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.

Cover of Body Composition and Physical Performance

Body Composition and Physical Performance: Applications For the Military Services.

Show details

4Body Composition In The Military Services: Standards And Methods

James A. Hodgdon


This paper will discuss two topics: the development of standards for body composition in the U.S. Navy and the methods of body composition assessment in use by the military services today.

In 1981, the Department of Defense (DOD) issued directive 1308.1 (DOD directive 1308.1, 1981). Part of the policy expressed in that directive was that the ''determining factor in deciding whether or not a service member is overweight is the member's percent body fat." (DOD Directive 1308.1, p. 2 Encl. 2, 1981). The military services were directed to determine body composition and fat standards consistent with the mission of the services.

The directive also indicated that there are three concerns relating to the need for establishing a weight control policy: first, body composition is an integral part of physical fitness and is, therefore, essential for maintaining combat readiness. This statement implies a relationship between fatness and military performance. Second, control of body fat (BF) is necessary to maintain appropriate military appearance. Third, control of BF is important in maintaining the general health and well-being of armed forces personnel.

The directive left the task of developing the most appropriate methodology for BF determination to the individual services. The directive required that fat measurement techniques must have a correlation coefficient of 0.75 or better with percent BF from underwater weighing. This coefficient has since been increased to 0.85. DOD percent BF goals were set at 20 percent BF for men and 26 percent BF for women.

Body Composition Standards

If body composition was presumed to affect military performance, military appearance, and general health and well-being, the basis for setting standards ought to lie with one of these three relationships. Below is the line of argument followed within the U.S. Navy to arrive at suitable standards for body composition.

Body Composition And Physical Performance

Performance on the U.S. Navy's biannual Physical Readiness Test (PRT) is taken to be an indicator of a sailor's readiness for combat. As an adjunct to setting standards for physical fitness and body composition, studies were carried out that investigated relationships between performance on the PRT items and performance of materials handling tasks. The Navy's PRT includes a body composition assessment, sit-reach distance, time for a 1.5-mile run, number of sit-ups performed in 2 minutes, and number of push-ups performed in 2 minutes. Work by Robertson and Trent (1985) at the Navy Personnel Research and Development Center showed that the majority of the physically demanding jobs performed by Navy personnel were materials handling tasks: lifting, carrying, and pulling, with the most common being carrying while walking (48 percent) and lifting without carrying (20 percent). Performance on such tasks might form a reasonable basis for setting standards for shipboard work.

Beckett and Hodgdon (1987) investigated associations between PRT items, body composition variables, and performance on two materials handling tasks. The two tasks were: the maximum weight of a box that could be lifted to elbow height (box-lift maximum weight) and the total distance a 34-kg box could be carried (box carry power) on alternate laps of a 51.4-m course during two 5-minute work bouts. The parameters of the carry task represented median values of the weight, distance, and timing of Robertson and Trent's survey of carry tasks performed aboard ship. Table 4-1 shows the correlations between PRT and body composition items, and performance on the lift and carry.

Table 4-1. Correlations, U.S. Navy Physical Readiness Test Items, and Body Composition with Materials Handling Tasks.

Table 4-1

Correlations, U.S. Navy Physical Readiness Test Items, and Body Composition with Materials Handling Tasks.

Table 4-1 shows percent BF to be only modestly correlated with these materials handling tasks. These modest correlations suggest that using relationships between these tasks and percent BF as the basis of setting percent BF standards would not be particularly fruitful. However, it might be noted that one of the body composition variables (fat-free mass [FFM]) is highly correlated with the box-lift maximum weight. In this study, FFM was also found to be highly correlated with other muscle strength measures. The possibility exists for using FFM as an approximation of overall strength in job assignment.

Body Composition And Appearance

The second stated reason for maintaining appropriate levels of BF is for proper military appearance. It is the Navy's policy that judgments about appearance are subjective and not necessarily strongly related to fatness. Current performance evaluation procedures allow for these subjective assessments, and they need not be anchored to other objective variables.

