔBM/ ΔAG1 for men is encouraging, the sample size is really too small for meaningful interpretations, and more data are required for confirmation. What can be stated with some confidence at this point is that the change in fat content at the abdominal site as measured by change in AG parallels, in general, the overall change in BM that occurs due to caloric restriction plus exercise or by caloric restriction alone.

Figure 8-1
Regression analysis for the change in two abdominal girths plotted as a function of change in body mass for female (top) and male (bottom) subjects. The 90 percent confidence bands are shown for the slope of the regression line.
The main requirement in future experiments should be to secure large samples (as in the military anthropometric studies) and include subjects of diverse body composition. A large data base that includes complimentary anthropometric data and a criterion measure of BF would permit a more accurate determination of the Q constants used in Step 1 for different levels of BF. It would also be desirable to secure anthropometric data and criterion BF measures during BM loss. These data would allow for the validation of the Q constants with changes in body composition.
In summary, the objective for individuals who need to reduce their percent BF would be to try and attain a target AG. If excess girth is not too large, attainment of the target AG should coincide with an a priori determined level of percent BF. However, if the excess AG is considerable, then the target AG becomes a "first approximation" with BM loss, and a further body composition evaluation is required to ensure congruence with the desired percent BF. If individuals can reduce their excess AG by bringing their AG in line with the target AG, their percent BF should coincide with the desired level of BF. The latter, however, is difficult to ascertain because one must define what in fact is normal or acceptable percent BF in relationship to age. This problem is further influenced by such factors as physical condition (varying from sedentary to very physically active) and race.
The Body Profile: An Enhancement Of The Somatogram
The concept of the somatogram (SOM) was established by Behnke et al. (1959) to describe body shape expressed in percentage deviation units from reference standards developed from military and civilian large-scale anthropometric surveys (Hertzburg et al., 1963; O'Brien and Shelton, 1941; Welham and Behnke, 1942). The basis of the SOM is the translation between a squared matrix of 12 girths and the previously described body size module F (square root of the quantity body mass in kg divided by stature in decimeters)1 into a graphic description of the percentage deviations from the reference standard.
To construct the SOM, each of 12 girths (g) are divided by their proportionality constants (k) to obtain a deviation (d) score (d = g/k). The k constant is computed as g/D, where D equals the sum of the g values divided by 100. The SOM presents the percentage deviation of each d quotient from D. This graphic approach has been used to show changes in body size during growth and development (Huenemann et al., 1974; Katch, 1985a), to depict progressive changes in overall body shape with aging (Behnke, 1963, 1969), and to describe gender differences in athletic groups (Behnke, 1963, 1968; Katch, 1985b; Katch and Katch, 1984). The SOM approach has now been enhanced, and the technique is referred to as the body profile, or more specifically, the ponderal SOM (PSOM) (Katch et al., 1987).
The SOM analysis did not permit translation of girth size into a volume or weight entity that relates to the body as a whole. The original SOM also did not differentiate between muscular and nonmuscular areas of the body; thus, nonmuscular girths such as the abdomen and hips were integrated with muscular parts such as the flexed biceps, thigh, and calf. Because the deviation of each d from D is based on the matrix of girths, each g is in fact related to itself because it is part of D. Although this discrepancy is probably of minor importance to the graphic representation of body shape, it still does not permit a clear-cut separation of the muscular and nonmuscular components.
The body profile is an extension of the SOM. Girth measures are converted to ponderal equivalent weight values. The matrix of girths can be separated into muscular and nonmuscular components and compared as mass equivalents. In this paper, the PSOM is presented for a world champion male body builder where there is excessive muscular development, especially in the biceps, chest, and shoulders.
