The primary causes of death in the United States have changed over the last century from acute diseases to more chronic diseases; diseases that could be prevented by proactively controlling risk factors. Body composition is an important indicator for the increased risk of such chronic diseases like obesity, diabetes, and cardiovascular disease. Therefore, a large benefit lies in measuring one’s body composition because in addition to assessing risk of chronic disease, the loss of muscle mass is one of the biggest challenges elderly individuals face during aging. In this article, different techniques for measuring body composition will be discussed along with the best-suited techniques for different populations and the possible errors they may inherently possess in their calculations. In addition, this article will cover some nutritional options for controlling body composition within an athletic population.
- Discuss the importance of regularly assessing body composition.
- Describe different techniques used to measure body composition along with their strengths and weaknesses.
- Provide guidance for which assessment technique best fits different populations.
- Provide practical recommendations for how to control body composition in athletic populations.
The Importance of Body Composition
When comparing the top causes of mortality in today’s world to those over 100 years ago, it is easy to see how much modern medicine has developed. The top killers in 1900 were primarily acute diseases like pneumonia, influenza, tuberculosis, and other infections. Fast forward to present day, and those acute diseases are almost non-existent in terms of overall mortality rates, whereas today’s killers are from chronic conditions like heart disease, cancer, Alzheimer’s and diabetes (Jones, Podolsky, & Greene, 2012).
The top killers of today are closely related to the prevalence of obesity and the associations of body weight with the diagnosis of chronic diseases and excess mortality rates (Flegal, Graubard, Williamson, & Gail, 2005). The threat of cardiovascular disease (CVD) alone exponentially increases by simply possessing greater than one traditional CVD risk factor which includes: high blood pressure, high blood cholesterol, cigarette smoking, diabetes, and obesity (Folsom, Yamagishi, Hozawa, & Chambless, 2009). Obesity is directly linked to diabetes, hypertension, CVD, stroke, gallbladder disease, some cancers, osteoarthritis, and psychosocial problems and is a very prominent risk factor for premature death (Orpana et al., 2010). Clearly there is a need to accurately assess and monitor body weight to help avoid these chronic diseases.
The most common method of comparing body composition across large populations is by using the body mass index (BMI) calculation, which takes into account a person’s height and weight. The BMI categorizes individuals into “underweight” (BMI of <18.5), “normal weight” (BMI of 18.5-24.9), “overweight” (BMI of 25-29.9), or “obese” (BMI of >30). Relative to the normal weight category, underweight and obese individuals are positively associated with increased mortality, especially the greater the obese BMI score (Flegal, Graubard, Williamson, & Gail, 2005; Orpana et al., 2010). In terms of specific causes of mortality, underweight BMI scores are associated with increased mortality from non-CVD, non-cancer causes and long-term cardiovascular events (Romero-Corral et al., 2006), whereas obese BMI scores are associated with increased CVD, some cancers, diabetes, and kidney disease mortality rates, with stronger associations occurring in higher obese BMI scores. Overweight BMI scores, however, are not significantly associated with increased CVD mortality rates, but are associated with increased diabetes and kidney disease mortality (Flegal, Graubard, Williamson, & Gail, 2007).
Muscular strength or lack thereof, is also a risk factor for mortality in older adults. Sarcopenia is a major contributor to this decline in muscular strength in aging individuals (Goodpaster et al., 2006). Underweight BMI scores are commonly indicative of sarcopenia, which may decrease exercise capacity and mobility or impair the body’s ability to cope with stress and disease due to decreased muscle mass, which may be the root of the increased mortality rates (Romero-Corral et al., 2006; Malavolti et al., 2003). This can be misleading for someone who possibly scores in the “normal” range (BMI 18.5-24.9) because they might not appear overweight but their percent body fat could still be high, which corresponds to a low amount of lean body mass.
Differentiating Body Composition
The use of BMI to determine disease risk is of benefit, but it is inaccurate in detecting differences in body fat versus lean body mass (muscle mass, bones, organs, tissues, etc). It is also especially inaccurate at detecting changes in percent body fat for athletes (Silva, Fields, Quitério, & Sardinha, 2009) since it only takes into account height and weight; it does not determine what that weight is (Romero-Corral et al., 2006). For athletes, percent body fat can be highly related to performance in different sports and tracking it allows coaches to monitor any necessary changes in nutrition or training programs in order to achieve an optimal body composition (Moon et al., 2009).
