PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Obesity (Silver Spring). Author manuscript; available in PMC 2017 May 1.
Published in final edited form as:
PMCID: PMC4846565
NIHMSID: NIHMS760371

Measuring Body Composition in Low-Resource Settings across the Life Course

Abstract

We explore recent advances in the field of body composition measurement that could be suitable for use in low-resource settings across the life-course. Our aim was three-fold: (i) to review the available literature and information on both current and novel technologies for body composition measurement, (ii) to present a decision schema that may assist in selecting the appropriate body composition technology, and (iii) which of the technologies available are suitable for low-resource settings based on cost, infrastructure needed, participant compliance needed for the measurement, quality assurance protocols in place, safety, accuracy of measurement and training required.

Keywords: body composition, measurement technology, adiposity, low-resource settings, decision tree

Introduction

Since World War II, dramatic lifestyle changes have altered eating habits and activity levels worldwide, driving an increase in obesity for many low- or middle-income countries (LMICs)(1). At the same time, protein-calorie malnutrition is still prevalent in some LMICs and can occur in combination with excess adiposity, a so-called double burden (2). African countries are illustrative of the demographic, economic and nutrition transitions LMICs are undergoing. In recent assessments, 66% of African countries exhibit a high to very high prevalence of stunting (≥ 30%)(3). Obesity is also significantly increasing in Africa, particularly in women, with a prevalence of 40% or more in South Africa and Egypt. With rising obesity, metabolic diseases such as type 2 diabetes are becoming more prevalent, mirroring rates in high-income countries (4).

There is a critical need for accurate quantitative markers of nutritional status appropriate for low-resource settings for both public health research and diagnostic and monitoring purposes. Anthropometric criteria (e.g., waist circumference or body mass index) have poor sensitivity for monitoring response to treatment and may not adequately reflect body composition, and the relationship of anthropometric markers at birth to adult body composition differs by region (5, 6). Hence, direct body composition markers could improve clinical management.

This report reviews literature and information on current and novel technologies for measuring body composition and presents a decision schema for selecting the most appropriate technology suitable for low-resource settings across the life-course. Our criteria include cost, infrastructure, participant compliance, quality assurance, safety, accuracy of measurement and device-specific training required. Our findings are presented in the context of recent advances in the field of body composition technologies suitable for use in low-resource settings.

Models and their applications in low-resource settings

There are many compartments within the human body of importance with respect to the evaluation and monitoring of nutritional status, including underweight syndromes and obesity. These compartments are distributed across five levels: atomic, molecular, cellular, tissue-organ, and whole-body (7). Each level includes multiple compartments, the sum of which is body mass. The molecular level is the most widely used in nutrition research, with major compartments including total body water, protein, minerals and fat. Many body composition methods provide estimates of fat mass (FM) and fat-free mass (FFM). There are no “direct” methods of quantifying body compartments in living humans (i.e., weighing the amount of body fat). Body composition models and equations allow “indirect” estimation of a compartment through measurement of some physical property such as body density or electrical impedance.

Known and relatively stable relationships exist between compartments at the five body composition levels (7). This information is leveraged to create models in which these relationships or “constants” are used to derive an “unknown” compartment of interest. Over the life-course, there are relative body composition changes that reach a steady-state in adulthood. This process involves a relative expansion of cell mass with reductions in extracellular fluid and follows a predictable course. Published tables and charts provide age- and sex-related model constants for body composition-estimating equations. Well-documented relative changes in body composition also occur during pregnancy. A need exists to develop standardized model coefficients so that body composition data can be effectively compared across varied resource settings. A critical property of these coefficients is that they account for potential effects of age, gender and race/ethnicity on the outcome variable of interest.

While “mechanistic” models form the basis of most body composition research methods, there are also “statistical” models for predicting unknown compartments (8). Mechanistic models are developed using an underling theory (e.g., the first law of thermodynamics) as their basis. The statistical approach involves use of a “reference” method for estimating a compartment of interest. A measured body property is then calibrated against this compartment with statistically-derived equations. No underlying physical or biological theories are used to create these models.