Original Somatogram Calculations
The left side of Table 8-6 lists the measurements and k constants for the reference man and woman (Behnke et al., 1978). To calculate SOM, each girth (g) is divided by k to obtain a ratio referred to as d (d = g/k). The reference value is then computed as D, where D is the sum of the girths (D = Σ girths) divided by the sum of the k values (Σ k = 100). The graphic representation of body shape is a plot of the percentage deviation of each d from D (percent deviation = [d - D]/D). If an individual's measurements conformed precisely to the reference values, there would be no deviations, and the SOM would plot as a vertical line. An example of a SOM for a 40-year-old man is shown in the left side of Figure 8-2. For a biceps of 40.2 cm and D = 6.771(Σ 11 g/100), d for the biceps is 7.60 (d = 40.2/5.29), where 5.29 is the k(biceps) value for the reference man listed in Table 8-6. Expressed as a deviation from D, d(biceps) is 12.2 percent larger ([7.600 minus 6.77]/6.771) multiplied by 100, and is plotted on the somatogram as +12.2 to the right of the zero axis. The d values for the other girths are plotted in a similar fashion.
Table 8-6
Measurements and Proportionality Constants for the Reference Man and Woman, and Conversion of Anthropometric Girths into Ponderal Equivalents.

Figure 8-2
Left. Original somatogram (SOM). There is no distinction between muscular and nonmuscular components. Scores within a range of +2 percentage deviations from the zero reference line are considered to be within normal limits of variation compared with the (more...)
Ponderal Sornatogram Calculation
The right side of Table 8-6 lists the constants to calculate the PSOM. There are two components: (1) ponderal equivalent muscular component (PEM), which includes the shoulder, chest, biceps, forearm, thigh, and calf, and (2) ponderal equivalent nonmuscular component (PENM), which includes two AG measures and their average, as well as hips, knee, wrist, and ankle.
The constants for the individual girths are calculated from the data of the reference man and woman as k = g/F, where g = individual girth in cm, and F = the square root of the reference man and woman median weight in kg divided by reference man and woman median stature in dm. For the reference man, the value for F is 2.000; for the reference woman F is 1.852 (Behnke et al., 1978).
The ponderal equivalent (PE), expressed in kg for each girth, is computed as the square of the quotient g/k multiplied by stature in dm. For example, the PE for the shoulders for the reference man is (g/k)2 multiplied by stature or (110.8/55.4)2 × 17.4 = 69.6 kg. For the reference man, the PE values for all of the girths are identical to the reference median weight of 69.6 kg; the same is true for the reference woman. All of the PE values for the girths are identical to the reference woman median weight of 56.2 kg. For the reference models, the deviations of each PE from their respective standards are necessarily zero because there is no deviation from group symmetry (the reference values represent the standard).
A unique aspect of the PSOM is the calculation of the d values. In the original SOM, it was not possible to separate the muscular from the non-muscular girths because D was calculated as the sum of all the girths/100. Thus, a particular d value was related to the sum of the girths that included that particular girth.
This complication is avoided in the PSOM by comparing the PEM girth values with the average of the PENM values, and vice versa. There will not be exact numerical equivalency between the total (cm/k)2 multiplied by stature and the average of the PE values because of differences in proportionality between the reference man and woman, and among individuals or groups of individuals. For the PSOM in Figure 8-2, the specific k values were from the PSOM listed in Table 8-6.
In summary, the original Behnke SOM to quantify body shape is a valid approach (Sinning and Moore, 1989) for partitioning a matrix of girths into PEM and PENM components that can be related to the body as a whole. In male body builders, for example, excess muscular development appears to predominate in the biceps without compensatory hypertrophy in the lower limbs. Even at the extremes, which includes the massively obese as well as diminutive and large adolescents (Katch et al., 1987), there appears to be a fundamental, intrinsic association between an individual's body weight and the squared matrix of girths multiplied by stature. Such relationships have useful clinical and research applications for the study of obesity and its relationship to growth and development, as well as various facets of human performance. The next section explores some of these relationships to muscular strength.