Techniques for Measuring Body Composition
There are many different clinical and field techniques commonly used to determine body composition, yet these techniques still provide a substantial amount of differences in measurements between individuals. The true value of total body fat is actually immeasurable in vivo (Wang et al., 1998; Laforgia, Dollman, Dale, Withers, & Hill, 2009), and every technique has a standard margin of error. Four-compartment criterion models that divide the body into fat, water, bone mineral, and residual components provide the best references for criterion of any clinical or field method, particularly because they require fewer assumptions than two or three-compartment models (Laforgia, Dollman, Dale, Withers, & Hill, 2009; Minderico et al., 2008; Moon et al., 2009).
There are five and six-compartment models that exist, but they have shown minimal added accuracy in comparison to four-compartment models (Moon et al., 2009). The difficulty lies in the practical use of four-compartment models for everyday individuals, large populations, or for quick references if trying to track small changes in body composition throughout a training program. It is time-intensive and requires a lot of equipment that many universities, trainers and coaches do not possess. Much of their available equipment may rely on two-compartment models which have additional margins of error.
Clinical-based methods for determining body composition have increased in usage over the last decade. Many times researches have used methods such as dual-energy x-ray absorptiometry (DEXA) or under-water weighing (UWW) as criterion or ‘gold’ standards to compare other methods (Laforgia, Dollman, Dale, Withers, & Hill, 2009). It is suggested, however, that any criterion method should consist of greater than two compartments in their calculations (Moon et al., 2009) because there are still limitations in these ‘gold’ standards due to inherent errors based on assumptions that can cause large individual errors in estimating body fat (Wang et al., 1998).
DEXA scans divide the body into three compartments; fat, bone mineral and fat-free tissue with the assumption of a constant hydration status within that fat-free tissue (Wang et al., 1998). Several studies have shown that DEXA significantly underestimates body fat in leaner individuals and may overestimate fat in obese individuals when compared to a four-compartment model. This may be due to the inability of DEXA to accurately detect body fat in individuals with larger tissue thickness because of a phenomenon known as ‘beam hardening’ during the DEXA scan (Laforgia, Dollman, Dale, Withers, & Hill, 2009; Minderico et al., 2008). Another limitation of DEXA scans is that different densitometers and software versions will give different individual estimates of body composition (Malavolti et al., 2003).
UWW is a two-compartment model that separates the body into either body fat, or fat-free mass. It utilizes the known densities of each and assumes these values when calculating weight with gravity compared to weight under water. The problem lies with the fact that there is a wide range of inter-individual variability in the density of fat mass and fat-free mass which would obviously affect each measurement (Minderico et al., 2008).
One of the most practical clinical techniques for measuring body composition is probably through segmental bioelectrical impedance analysis (BIA). Most water in the body is found within muscle tissue making it a good conductor of electricity, whereas fat tissue contains almost no water, causing it to actually impede electrical flow (Benardot, 2012). BIA works through low-frequency electrodes placed on the hands and feet that detect the amount of total body water based on the resistance of each individual. It was found that the use of an eight-polar electrode BIA meter is accurate in estimating total body water (Bedogni et al., 2002), but since BIA is dependent on body water, the hydration status of the subject can alter the reading. If the subject is dehydrated the electrical current will not flow through lean tissue as well, therefore making it appear the subject has more fat mass (Benardot, 2012). Segmental BIA, like other techniques, still relies on assumptions that are partly age-dependent, but in a study with 110 individuals aged 21-82 years, BIA was as accurate in estimating body composition when compared to DEXA scans. The ease of operation combined with the rapidity of the measurement make it an attractive alternative for trainers and coaches compared to other more expensive and time-intensive methods (Malavolti et al., 2003).
There are many popular forms of anthropometric measurements used to track body composition. The most commonly used is the 3-site skinfold (men: chest, abdomen, thigh / women: tricep, abdomen, thigh) equations from Jackson & Pollock (1985), but there are some equations that use 7-sites, and others even use 10 or 12-sites in their calculations. For the sake of simplicity, there are even single-site calculations such as the Accu-Measure (one skinfold measurement taken one inch superior to the iliac crest) which has been proven to be just as accurate as a three-site equation in estimating body fat (Eckerson, Stout, Evetovich, Housh, Johnson, & Worrell, 1998).