The soundness of this approach was recently tested by Hodgdon and colleagues (1990). A panel of 11 U.S. Army headquarters staff (5 women, 6 men; 6 officers, 5 enlisted; and including both Black and White members) rated the "military appearance" of 1,075 male and 251 female U.S. Army personnel dressed in Class A uniform. Physical characteristics of this population sample are provided in Table 4-2. A 5-point scale was used for the ratings. In this scale, a value of 1 was labeled "poor"; a value of 2, "fair"; a value of 3, "good''; a value of 4, ''very good"; and a value of 5, "excellent". The raters were instructed to rate the "military appearance" of the soldier according to their own personal standards, and instructed to evaluate how the individual looked in uniform, not how the uniform looked. The personnel who were rated also had their percent BF determined from underwater weighing. The inter-rater reliability of the ratings was quite good (alpha = 0.86 for rating of men, 0.87 for rating of women). The results of the regression analysis to predict measured percent BF from the ratings of appearance are provided in Table 4-3. The correlation between appearance ratings and percent fat was modest: 0.53 for ratings of male personnel, and 0.46 for ratings of female personnel. The square of the correlation coefficient indicates the percent of the total variance in one variable accounted for by the other. Percent fat accounts for only 28 percent of the measured variance in appearance for men, only 22 percent for women. It does not appear from this study that percent BF, by itself, constitutes a reasonable indicator of military appearance. Clearly, other factors play a role in such judgments.

Table 4-3. Prediction of Appearance from Percent Fat Scores in U.S. Army Personnel Appearance Study.

Table 4-3

Prediction of Appearance from Percent Fat Scores in U.S. Army Personnel Appearance Study.

Body Composition And Health

The DOD directive points out that one of the reasons for wanting to set BF standards is the maintenance of health and well-being of the service members. It is in the relationship between health and fatness that the Navy has anchored its body composition standards.

On February 11-13, 1985, the National Institutes of Health (NIH) Office of Medical Applications of Research; the National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases; and the National Heart, Lung, and Blood Institute convened a consensus development conference on the health implications of obesity (National Institutes of Health, 1985). The conferees determined that obesity is related to a significant impairment of health, particularly in terms of increased risk of diabetes, hypertension, coronary artery heart disease, and cancer. They also agreed that obesity could be defined as a weight-for-height 20 percent above the midpoint weight listed in the 1983 Metropolitan Life Insurance tables for the medium-frame individual (Metropolitan Life Insurance Company, 1984).

Armed with this definition, and the information that obesity could be considered a health risk, the following study determined whether or not these weight-for-height tables had any reasonable expression in percent BF. Using the Navy anthropometry data set, the regression between weight and height and percent BF was determined. Table 4-4 describes the data set used for development of the regressions.

Table 4-4. Regression Sample Descriptions.

Table 4-4

Regression Sample Descriptions.

The regressions that were developed were:

Image img00006.jpg


Image img00007.jpg

where R = multiple correlation coefficient and SEE = standard error of the estimate.

Using these equations, we then determined the percent BF value associated with the NIH critical weights at each height for both men and women. The results are provided graphically in Figure 4-1. As can be seen from Figure 4-1, the "critical" percent BF values are rather constant across heights, especially the values for women. Mean values for critical percent BF across height were 22.0 ± 1.20 for men and 33.5 ± 0.18 for women.

Figure 4-1. National Institutes of Health critical weights for each height expressed as percent fat.

Figure 4-1

National Institutes of Health critical weights for each height expressed as percent fat.

Standards for percent BF for Navy personnel were derived from these mean values. The circumference equations used by the U.S. Navy to predict BF have standard errors of measurement of approximately 3.5 percent BF. It was decided that the standard for administrative action should be approximately one standard error above the critical percent BF to minimize the number of false positives for individuals exceeding the NIH obesity definition. The values of 26 percent BF for men and 36 percent BF for women were thus adopted. Any sailor or officer exceeding these limits for three successive administrations of the PRT is subject to administrative action. In addition, an "overfat" category was defined. Individuals exceeding values of 22 percent BF, if men, or 30 percent BF if women, are required to go on a fat reduction program. This approach allows remedial action on BF reduction to begin prior to exceeding the limits for administrative action.

The finding that the NIH critical weights represent a relatively constant percent BF for men and women is intriguing, especially when one considers that those weights derive from the empirically determined Metropolitan Life Insurance Tables (Metropolitan Life Insurance Company, 1984). However, there is a paucity of data relating body composition variables themselves to mortality and morbidity outcomes. Such epidemiological studies need to be done.