Some Relationships To Dynamic Muscular Strength
Traditional views suggest that an individual's body size is directly related to muscular strength. However, there are conflicting data concerning the relationships among muscular strength and various measures of body size, including limb dimensional characteristics. Some studies report correlations of r = .61 to .96 among strength and body size and limb dimensions (Clarkson et al., 1982; Ikai and Fukunaga, 1968; Nygaard et al., 1983; Sale et al., 1987; Schantz et al., 1983; Tappen, 1950; Tsunoda et al., 1985; Young et al., 1982). It is likely, however, that these correlations are spuriously inflated because subsets of samples are included that combined men with women and trained with untrained subjects of different ages. In contrast, other studies report correlations of r = -0.25 to 0.52 among body size variables and muscular strength (Cureton, 1947; Ergen et al., 1983; Katch and Michael, 1973; Smith and Royce, 1963; Watson and O'Donovan, 1977). Such results suggest that additional factors, such as muscle fiber type, neural control of force production, and biomechanical alignment of muscles and joint levers, help to explain a greater proportion of the variance in strength.
In a recent study (Hortobagyi et al., 1990), the relationship was examined between individual differences in muscular strength versus body size, body shape, limb volume, muscle plus bone cross-sectional area, and the
PSOM. This study was done with two homogeneous groups using a statistical approach that avoided the confounding influence of individual differ
ences in age, gender, and training status.
Experimental Procedures
The subjects were 42 Caucasian men from physical activity classes at the University of Massachusetts. Two different test protocols were used to assess muscular strength:
- Bench press (BP) and squat (SQ) were measured with an isokinetic dynamometer during three sets of two repetitions for BP and three sets of three repetitions for SQ. There was a 1-minute rest between each set, and approximately a 3-second pause between repetitions.
- BP and knee extension (KE) were measured with a hydraulic dynamometer. Subjects performed one set of five continuous repetitions for the seated BP and right KE. Based on the strength scores, subjects were placed into high strength (HS) and low strength (LS) groups by converting each of the four measures of strength into
-scores that were then averaged and ranked. The
-score procedure was used to characterize muscular ''strength'' as an overall component without weighting of the strength measures.
Anthropometric Assessment
The measurements included six fatfolds, 11 girths, and two segment lengths.
Calculations
Muscle plus bone cross-sectional area for the biceps (MCSABi) was calculated as:

where r is the radius of the upper arm calculated from biceps girth, BiFF is biceps fatfold, and TrFF is triceps fatfold.
Muscle plus bone cross-sectional area for the thigh (MCSAThi) was estimated as:

where r is radius of the thigh and ThFF is thigh fatfold.
The volume of the upper arm and thigh was estimated as:

where h is the length of the upper arm or thigh in cm, R is upper ann or thigh girth, and r is elbow or knee girth.
FFM (fat-free mass) was predicted by the method of Wilmore and Behnke (1969), where FFM = 1.08BM + 44.6 - 0.74 (AG); BM is body mass in kg, and AG is abdominal girth (umbilicus) in cm. Body shape was described by the PSOM outlined in the previous section.
For the statistics, a priori and postmortem sample size estimations using Cohen's Case 2 formula with an alpha level of 0.05 and power of 0.80 required a minimum sample size of 12 subjects per group. The a priori effect size was a 25 percent difference between HS and LS in muscular strength. The final sample size (n = 21 per group) was nearly twice that required.
Between-group differences for single pairs of variables were evaluated with a two-tailed independent t-test. A one-way multivariate analysis of variance (MANOVA) was used to compute the differences between the two groups in the overall PSOM and pairs of various dependent variables. If the Hotelling's T 2 value was significant, a two-tailed independent t-test was used for pairwise contrasts with an adjusted confidence level. Pearson product-moment correlations were computed among selected variables, and the differences in correlations between the groups were compared by z-transformation.
Results
As expected, HS had significantly greater strength for isokinetic SQ (21.1 percent), BP (23.3 percent), hydraulic BP (16.7 percent), and hydraulic KE (24.2 percent).
Anthropometry
There were no significant differences between HS and LS for age (1.5 percent), stature (2.2 percent), or sum of six fatfolds. In contrast, there were significant differences between HS and LS in BM (6.9 percent; p < 0.05), FFM (6.6 percent; p < 0.01), and 11 girths. For the girths, applying the pairwise follow-up, 7 of the 11 girths were significantly larger for HS, including the shoulders (3 percent), chest (4.4 percent; p < 0.05), biceps (5.6 percent), forearm (5.9 percent), knees (4.5 percent; p < 0.05), wrists (4.0 percent; p < 0.001), ankles (7.7 percent; p < 0.01), and sum of 11 girths (3.6 percent; p < 0.01). There were no significant differences in thigh or calf girths (~3.0 percent).
