Introduction
Body composition testing is widely used in research, sport and clinical practice. But what are these methods actually measuring, and how much confidence should we place in the numbers they produce?
In this episode, we examine the assumptions behind common approaches including DXA, bioelectrical impedance, and skinfold measurements. We discuss the distinction between accuracy and reliability, the value and limitations of repeated measurements, and how coaches, clinicians, researchers and individuals can select an appropriate method and interpret the data without overstating its precision.
Joining Danny to discuss these issues is Professor Grant Tinsley, Professor of Human Performance at Texas Tech University and Director of its Energy Balance and Body Composition Laboratory.
[03:04] Grant Tinsley’s research
[05:25] What body composition means
[08:50] Skeletal muscle mass vs. fat-free mass vs. lean soft tissue
[11:51] How different methods estimate body composition
[23:38] Group vs individual accuracy
[31:09] Standardizing testing conditions and assessments
[36:36] Reliability testing thresholds
[40:08] Understanding phase angle in bioelectrical impedence results
[45:02] Consumer devices
[50:55] Not all benefit from monitoring
[54:14] Using data wisely
[01:02:55] Key Ideas (premium-only)
Episode Resources
- Join the Sigma newsletter for free
- Subscribe to Sigma Nutrition Premium
- Enroll in the next cohort of our Applied Nutrition Literacy course
- Prof. Tinsley’s website
- LinkedIn – Prof. Tinsley
- Instagram: @grant_tinsley_phd
Research Relevant to this Episode:
- Wang et al. (1992): The five-level model: a new approach to organizing body-composition research.
- Methodological standards for body composition: levels, models and terminology.
- Tinsley et al.: Tracking changes in body composition and the influence of pre-assessment standardization.
- Siedler et al.: Reliability and validity of 14 consumer-grade BIA devices.
- Rodriguez et al.: Influence of acute localized resistance exercise on body-composition estimates.
- Tinsley et al.: Agreement of raw bioelectrical variables from supine and standing analyzers.
- Mattiello et al.: Phase-angle reference values in more than 250,000 participants.
- Tinsley (2024): Fundamental body-composition principles in the interpretation of weight-loss changes.
Guest Information
Professor Grant Tinsley is a Professor of Human Performance at Texas Tech University and Director of its Energy Balance and Body Composition Laboratory. His research focuses on assessing and modifying human body composition through nutrition, exercise, energy-balance interventions and obesity medications.
He has published more than 180 peer-reviewed papers and is a Fellow of both the American College of Sports Medicine and the International Society of Sports Nutrition. He also serves as Chief Science Officer at Vineyard and advises companies across the health, nutrition and fitness sectors.
Study Notes
Overview
Body composition describes the physical makeup of the body, but a reported value such as body-fat percentage is not a direct reading of that makeup. In a living person, most techniques measure an accessible property, such as electrical impedance, X-ray attenuation, body density, circumferences or skinfold thickness, and then use models and assumptions to estimate the component of interest.
- Different terms refer to different biological levels. Fat-free mass, lean soft tissue and skeletal muscle mass overlap, but they are not interchangeable.
- There is no universally best method. The appropriate choice depends on the question, the component of interest, access, cost, testing frequency, operator expertise and the magnitude of change that needs to be detected.
- Group-level validity does not ensure individual accuracy. Overestimation in some people and underestimation in others can cancel out, producing an excellent group mean while individual errors remain large.
- Longitudinal interpretation requires standardization. Food, fluid, glycogen, recent exercise, blood flow and other biological factors can create apparent changes or conceal real ones.
- Displayed precision is not measurement certainty. A result reported to two decimal places may still carry an uncertainty of several percentage points.
- Testing is optional for many people. It is valuable only when it supplies information that changes interpretation or management without creating disproportionate anxiety or fixation.
Useful Terminology
- Body Composition: The makeup of the body expressed as defined components of body mass.
- Body-Composition Model: A conceptual framework that partitions body mass into components. A two-component molecular model divides mass into fat mass and fat-free mass; more complex models separate additional components.
- Fat Mass: At the molecular level, the mass of non-polar lipids extracted from the body. It is not identical to adipose tissue.
- Fat-Free Mass (FFM): Everything in body mass other than molecular fat. It includes water, protein, mineral, glycogen and other constituents across all tissues.
- Lean Soft Tissue (LST): Fat-free mass excluding bone mineral content. This is a common DXA output.
- Skeletal Muscle Mass: The mass of anatomically defined skeletal muscles. It belongs to the organ-tissue level, rather than the molecular level.
- Adipose Tissue: An anatomical tissue consisting mainly of lipid but also containing water, protein, cells and supporting structures. Consequently, adipose tissue contributes to both molecular fat mass and fat-free mass.
- Criterion Method: A method used as the best available comparator for a particular purpose. The term does not imply perfect truth.
- Technical Error: Variation attributable to the instrument, software, positioning, operator, calibration or analysis procedure.
- Biological Variability: Real short-term variation in body water, glycogen, gastrointestinal contents, blood flow and other physiological quantities that can affect an estimate without representing the tissue change of interest.
- Reliability or Precision: The consistency of repeated measurements under specified conditions.
- Validity or Agreement: How closely a method agrees with an appropriate reference method for the intended quantity and population.
- Proportional Bias: Error that changes according to the level of the measured quantity, such as systematically poorer agreement at very low or very high body-fat percentages.
- Phase Angle: A raw bioimpedance-derived quantity calculated from reactance and resistance. It is expressed in degrees and is commonly interpreted as reflecting aspects of cell-membrane function, tissue composition and fluid distribution.
Body Composition Can Be Described at Different Levels
The five-level model proposed by Wang and colleagues organizes body composition at the atomic, molecular, cellular, tissue-system and whole-body levels. Internal consistency matters: components from different levels cannot simply be added, subtracted or treated as synonyms.
The molecular level
- Most familiar field and laboratory outputs, including body-fat percentage, fat mass and fat-free