In summary, the U.S. Navy, finding a lack of basis for setting body composition standards based on either performance or appearance, has chosen to base its standards on health considerations. The standards are derived from the NIH consensus definition of obesity.

Body Composition Measurement

The criteria for selecting methods for assessment of body composition in the military were that the measures must be: usable easily in the field, able to be made reliably, and be valid indicators of fatness. It was also important that skill in measurement be relatively easily acquired. To meet these measurement technique requirements, all four services have adopted circumference measurements, often in conjunction with height and weight, as the basis for predicting percent BF.

Reliability And Trainability

In 1987, Mueller and Malina determined intra- and interexaminer reliabilities of skinfold and circumference measurements. They found both techniques to be quite reliable but circumferences to be more reliably measured than skinfold thicknesses (0.97 and 0.96 for circumference intra-and interexaminer reliabilities, respectively, and 0.94 and 0.92 for skinfold reliabilities).

In addition to being slightly more reliably made, circumference measurements appear to be more easily learned. J. H. Heaney and coworkers (Naval Health Research Center, San Diego, unpublished manuscript) investigated the time course for acquiring skill in circumference and skinfold thickness measurement. Thirty-eight active duty Navy personnel were provided six 1-hour training sessions during which they were trained and evaluated in skinfold measurements at two sites and circumference measurements at three sites. Heaney and coworkers found that after 75 skinfold measurements at each site (150 total measurements), only 24 percent of the study participants had reached proficiency in skinfold measurement. In contrast, 68 percent of the participants had reached proficiency after 45 circumference measurements at each site (135 total measurements). In this study, circumference measurement was clearly the more easily learned technique.

Equation Validity

Each of the services developed regression equations involving body circumference measurements, sometimes in conjunction with height or weight or both. The regression equations predict either body density, percent BF, or FFM. For the U.S. Army, Navy, and Marine Corps, the criterion measurement for equation development was either body density from underwater weighing or percent BF using the Siri (1961) equation to convert body density to percent BF. The U.S. Air Force equations use as a criterion measure FFM determined from tritiated water dilution or from body volume and weight (Allen, 1963). Tables 4-5a and 4-5b contain the equations and descriptive data from the equation development samples for the military services. Sample descriptors are shown as mean plus or minus standard deviation.

Table 4-5a. Body Composition Equations for the Military Services.

Table 4-5a

Body Composition Equations for the Military Services.

Table 4-5b. Body Composition Development Data for the Military Services.

Table 4-5b

Body Composition Development Data for the Military Services.

U.S. Army

The U.S. Army equations were developed by Vogel and coworkers (1988) at the U.S. Army Research Institute of Environmental Medicine on a large sample of Army personnel. The sample was not stratified to reflect distributions of demographic variables (for example, age, gender, race, job classification) within the Army population.

These equations are used in conjunction with weight-for-height tables that serve as an initial screening tool in detecting overfat. Current Army BF retention standards are based on age (AR 600-9, 1986). Standards for men are 20 percent BF for ages 16-20 years, 22 percent BF for ages 21-27 years, 24 percent BF for ages 28-39 years, and 26 percent BF for ages 40 years and older. Standards for women are 28 percent, 30 percent, 32 percent, and 34 percent BF respectively, for the same age groupings as the men.

U.S. Navy

The U.S. Navy equations were developed by Hodgdon and Beckett (1984a,b) at the Naval Health Research Center. Their large sample of U.S. Navy personnel was also non stratified with respect to Navy demographics. Within the Navy every service member has his or her BF estimated twice each year using these equations (U.S. Department of the Navy, 1986a). There are no weight-for-height screening tables used. As noted above, the current retention standards are 26 percent BF for men and 36 percent BF for women, irrespective of age.

U.S. Marine Corps

The Marine Corps was the first service to use body composition estimation from circumferences. The Marine Corps equations were developed by Wright and coworkers (1980, 1981) of the Institute of Human Performance from data collected by Wright and Wilmore (1974) on Marine Corps personnel. The Marine Corps uses weight-for-height tables as the basis for weight control decisions (U.S. Department of the Navy, 1986b). If a Marine is overweight by the tables but does not appear to be fat, he or she may have a BF estimation done. If the individual's BF is less than the Marine Corps standards of 18 percent BF for men and 26 percent BF for women, a new maximum allowable weight is calculated and entered into the Marine's record.