There were differences between HS and LS in thigh volume (13.2 percent; p < 0.01) and upper arm volume (7.2 percent; p < 0.05). HS had a 14.8 percent greater MCSABi (p < 0.001) and 13.8 percent greater MCSAThi than LS (p < 0.05). The mean values for thigh length were significantly different between HS (41.7 cm) and LS (39.9 cm), but not for upper arm length (17.9 cm for HS versus 18.3 cm for LS).
There were significant differences in the PSOM between HS and LS. The sum of the muscular components (shoulders, chest, biceps, forearm, thigh, calf) was also significantly larger by 2.8 percent for HS. In addition, the sum of the five nonmuscular components (abdomen, hips, knee, wrist, ankle) was significantly larger by 10.1 percent for HS.
The percent deviations for the PEM from the PENM for PSOM varied from 0.4 percent (thigh) to 12.3 percent (biceps) for the PEM for LS, to minus 1.8 to 16.0 percent for HS. The deviations of the PE NM from PEM ranged from minus 11.7 to 2.2 percent for LS, and minus 8.9 to minus 1.8 percent for HS. None of the between-group comparisons were significant.
Correlations
Table 8-7 presents the intercorrelations between strength, BM, FFM, fatfolds, limb volume, and limb CSA for the total sample and the two subsamples. For HS and LS, all of the r values were less than r = 0.56. There were no significant differences in any of the r values between HS and LS. The observed r values between BM and the four measures of strength averaged r = 0.23 (p > 0.05) for HS and LS; they did not increase significantly and were not significantly higher after applying the Guilford correction for restriction of range ( r c = 0.30; p > 0.05). The r values between limb volume and the strength measures averaged r = 0.31 and r c = 0.41 (p > 0.05). The corresponding r values for estimated FFM versus strength were r = 0.27 and r c = 0.34 (p > 0.05). Similarly, low correlations were obtained for the comparisons of MCSA versus strength (r = 0.43 and r c = 0.59; p < 0.05), and for the sum of six fatfolds versus strength (r = 0.27 versus r c = 0.36; p > 0.05). Thus, the observed r values between strength and anthropometry averaged r = 0.30, and the average corrected r values increased slightly to only r c = 0.40 (p > 0.05).
Table 8-7
Correlations (r) and the Correlations Corrected for Restriction of Range (r c) Between Strength and Body Size and Segment Size for the Low- (n = 21) and High-Strength (n = 21) Groups, and for the Total Sample (n = 42).
It was postulated that subjects classified as high and low strength would shed light on the relationship between size and strength. In the studies where the average r between estimates of strength and BM and estimates of strength and muscle size exceeded r ≥ 0.80, it is likely that these r values were spuriously inflated due to at least five factors:
- large individual variations in body size and physique among comparison groups,
- combining men and women in the salient comparisons,
- mingling trained and untrained subjects,
- pooling young and old subjects in the data analyses, and
- using the mean value to correlate strength and body size for nine different sport groups, rather than using the data of individuals for each group.
Thus, the moderate to high r values were probably generated because of a "range of talent" effect in body size and physique, gender, training status, and age. In studies that used more homogeneous samples, lower r values averaging r ≤ 0.60 were reported between strength, BM, and FFM in men and women. The low r values observed in the present study support these latter findings.
In the present data, the r values were all low between BM and muscular strength regardless of strength level (r = 0.29 for HS, and r = -0.12 for LS) or between estimated FFM and strength (r = 0.33 for HS, and r = -0.07 for LS; Table 8-7). There also was no improvement in the r values when arm or leg strength was related to muscle CSA (average r = 0.38 for HS, and r = 0.27 for LS) or to segmental volume (r = 0.12 for HS, and r = 0.27 for LS).