Both mechanistic and statistical models are used in low-resource settings. A mechanistic model example is the use of measured deuterium dilution space to estimate total body fat. Here we have one “constant” describing the overestimate of total body water (TBW) as measured by deuterium dilution and another related to the age-specific hydration of FFM. A statistical model example is the use of bioimpedance analysis to quantify FFM and total body fat. In this case a method such as dual-energy x-ray absorptiometry (DXA) or deuterium dilution can be used as the reference for estimating total body fat and FFM in a well-defined subject group. Measured impedance values are then calibrated against the reference values to derive and cross-validate a body fat and/or FFM prediction equation. Multiple measured body properties can be combined to create robust models and predictions. For example, in low-resource settings, multiple anthropometric dimensions captured by optical scanning might be used along with measured impedance values to “predict” compartment such as FFM and skeletal muscle mass. These are important opportunities given the newer approaches applicable in low-resource settings described below.

In addition to standardized models, a need exists to use uniform measurement approaches in low-resource settings. Some standardized operating procedures developed in this context by the International Atomic Energy Agency have been previously reported (9).

Technologies with their Equivalences

Table 1 provides a decision tree for choosing an appropriate technology. The pragmatic needs of large epidemiological population studies differ from those of smaller scale clinical research. Hence, the initial steps are defined by the research question and available resources; is the required information for whole populations average effects or for descriptors of individuals, and is the experimental design cross-sectional or longitudinal? Often, whole body two-compartment information (e.g., FFM and FM) is sufficient to answer the research question. Alternatively, it might be important to define where in the body a loss in adipose tissue is occurring. The next decision is thus whether the primary outcome is whole body or regional body composition data. Because measuring body composition in a young child presents particular technological, practical, logistical and ethical challenges, we also divide technologies into those suitable for use below and beyond 5 years of age.

Table 1
Decision tree for choosing a technology for a low-resource setting

Several techniques exist to measure non-bone soft tissue compartments, but if bone mineral content or density is the primary research interest (e.g., development of osteopenia in the elderly), then only DXA is practical. If soft tissue information is required, then other technologies become available in addition to DXA. Only technologies suitable in low-resource settings are included in Table 1; we exclude technologies such as magnetic resonance imaging (MRI) or computed tomography (CT) that are only available in specialist centres.

Technologies suitable for low-resource settings

Table 1 lists the most suitable body composition methods by technological category. See Mattson and Thomas for a detailed methodological review (10), but some general points should be noted. BIA encompasses all variants of the technology: hand to hand, foot to foot, wrist to ankle and both lead and stand-on type devices; all measure tissue electrical impedance and transform these data into a prediction of body composition, typically TBW or FFM. Each have advantages and disadvantages and readers are referred to the specialist literature for further information (11). Irrespective of device type, available data (12) show that BIA provides no advantage for predicting body composition in the very young, and because of the likelihood of movement artefacts it is not recommended for use in this age group. Body density includes both underwater weighing and air displacement plethysmography (ADP) techniques although the former is rarely used today. ADP is theoretically usable in all age groups, but while special small-chamber systems have been validated for children less than 8 kg and 6 months of age, currently available instruments do not readily accommodate children between 6 months and 4 years of age. Subcutaneous adipose measurements include both skinfold calliper and ultrasound techniques, which are equivalent approaches. New and inexpensive ultrasound devices (13) may eventually supplant calliper techniques, since they do not require the training necessary to attain good quality control with calliper techniques. Anthropometry has also been extended to techniques using 2-dimensional (14) and 3-dimensional photogrammetry methods (15), which can provide whole body and regional volume and mass measurements. Whole body 3D optical images can provide over 400 automated anthropometric measures. Total body potassium measurements (TBK) and dilution techniques (deuterium dilution for TBW and bromide dilution for extracellular water) are reference techniques (9, 16) against which indirect field methods are calibrated and validated. Recent advances in analytical methods for Br and deuterium (17) mean these may become usable in field studies.

Special considerations for measuring body composition using specific technologies

There are a number of practical and technical issues to consider when selecting an appropriate methodology for measuring body composition in a low-resource setting (Table 2). Factors to consider include: cost, compliance needed, safety, accuracy, training required, infrastructure and technical support needed as well as quality control.

Table 2
Special considerations for measuring body composition with specific technologies.