U.S. Air Force

The U.S. Air Force body composition equation for men was developed by Fuchs and coworkers (1978) at the U.S. Air Force School of Aerospace Medicine. The equation for women was developed by Brennan (1974) as part of her master's work at the Incarnate Word College in San Antonio. Unlike the equations of the other services, the U.S. Air Force equations predict FFM. Also, the development sample for the women's equation contained some non-service personnel.

Like the Marine Corps, the U.S. Air Force has a weight-for-height standard (AFR 35-11, 1985). Individuals whose weight exceeds the standard will have their body composition determined. If they do not exceed the U.S. Air Force BF standards (20 percent BF for men less than 30 years of age, 26 percent BF for men older than 30 years; 28 percent BF for women less than 30 years, 34 percent BF for women older than 30 years), new maximum allowable weight can be assigned. Similarly, individuals whose weight does not exceed the standard, but who appear obese, can have a new allowable weight assigned based on BF measurement.


To provide a basis for comparing the performance of these equations on a general military population, each of the equations was cross-validated on the Navy anthropometric sample described in Table 4-2. If the equation did not predict percent BF directly, the equation output was converted to percent BF. Predicted percent BF was correlated with percent BF derived from underwater weighing using the Siri (1961) equation. Table 4-4 shows the results of this cross-validation. Note the U.S. Air Force equation for men is only cross-validated on a subset of the U.S. Navy sample. This is because flexed biceps measurements were only made on a few of the Navy subjects.

Table 4-2. Participant Characteristics in U.S. Army Personnel Appearance Study.

Table 4-2

Participant Characteristics in U.S. Army Personnel Appearance Study.

It is apparent from Table 4-4 that predicted BF was rather highly correlated with hydrostatic BF in all of the equations. More importantly, the standard errors of measurement seen here with these equations are comparable to those seen with other generalized equations in common use, including those using skinfolds (Durnin and Womersley, 1974; Jackson and Pollack, 1978; Jackson et al., 1980). Hodgdon and Beckett (1984a,b) and Wright et al. (1980) have already shown that generalized circumference and skinfold equations have similar validities when applied to these military population samples.

Table 4-6Cross-Validation of Military Equations, U.S. Navy Sample

Correlation CoefficientMean Difference (percent fat)Standard Error of Measurement (percent fat)
U.S. Army
U.S. Navy
U.S. Marine Corps
U.S. Air Force

Cross-validation on only 52 Navy subjects.


Two major summary points can be made: first, there is admittedly a need to further validate the relationship between body composition and health outcomes. However, as evidenced by the studies presented here, it would appear at present that health considerations are the most rational scientific basis for setting body composition standards. Second, the military services have used standard techniques to derive equations to estimate relative BF from anthropometric measures: body circumferences, height, and weight. When applied to a general military population sample, these equations have validities and standard errors of measurement similar to other published, generalized anthropometric equations and would appear to be reasonable, useful estimators of body composition.