Additional comparisons between HS and LS revealed some interesting findings. For the girth comparisons between the groups, 7 of 11 girths were significantly larger for HS than LS (range 3.0 to 7.7 percent). Also, thigh and calf girth did not differ significantly between HS and LS, but the non-muscular knee and ankle girths of the lower body did. Perhaps HS subjects had a propensity for upper body development that produced the significantly larger muscular upper body components.
The relationship among the PSOM and the various expressions of muscular strength revealed that the PE (ponderal equivalents) for HS and LS showed identical patterns; that is, the same sites for girths and PE were different between HS and LS. Such data support the intrinsic validity of PSOM.
In summary, the present data illustrate the relative independence of individual differences in strength and measures of anthropometry and body composition. Thus, other factors must be responsible for explaining the approximately 21 percent difference in muscular strength between HS and LS. Large individuals were not superior in overall muscular strength compared to their counterparts of smaller body size and shape. Such differences in strength cannot be explained by individual differences in girths, fat-folds, CSA, segmental volume, or PSOM. The present conclusions may not apply to groups of highly trained individuals with extreme muscular development and strength. Factors other than muscle size alone, for example, neural and/or muscle and joint mechanical properties, may play at least an equally important role in explaining individual differences in muscular strength.
Acknowledgments
The strength studies were supported in part by a grant from Hydra-Fitness Industries, Belton, Texas and Contract DAMD 17-80-C-0108, U.S. Army Medical Research and Development Command, Ft. Derrick, Maryland.
References
- Barrows, K., and J. T. Shook. 1987. Effect of a high-protein, very-low calorie diet on body composition and anthropometric parameters of obese middle-aged women. Am. J. Clin. Nutr. 45:381-390. [PubMed: 3812337]
- Behnke, A. R. 1963. Anthropometric evaluation of body composition throughout life. Ann. N.Y. Acad. Sci. 110:75-78. [PubMed: 14061658]
- Behnke, A. R. 1968.Physique and exercise. Pp. 359-386 in Exercise Physiology, H. Falls, editor. , ed. New York: Academic Press.
- Behnke, A. R. 1969. New concepts in height-weight relationships. Pp. 25-35 in Obesity, N. L. Wilson, editor. , ed. Philadelphia: F. A. Davis.
- Behnke, A. R., and J. H. Wilmore. 1974. Pp. 45-84 in Evaluation and Regulation of Body Build and Composition. Englewood Cliffs, N.J.: Prentice-Hall.
- Behnke, A.R., O. E. Guttentag, and C. Brodsky. 1959. Quantification of body weight and configuration from anthropometric measurements. Hum. Biol. 31:213-234. [PubMed: 13798291]
- Behnke, A. R., F. I. Katch, and V. Katch. 1978. Routine anthropometry and arm radiography in assessment of nutritional status; its potential. J. Parenter. Enteral. Nutr. 2:532-553. [PubMed: 569722]
- Borkan, G. A., D. E. Hults, S. G. Gerzof, and A. H. Robbins. 1985. Comparison of body composition in middle-aged and elderly males using computed tomography. Am. J. Phys. Anthropol. 66:289-295. [PubMed: 3985138]
- Churchill, E., J. T. McConville, L. L. Laubach, and R. M. White. 1971. Anthropometry of U.S. Army Aviators. Technical Report No. 72-52-CE. Clothing and Personal Life Support Equipment Laboratory, U.S. Army Natick Laboratories, Natick, Mass.
- Clarkson P. M., W. Kroll, and A. M. Melchionda. 1982. Isokinetic strength, endurance, and fiber type composition in elite American paddlers. Eur. J. Appl. Physiol. 48:67-76. [PubMed: 7199455]
- Clauser, C. E., J. T. McConville, E. Churchill, and L. Lanbach. 1972. Pp. 1054-1069 in Anthropometry of Air Force Women. Technical Report No. 70-5. Aerospace Medical Research Laboratories, Wright Patterson Air Force Base, Ohio.