Anytime a skinfold measurement is used to determine body fat, there is always an underlying biological error because of differences in individual fat distribution patterns between subcutaneous fat (the tissue that skinfold measurements use) and visceral fat (Silva, Fields, Quitério, & Sardinha, 2009). Also, most of the skinfold equations used to estimate body fat are derived from UWW, which is a two-compartment model that contains its own errors in assumptions of fat-free mass densities, hydration status, and bone mineral content (Moon et al., 2009). The low cost and simplicity of skinfold measurement make it an attractive way to track body composition, but results should be noted with caution. In a study which looked at 97 women who were partaking in an exercise program, both skinfold measurements and BIA underestimated small changes in body fat compared to DEXA post training (Sillanpää, Häkkinen, & Häkkinen, 2013).
Selecting techniques for different populations
Depending on the population being tested, various techniques may or may not have the best application. Although four-compartment models have proven to be the most accurate in determining body composition, it is not practical or realistic to assume everyone has access to such equipment. Due to the standard error of estimation within any technique for determining body composition and due to the high amount of individual variance of testing results, different techniques should not be used interchangeably to detect changes in body composition (Minderico et al., 2008). Therefore, whatever technique is chosen initially to determine body composition should also be used as the follow up technique to reassess progress.
Some techniques may be best suited for a specific population in terms of practicality and accuracy, while other techniques may not be favorable. For example, the Accu-Measure single-site measurement was proven to be quicker and just as accurate as a three-site calculation, but it is important to note the sample population used for the study was only Caucasian collegiate students, which is most likely a statistically leaner population in comparison to a middle-aged sample. If someone has more visceral fat, it would be hard to accurately estimate body fat due to the fact that a single-site location only measures subcutaneous fat in one region of the body. In a study of obese children who partook in an exercise program, changes in abdominal skinfolds were not accurate in predicting changes in total fat mass when compared to DEXA scans and the researchers questioned the validity of using skinfold measurements on obese or elderly populations (Watts et al., 2006).
However, there are even inaccuracies in the ‘gold’ standard methods as well. In a study detecting small changes in body composition in elite male judo athletes, they compared DEXA scans to a 4-compartment reference model (taking into account fat-free mass, bone mineral density, body volume, and total body water) and found that the DEXA scan had large and unexpected individual errors in tracking physiological changes in the athletes. The researchers concluded that DEXA may not be accurate for athletes needing to achieve a specific target weight or body composition for their competitions (Santos, Silva, Matias, Fields, Heymsfield, & Sardinha, 2010).
Inaccuracies in athletic populations may be due to the fact that some elite athletes possess atypical body compositions. Due to intense physical training and the extreme physiques required for various sports, athletes have been shown to have systemic deviations in the densities of fat-free mass (Prior et al., 2001). Two-compartment models are able to produce a low value of total error when compared to a multi-compartment model, but due to the large limits of agreement, two-compartment models may not be the most appropriate for athletic populations to accurately track changes in body composition during diet and exercise interventions due to individual variations in fat-free mass density and total body hydration (Moon et al., 2009). However, in collegiate female athletes, it was shown that skinfold equations can provide accurate estimations of body fat when compared to a four-compartment model and the use of a seven-site equation did not provide any better accuracy compared to a three-site equation. The researchers concluded that trainers and coaches should continue the use of three-site equations in this type of population, but individual errors may be as high as +/- 5.34% body fat (Moon et al., 2009).
Practical Applications: Controlling Body Composition
In sports that include weight class divisions or emphasize leanness, and for athletes who wish to improve their power to weight ratio (reducing overall body mass without compromising their performance), a low-carbohydrate, ketogenic diet may be suitable. A study investigating the affects of a ketogenic diet for 30 days on elite gymnasts consisted of a diet in which they ate 54.8% fat, 40.7% protein, and 4.5% carbohydrates (<28 grams/day). They continued their normal training program for the period of the study, and showed no loss in strength performance, while significantly reducing their fat mass, with an insignificant increase in muscle mass (Paoli et al., 2012).
Rapid weight loss through dehydration, fasting, vomiting, or even laxative use can all negatively impact sport performances, and should be avoided (Garthe, Raastad, & Sundgot-Borgen, 2011). Healthy weight reduction has typically been recommended at a rate of a 0.5-1.0 kg /week. In a study comparing the long-term effects of two different weight-reduction protocols on different elite athletes using either a slow-reduction (decrease of 0.7% body mass/week) or a fast-reduction (decrease of 1.4% body mass/week) of weight, the slow-reduction group significantly increased lean body mass from baseline to post-intervention, and tended to maintain strength (determined by 1 repetition-maximum tests) better than the fast-reduction group, who demonstrated no changes in lean body mass (Garthe, Raastad, & Sundgot-Borgen, 2011). Interestingly enough, after six and 12 month follow up testing; there were no significant differences in either group, suggesting maintaining a new body mass is difficult, and many different factors other than the rate of weight loss contribute to the maintenance of body composition.