  • AFR 35-11, 1985. See U.S. Department of the Air Force. 1985.
  • AR 600-9, 1986. See U.S. Department of the Army. 1986.
  • DOD Directive 1308.1, 1981. See U.S. Department of Defense. 1981.
  • Allen, T. H. 1963. Measurement of human body fat: A quantitative method suited for use by aviation medical officers. Aerospace Med. 34:907.
  • Beckett, M. B., and J. A. Hodgdon. 1987. Lifting and carrying capacities relative to physical fitness measures. Report No. 87-26. Naval Health Research Center, San Diego, Calif.
  • Brennan, E. H. 1974. Development of a binomial involving anthropometric measurements for predicting lean mass in young women. M.S. thesis. Incarnate Word College, San Antonio, Tex.
  • Durnin, J. V. G. A., and J. Womersley. 1974. Body fat assessed from total body density and its estimation from skinfold thickness: Measurements on 481 men and women aged 16 to 72 years. Br. J. Nutr. 32:77-97. [PubMed: 4843734]
  • Fuchs, R. J., C. F. Theis, and M. C. Lancaster. 1978. A homogram to predict lean body mass in men. Am. J. Clin. Nutr. 31:673-678. [PubMed: 637039]
  • Hodgdon, J. A., and M. B. Beckett. 1984. a. Prediction of percent body fat for U.S. Navy men from body circumferences and height. Report No. 84-11. Naval Health Research Center, San Diego, Calif.
  • Hodgdon, J. A., and M. B. Beckett. 1984. b. Prediction of percent body fat for U.S. Navy women from body circumferences and height. Report No. 84-29. Naval Health Research Center, San Diego, Calif.
  • Hodgdon, J. A., P. I. Fitzgerald, and J. A. Vogel. 1990. Relationships between body fat and appearance ratings of U.S. Soldiers. Report No. 90-01. Naval Health Research Center, San Diego, Calif.
  • Jackson, A. S., and M. L. Pollack. 1978. Generalized equations for predicting body density of men. Br. J. Nutr. 40:497-504. [PubMed: 718832]
  • Jackson, A. S., M. L. Pollack, and A. Ward. 1980. Generalized equations for predicting body density of women. Med. Sci. Sport Exerc. 12:175-182. [PubMed: 7402053]
  • Metropolitan Life Insurance Company. 1984. 1983 Metropolitan height and weight tables. Statistical Bulletin of the Metropolitan Life Insurance Company 64:2-9.
  • Mueller, W. H., and R. M. Malina. 1987. Relative reliability of circumferences and skinfolds as measures of body fat distribution. Am. J. Phys. Anthropol. 72:437-439. [PubMed: 3605318]
  • National Institutes of Health. 1985. Health implications of obesity. National Institutes of Health Consensus Development Conference Statement, vol. 5, no. 9. U.S. Department of Health and Human Services, Washington, D.C.
  • Robertson, D. W., and T. T. Trent. 1985. Documentation of muscularly demanding job tasks and validation of an occupational strength test battery (STB). Report No. 86-1. Naval Personnel Research and Development Center, San Diego, Calif.
  • Siri, W. E. 1961. Body composition from fluid spaces and density: Analysis of methods. Pp. 223-244 in Techniques for Measuring Body Composition, J. Brozek, editor; and A. Henschel, editor. , eds. Washington, D.C: National Academy of Sciences.
  • U.S. Department of the Army. 1986. Army Regulation 600-9. ''The Army Weight Control Program.'' September 1. Washington, D.C.
  • U.S. Department of the Air Force. 1985. Regulation 35-11. "The Air Force Weight and Fitness Programs." April 10.
  • U.S. Department of Defense. 1981. "Physical Fitness and Weight Control Programs." Directive No. 1308.1. June 29. Washington, D.C.
  • U.S. Department of the Navy, Navy Military Personnel Command, Code 6H. 1986. a. "Physical Readiness Program." Office of the Chief of Naval Operations Instruction; 6110.1C. August 7.
  • U.S. Department of the Navy, Headquarters Marine Corps, Training Department. 1986. b. "Weight Control and Military Appearance." Marine Corps Order 6100.10a. July 24.
  • Vogel, J. A., J. W. Kirkpatrick, P. I. Fitzgerald, J. A. Hodgdon, and E. A. Harman. 1988. Derivation of anthropometry based body fat equations for the Army's weight control program. Technical Report No. T17-88. U.S. Army Research Institute of Environmental Medicine, Natick, Mass.
  • Wright, H. F., and J. H. Wilmore. 1974. Estimation of relative body fat and lean body weight in a United States Marine Corps population. Aerospace Med. 45:301-306. [PubMed: 4816552]
  • Wright, H. F., C. O. Dotson, and P.O. Davis. 1980. An investigation of assessment techniques for body composition of women marines. U.S. Navy Med. 71:15-26.
  • Wright, H. F., C. O. Dotson, and P.O. Davis. 1981. Simple technique for measurement of percent body fat in man. U.S. Navy Med. 72:23-27. [PubMed: 6942599]
Copyright 1992 by the National Academy of Sciences.
Bookshelf ID: NBK235939


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (4.3M)

Related information

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...