- Cureton, T. K. 1947. Pp. 356-389 in Physical Fitness Appraisal and Guidance. St. Louis, Mo.: C. V. Mosby.
- Deurenberg, P., K. van der Kooij, P. Evers, and T. Hulshof. 1990. Assessment of body composition by bioelectrical impedance in a population aged greater than 60 y. Am. J. Clin. Nutr. 51:3-6. [PubMed: 2296928]
- Ergen, E., N. Gambuli, L. M. Leonardi, and A. Dal Monte. 1983. Relationship between body composition, leg strength, and maximum alactacid anaerobic power in trained subjects. J. Sports Med. 23:399-403. [PubMed: 6677805]
- Hertzburg, H. T. E., G. S. Daniels, and E. Churchill. 1963. Anthropometry of Turkey, Greece, and Italy. AGARDograph 73. Elmsford, N.Y.: Pergamon.
- Hortobagyi, T., F. I. Katch, V. L. Katch, P. F. LaChance, and A. R. Behnke. 1990. Relationships of body size, segmental dimensions, and ponderal equivalents to muscular strength in high and low strength subjects. Int. J. Sports Med. 11:349-356. [PubMed: 2262226]
- Huenemann, R. L., M. C. Hampton, A. R. Behnke, L. R. Shapiro, and B. Mitchell. 1974. Teenage Nutrition and Physique. Springfield, Ill.: Charles C. Thomas.
- Ikai M., and T. Fukunaga. 1968. Calculation of muscle strength per unit cross-sectional area of human muscle by means of ultrasonic measurement. Int. Z. angew. Physiol. einschl. Arbeitsphysiol. 26:26-32. [PubMed: 5700894]
- Jackson, A. S., and M. L. Pollock. 1978. Generalized equations for predicting body density of men. Br. J. Nutr. 40:497-504. [PubMed: 718832]
- Katch, V. L. 1985. a. Assessment of body composition; comments on prediction. Euro-Nut Publication Series; Body Composition and Fat Patterning. London: Ciba Foundation, London.
- Katch, V. L. 1985. b. Body composition and sports medicine: Clinical Considerations. Pp. 73-78 in Body Composition Assessments in Youth and Adults, A. F. Roche, editor. , ed. Report of the Sixth Ross Conference on Medical Research. Columbus, Ohio: Ross Laboratories.
- Katch, F. I., and V. L. Katch. 1980. Measurement and prediction errors in body composition assessment and the search for the perfect prediction equation. Res. Q. Exerc. Sport 51:249-260. [PubMed: 7394287]
- Katch, F. I., and V. L. Katch. 1984. The body composition profile. Techniques of measurement and applications. Clin. Sports Med. 3:31-63. [PubMed: 6400334]
- Katch, F. I., and W. D. McArdle. 1973. Prediction of body density from simple anthropometric measurements in college-age men and women. Hum. Biol. 45:445-454. [PubMed: 4750412]
- Katch, V. L., and E. D. Michael. 1973. The relationship between various segmental leg measurements, leg strength, and relative endurance performance of college females. Hum. Biol. 45:371-383. [PubMed: 4750407]
- Katch, F. I., A. R. Behnke, and V. L. Katch. 1979. Estimation of body fat from skinfolds and surface area. Hum. Biol. 51:411-414. [PubMed: 533717]
- Katch, F. I., A. R. Behnke, and V. L. Katch. 1987. The ponderal somatogram: Evaluation of body size and shape from anthropometric girths and stature. Hum. Biol. 59(3):439-458. [PubMed: 3610119]
- Katch, F. I., V. L. Katch, and A. R. Behnke. 1989. A new approach for estimating excess body fat from changes in abdominal girth. Am. J. Hum. Biol. 2:125-131. [PubMed: 28590534]
- Kvist, H., L. Sjostrom, and U. Tylen. 1986. Adipose tissue volume determinations in women by computed tomography: Technical considerations. Int. J. Obes. 10:53-57. [PubMed: 3710689]
- Lukaski, H. C. 1987. Methods for the assessment of human body composition: Traditional and new. Am J. Clin. Nutr. 46:537-556. [PubMed: 3310598]
- Nygaard, E., M. Houston, Y. Suzuki, K. Jorgensen, and B. Saltin. 1983. Morphology of the brachial biceps muscle and elbow flexion in man. Acta Physiol. Scand. 117:287-292. [PubMed: 6223510]
- O'Brien, R., and W. C. Shelton. 1941. Women's Measurements for Garment and Pattern Construction. U.S. Department of Agriculture. Misc. Publ. 454. Washington, D.C.: U.S. Government Printing Office.