Cost will often be a primary consideration. Costs include the initial capital outlay to purchase equipment and costs associated with maintenance and servicing, costs of data acquisition such as consumables needed to perform testing and costs of data analysis. Procedures such as DXA and TBK are more expensive due to initial and ongoing costs needed to support the testing apparatus and high costs per test. Participants need to be tested in a clinical/laboratory-based setting by trained technicians. Measurements of body fat obtained with BIA or subcutaneous skinfold thickness, which have considerably lower costs associated with initial equipment purchases, may be more cost effective, especially for large studies or studies requiring extensive follow-up. DXA and TBK have the highest degree of measurement accuracy. Less cost-prohibitive methods with a high level of measurement accuracy are isotope dilution techniques. For example, deuterium dilution measured with portable FTIR systems (18) might be considered for large population-based studies, because after the initial purchase of the instrument (approximately US$40,000 per unit), the deuterium dosing and collection of samples and analysis can be conducted at extremely low cost. This has led to the current use of deuterium dilution in large population-based studies in developing countries.

A user of technology also needs to evaluate the level of cooperation needed from the test subject. DXA or ADP measurements are challenging for participants aged 6 months to 5 years, owing to the need for the test subject to be still during the procedure. In this age group, anthropometric measurements and isotope dilution methods are more practical. However, DXA, ADP and isotope dilution are all highly correlated to the 4-compartment criterion method in children ages 9 to 14 years, with coefficients of determination (R2) between 0.95 and 0.98 (19). Impedance methods also require compliance, with pre-testing instructions such as fasting, recent participation in physical activity and hydration status. The accuracy of impedance methods are particularly sensitive to these factors, with between-day test-retest precision ranging from 2 to 6% (20), so we advise that standard operating procedures and quality control measures be established when using this methodology. Another important consideration for younger participants is safety. For example, DXA emits a low level of ionizing radiation (~1 mR or ≤ 1 day of background radiation), and ethics boards may not permit repeated testing in young children within short periods of time.

Finally, the supporting infrastructure needed to conduct testing varies considerably. Methods suitable for field-testing or where access to power or a typical clinic/research laboratory is not available include basic anthropometry, height/weight indices and skinfold thickness, or impedance devices that use a battery source for power. Methods such as DXA, ADP, 2D/3D optical spectroscopy, TBK and isotope dilution techniques require a clinical/laboratory-based setting with controlled ambient conditions and access to power sources, and in the cases of DXA and TBK, specifically trained personnel or technical staff. Emerging technologies available through smartphones may also provide effective measurements of body composition in field settings, particularly in conjunction with other existing and validated methodologies.

Conclusion

A wide variety of body composition technologies exist for low-resource settings. The decision schema and special considerations presented provide the researcher with a novel tool needed to make an appropriate decision concerning which to utilize for specific needs.

Study Importance Questions

What is already known?

  • Total body composition is an important measure of nutritional status, including underweight syndromes and obesity.
  • Regional measures of body composition, such as truncal fat and skeletal muscle mass, are useful indicators of risk for metabolic conditions.
  • Criterion methods of body composition assessment exist but are typically costly and not accessible in low-resource settings.

What this body of work adds?

  • New and novel technologies are becoming available to measure body composition that may be suitable for use in low-resource settings.
  • A decision tree that provides a structure for choosing a body composition technology with respect to population, study design, body region of interest, age of participants and tissue type is presented.
  • Other considerations should also be taken into account, including cost, participant compliance, support infrastructure, measurement accuracy, maturity of quality assurance procedures and the need for training.

Acknowledgments

This Perspective is based on discussions during a Consultants’ Meeting at the International Atomic Energy Agency, Vienna, 2–4 June 2015. Support by grants from the National Institutes of Health was provided to LMR (U01DK094418, R01DK099175) and JAS (P30DK098722-01). SAN is supported by the UK MRC/DfID African Research Leader Scheme.

Footnotes

Disclosure: The authors declare no conflict of interest.