Mechanisms that regulate total energy expenditure, body composition and eating behavior all impact body mass, along with the endocrine system and its regulatory hormones that stimulate greater food intake after a period of reduced energy consumption. For example, when obese subjects were put on extremely low-caloric diets to lose weight, the subjects did lose weight initially but the extreme calorie restriction caused an imbalance in their metabolic hormones; Among them, leptin which is responsible for inhibiting appetite was decreased, while gherlin and gastric inhibitory polypeptide, which are both responsible for increasing appetite, were both elevated above baseline for up to 12 months after their initial weight loss and resulted in many subjects regaining weight back over those months (Sumithran et al., 2011). This may be why the athletes in the Garthe, Raastad, & Sundgot-Borgen (2011) study noted that it was very difficult to maintain their new body mass for a prolonged period and therefore were back at their baseline weight after 12 months. Nevertheless, consistently tracking body composition with the same methods is the best way to detect change over time.
The importance of determining an individual’s body composition cannot be overstated. Specific BMI scores and body compositions significantly increase the risk of chronic diseases, which can affect quality of life or even attribute to an early death. Although BMI scores provide generic information in regards to CVD risk, there are errors in the fact that it only takes into account height and weight, therefore it is important to determine the difference in that weight by measuring % body fat.
There are many different techniques used to determine % body fat. A four-compartment model provides the most accurate results and should therefore be used as a criterion for other methods. Individuals may have limited availability for the most accurate testing techniques but the most practical technique to determine % body fat in terms of cost and ease of operation in a clinical setting is the BIA for most populations and in a field setting, skinfold measurements have proven to be accurate specifically for more athletic populations.
There is inherent error within any chosen technique for determining body composition, therefore different techniques should not be used interchangeably when attempting to track changes in body composition. Once a technique is chosen, it is advised to stay with that technique to appropriately measure changes.
Finally, if athletes wish to improve their body composition without sacrificing strength and power, a low-carbohydrate, ketogenic diet may prove beneficial in improving their overall power to weight ratio.
Bedogni, G., Malavolti, M., Severi, S., Poli, M., Mussi, C., Fantuzzi, A., & Battistini, N. (2002). Accuracy of an eight-point tactile-electrode impedance method in the assessment of total body water. European Journal Of Clinical Nutrition, 56(11), 1143-1148.
Benardot, D. (2012). Advanced sports nutrition (2nd ed.). Champaign, IL: Human Kinetics.
Björntorp, P. (1997). Hormonal control of regional fat distribution. Human Reproduction (Oxford, England), 12 Suppl 121-25.
Eckerson, J., Stout, J., Evetovich, T., Housh, T., Johnson, G., & Worrell, N. (1998). Validity of self-assessment techniques for estimating percent fat in men and women. Journal Of Strength & Conditioning Research, 12(4), 243-247.
Flegal, K., Graubard, B., Williamson, D., & Gail, M. (2005). Excess deaths associated with underweight, overweight, and obesity. JAMA: Journal Of The American Medical Association, 293(15), 1861-1867.
Flegal, K., Graubard, B., Williamson, D., & Gail, M. (2007). Cause-specific excess deaths associated with underweight, overweight, and obesity. JAMA: Journal Of The American Medical Association, 298(17), 2028-2037.
Folsom, A., Yamagishi, K., Hozawa, A., & Chambless, L. (2009). Absolute and attributable risks of heart failure incidence in relation to optimal risk factors. Circulation. Heart Failure, 2(1), 11-17.
Garthe, I., Raastad, T., & Sundgot-Borgen, J. (2011). Long-term effect of weight loss on body composition and performance in elite athletes. International Journal Of Sport Nutrition & Exercise Metabolism, 21(5), 426-436.
Goodpaster, B., Park, S., Harris, T., Kritchevsky, S., Nevitt, M., Schwartz, A., & ... Newman, A. (2006). The loss of skeletal muscle strength, mass, and quality in older adults: the health, aging and body composition study. Journals Of Gerontology Series A: Biological Sciences & Medical Sciences, 61A(10), 1059-1064.
Jackson, A. S., & Pollock, M. L. (1985). Practical assessment of body composition. Phys Sport Medicine, 13, 76-90.