- Pollock, M. L., T. Hickman, and Z. Kendrick. 1976. Prediction of body density in young and middle-aged men. J. Appl. Physiol. 40:300-304. [PubMed: 931840]
- Sale, D. G., J. D. MacDougall, S. E. Alway, and J. R. Sutton. 1987. Voluntary strength and muscle characteristics in untrained men and women and male body builders. J. Appl. Physiol. 62:1786-1793. [PubMed: 3597252]
- Schantz, P., E. Randall-Fox, W. Hutchison, A. Tyden, and P-O. Åstrand. 1983. Muscle fiber type distribution, muscle cross-sectional area, and maximum voluntary strength in humans. Acta Physiol. Scand. 117:219-226. [PubMed: 6223509]
- Segal, K. R., B. Gutin, E. Presta, J. Wang, and T. B. Van Itallie. 1985. Estimation of human body composition by electrical impedance methods: A comparative study. J. Appl. Physiol. 58:1565-1571. [PubMed: 3997721]
- Sinning, W. E., and C. Moore. 1989. A validation of the ponderal somatogram for evaluating muscularity. Med. Sci. Sports Exert. 21:S101.
- Smith, L. E., and J. Royce. 1963. Muscular strength in relation to body composition. Ann. N.Y. Acad. Sci. 110:809-813. [PubMed: 14061683]
- Tappen, N. C. 1950. An anthropometric and constitutional study of champion weight lifters. Am. J. Phys. Anthropol. (New Series) 8:49-64. [PubMed: 15410863]
- Tsunoda, N., S. Ikewaga, H. Yata, M. Kondo, H. Kanehisa, T. Fukunoga, and T. Asami. 1985. Structural and functional characteristics of thigh muscle in athletes. Pp. 15-20 in Biomechanics IX-A, D. A. Winter, editor. , ed. Champaign, Ill.: Human Kinetics Publishers.
- Watson, A. W. S., and D. J. O'Donovan. 1977. Factors relating to strength of male adolescents. Appl. Physiol. 43:834-838. [PubMed: 591477]
- Welham, W. C., and A. R. Behnke. 1942. The specific gravity of healthy men. J. Am. Med. Assoc. 118:498-501.
- White, R. M., and E. Churchill. 1971. The Body Size of Soldiers: U.S. Army Anthropometry—1966. Technical Report No. 72-51-CE. U.S. Army Natick Laboratories, Natick, Mass.
- Wilmore, J. H., and A. R. Behnke. 1968. Predictability of lean body weight through anthropometric assessment in college men. J. Appl. Physiol. 25:349-355.
- Wilmore, J. H., and A. R. Behnke. 1969. An anthropometric estimation of body density and lean body weight in young men. J. Appl. Physiol. 27:25-31. [PubMed: 5786965]
- Wilmore, J. H., and A. R. Behnke. 1970. An anthropometric estimation of body density and lean body weight in young women. Am. J. Clin. Nutr. 23:267-272. [PubMed: 5436635]
- Young, A., M. Stokes, and M. Crowe. 1982. The relationship between quadriceps size and strength in elderly women. Clin. Sci. 63:35P-36P.
Footnotes
- 1
Note that for the SOM concept, stature in decimeters replaces stature in meters in the calculation of F.
- PubMedLinks to PubMed
- New Approaches To Body Composition Evaluation And Some Relationships To Dynamic ...New Approaches To Body Composition Evaluation And Some Relationships To Dynamic Muscular Strength - Body Composition and Physical Performance
Your browsing activity is empty.
Activity recording is turned off.
See more...