References

1. Godfrey KM, Lillycrop KA, Burdge GC, Gluckman PD, Hanson MA. Epigenetic mechanisms and the mismatch concept of the developmental origins of health and disease. Pediatric Research. 2007;61:5R–10R. [PubMed]
2. Gillespie S, Haddad L, Allen L, Babu S, Horton S. Attacking the double burden of malnutrition in Asia and the Pacific. 2001.
3. Popkin BM, Adair LS, Ng SW. Global nutrition transition and the pandemic of obesity in developing countries. Nutr Rev. 2012;70:3–21. [PMC free article] [PubMed]
4. Macaulay S, Dunger DB, Norris SA. Gestational diabetes mellitus in Africa: a systematic review. PLoS One. 2014;9:e97871. [PMC free article] [PubMed]
5. Kuzawa CW, Hallal PC, Adair L, Bhargava SK, Fall CH, Lee N, et al. Birth weight, postnatal weight gain, and adult body composition in five low and middle income countries. American Journal of Human Biology. 2012;24:5–13. [PMC free article] [PubMed]
6. Sachdev HS, Fall CH, Osmond C, Lakshmy R, Biswas SKD, Leary SD, et al. Anthropometric indicators of body composition in young adults: relation to size at birth and serial measurements of body mass index in childhood in the New Delhi birth cohort. The American journal of clinical nutrition. 2005;82:456–466. [PubMed]
7. Wang ZM, Pierson RN, Jr, Heymsfield SB. The five-level model: a new approach to organizing body-composition research. Am J Clin Nutr. 1992;56:19–28. [PubMed]
8. Wang ZM, Heshka S, Pierson RN, Jr, Heymsfield SB. Systematic organization of body-composition methodology: an overview with emphasis on component-based methods. Am J Clin Nutr. 1995;61:457–465. [PubMed]
9. IAEA Human Health Campus (Resources for Health Professionals): IAEA. 2015 [cited 2015 06-05-2015]. Available from: http://nucleus.iaea.org/HHW/Nutrition/index.html.
10. Mattsson S, Thomas BJ. Development of methods for body composition studies. Phys Med Biol. 2006;51:R203–228. [PubMed]
11. Ward LC. Segmental bioelectrical impedance analysis: an update. Current Opinion in Clinical Nutrition & Metabolic Care. 2012;15:424–429. [PubMed]
12. Lingwood BE, Storm van Leeuwen AM, Carberry AE, Fitzgerald EC, Callaway LK, Colditz PB, et al. Prediction of fat-free mass and percentage of body fat in neonates using bioelectrical impedance analysis and anthropometric measures: validation against the PEA POD. Br J Nutr. 2012;107:1545–1552. [PubMed]
13. Johnson KE, Miller B, Juvancic-Heltzel JA, Agnor SE, Kiger DL, Kappler RM, et al. Agreement between ultrasound and dual-energy X-ray absorptiometry in assessing percentage body fat in college-aged adults. Clin Physiol Funct Imaging. 2014;34:493–496. [PubMed]
14. Xie B, Avila JI, Ng BK, Fan B, Loo V, Gilsanz V, et al. Accurate body composition measures from whole-body silhouettes. Medical physics. 2015;42:4668–4677. [PubMed]
15. Lee JJ, Freeland-Graves JH, Pepper MR, Yu W, Xu B. Efficacy of thigh volume ratios assessed via stereovision body imaging as a predictor of visceral adipose tissue measured by magnetic resonance imaging. American Journal of Human Biology. 2015 [PMC free article] [PubMed]
16. Murphy AJ, Ellis KJ, Kurpad AV, Preston T, Slater C. Total body potassium revisited. Eur J Clin Nutr. 2014;68:153–154. [PubMed]
17. Kehayias CE, Kempf KG, Roto AV, Kehayias JJ. A Simple, non-Destructive Method for Bromide Analysis in Plasma for Field Body Composition Studies. FASEB Journal. 2013;27:616.616.
18. Rein A, Higgins F, Preston T. [06-05-2015];Application of the Agilent 4500 Series FTIR to the Stable Isotope Technique for Assessing Intake of Human Milk in Breastfed Infants. 2015 Available from: http://www.chem.agilent.com/Library/applications/5991-5140EN.pdf.
19. Fields DA, Goran MI. Body composition techniques and the four-compartment model in children. J Appl Physiol. 2000;89:613–620. [PubMed]
20. Andersen TB, Jødal L, Arveschoug A, Eskild-Jensen A, Frøkiær J, Brøchner-Mortensen J. Precision and within-and between-day variation of bioimpedance parameters in children aged 2–14 years. Clin Nutr. 2011;30:326–331. [PubMed]