Body Composition

Body composition assessment is the umbrella term for measuring what your weight is made of rather than simply how much of it there is. A typical report separates fat mass from lean or fat-free mass, adds an estimate of total body water, and often reports a segmental breakdown by limb and trunk plus a visceral-fat index. The methods sitting under that umbrella are genuinely different from one another. Recovery studios, gyms, and med-spas most often use bioelectrical impedance analysis (BIA) on a standing device such as an InBody or Evolt, which passes a small imperceptible current through the body and infers composition from how that current is impeded. Dedicated scan providers, sports-performance labs, and longevity clinics more often use dual-energy X-ray absorptiometry (DXA or DEXA), which uses two low-dose X-ray energies to separate fat, lean tissue, and bone and can also report regional and bone-density values. Air-displacement plethysmography (BodPod), 3D optical scanning, and skinfold calipers fill out the category. What the research supports well is that the underlying metrics matter and that a repeated measurement is informative. Body-fat percentage and fat distribution are consistently associated with long-term health outcomes in large observational cohorts, and in a 2025 NHANES-linked analysis of adults aged 20 to 49 the BIA-derived body-fat percentage predicted 15-year mortality while BMI in the same model did not reach significance. What the research does not support is treating a single reading as a precise fact about your body. A 2026 systematic review comparing BIA devices against a four-compartment reference model found modest average bias but limits of agreement typically spanning 15 to 20 percentage points of body fat, which the authors read as clinical non-equivalence at the individual level. DXA is substantially more repeatable, with same-day precision error usually under two percent for fat and lean mass, though its visceral-fat estimate is far noisier. The practical framing is that this is a measurement service, not a treatment. The scan itself changes nothing; its value comes entirely from the training, nutrition, or medical decisions made around it and from measuring again under matched conditions later. Same device, same time of day, similar hydration, similar time since eating and since exercising. Read it as a trend line, not a verdict, and take any finding that looks clinically concerning to a qualified clinician rather than to the scan operator.

How body composition works

The measurement principles differ by method and so does the quality of the evidence behind them. BIA passes a low-level alternating current through the body; lean tissue conducts well because it holds most of the body's water and electrolytes, while fat resists, and a device-specific regression equation converts the measured impedance into fat and lean estimates. DXA instead measures how two X-ray energies are attenuated by tissue and solves for fat, lean soft tissue, and bone mineral directly, which is why it is commonly used as a practical reference method rather than as an estimate. Evidence that the metric is worth tracking is strong and comes from large prospective cohorts. A 2025 analysis in Annals of Family Medicine followed 4,252 US adults aged 20 to 49 from NHANES 1999-2004 for roughly 15 years and found that a high BIA-measured body-fat percentage (27 percent or more in men, 44 percent or more in women) was associated with higher all-cause mortality (HR 1.78, 95% CI 1.28-2.47) and heart-disease mortality (HR 3.62, 95% CI 1.55-8.45), while a BMI of 25 or above did not reach statistical significance in the same adjusted model (HR 1.25, 95% CI 0.85-1.84). Earlier NHANES III linkage work reported a U-shaped relationship between body-fat percentage and mortality, so the association is not simply linear. Evidence for the accuracy of any one reading is weaker, and this is the honest limitation of the category. A 2026 systematic review in the Journal of Functional Morphology and Kinesiology pooled 12 studies comparing BIA against a four-compartment criterion model in healthy adults: mean bias for body-fat percentage ranged from -3.5 to +4.4 percentage points and mean bias for fat-free mass from -3.9 to +1.8 kg, but limits of agreement typically spanned 15 to 20 percentage points of body fat and often exceeded plus or minus 6 kg of fat-free mass. The authors concluded the devices are not clinically equivalent at the individual level despite low average bias. A 2025 systematic review of 28 studies in children and adolescents with obesity found BIA consistently underestimated fat mass and body-fat percentage against DXA. A separate 2025 review characterized BIA in the health field as a still scientifically fragile tool for individual body-composition assessment. Repeatability is where the field is most usable. A reliability study of weight-stable adults on a GE iDXA reported same-day coefficients of variation under two percent for the main compartments, rising to roughly 0.8 to 5.9 percent over three months as biological variation enters, with visceral adipose tissue a clear outlier at about 16 percent same-day and 42 percent over three months. Standard densitometry practice converts precision error into a least significant change (2.77 times the precision error at 95 percent confidence), which is the formal reason a small movement between two scans should not be read as a real change. Taken together, the evidence supports using body composition as a repeated trend on one device under standardized conditions, and does not support reading a single absolute number as accurate to the decimal place.