Jones, D., Podolsky, S., & Greene, J. (2012). The burden of disease and the changing task of medicine. New England Journal Of Medicine, 366(25), 2333-2338.
Laforgia, J., Dollman, J., Dale, M., Withers, R., & Hill, A. (2009). Validation of DXA Body Composition Estimates in Obese Men and Women. Obesity (19307381), 17(4), 821-826.
Malavolti, M. M., Mussi, C. C., Poli, M. M., Fantuzzi, A. L., Salvioli, G. G., Battistini, N. N., & Bedogni, G. G. (2003). Cross-calibration of eight-polar bioelectrical impedance analysis versus dual-energy X-ray absorptiometry for the assessment of total and appendicular body composition in healthy subjects aged 21-82 years. Annals Of Human Biology, 30(4), 380-391.
Minderico, C., Silva, A., Keller, K., Branco, T., Martins, S., Palmeira, A., & Sardinha, L. (2008). Usefulness of different techniques for measuring body composition changes during weight loss in overweight and obese women. British Journal Of Nutrition, 99(2), 432-441.
Moon, J. R., Eckerson, J. M., Tobkin, S. E., Smith, A. E., Lockwood, C. M., Walter, A. A., & ... Stout, J. R. (2009). Estimating body fat in NCAA Division I female athletes: a five-compartment model validation of laboratory methods. European Journal Of Applied Physiology, 105(1), 119-130.
Moon, J., Tobkin, S., Smith, A., Lockwood, C., Walter, A., Cramer, J., & ... Stout, J. (2009). Anthropometric estimations of percent body fat in NCAA Division I female athletes: a 4-compartment model validation. Journal Of Strength & Conditioning Research, 23(4), 1068-1076.
Orpana, H., Berthelot, J., Kaplan, M., Feeny, D., McFarland, B., & Ross, N. (2010). BMI and mortality: results from a national longitudinal study of Canadian adults. Obesity, 18(1), 214-218.
Paoli, A., Grimaldi, K., D'Agostino, D., Cenci, L., Moro, T., Bianco, A., & Palma, A. (2012). Ketogenic diet does not affect strength performance in elite artistic gymnasts. Journal Of The International Society Of Sports Nutrition, 9(1), 34-42.
Prior, B. M., Modlesky, C. M., Evans, E. M., Sloniger, M. A., Saunders, M. J., Lewis, R. D., & Cureton, K. J. (2001). Muscularity and the density of the fat-free mass in athletes. Journal Of Applied Physiology, 90(4) 1523-1531.
Romero-Corral, A., Montori, V., Somers, V., Korinek, J., Thomas, R., Allison, T., & ... Lopez-Jimenez, F. (2006). Association of bodyweight with total mortality and with cardiovascular events in coronary artery disease: a systematic review of cohort studies. Lancet, 368(9536), 666-678.
Santos, D., Silva, A., Matias, C., Fields, D., Heymsfield, S., & Sardinha, L. (2010). Accuracy of DXA in estimating body composition changes in elite athletes using a four compartment model as the reference method. Nutrition & Metabolism, (7) 22.
Sillanpää, E., Häkkinen, A., & Häkkinen, K. (2013). Body composition changes by DXA, BIA and skinfolds during exercise training in women. European Journal Of Applied Physiology, 113(9), 2331-2341.
Silva, A., Fields, D., Quitério, A., & Sardinha, L. (2009). Are skinfold-based models accurate and suitable for assessing changes in body composition in highly trained athletes?. Journal Of Strength & Conditioning Research, 23(6), 1688-1696.
Sumithran, P., Prendergast, L., Delbridge, E., Purcell, K., Shulkes, A., Kriketos, A., & Proietto, J. (2011). Long-term persistence of hormonal adaptations to weight loss. New England Journal of Medicine, 365(17), 1597-1604.
Wang, Z., Deurenberg, P., Guo, S., Pietrobelli, A., Wang, J., Pierson, R., & Heymsfield, S. (1998). Six-compartment body composition model: inter-method comparisons of total body fat measurement. International Journal Of Obesity And Related Metabolic Disorders: Journal Of The International Association For The Study Of Obesity, 22(4), 329-337.
Watts, K., Naylor, L. H., Davis, E. A., Jones, T. W., Beeson, B., Bettenay, F., Siafurikas, A., Bell, L., Ackland, T., & Green, D. J. (2006). Do skinfolds accurately assess changes in body fat in obese children and adolescents? Medicine & Science In Sports & Exercise, 38, 439-444.