InBody devices are reasonably accurate for tracking body fat trends over time, but they consistently underestimate body fat percentage when compared to DEXA scans, which are considered the clinical gold standard. Studies comparing the InBody 770 to DEXA show strong correlations (R² values of 0.62 to 0.85) but also significant and systematic underestimation of actual body fat. In practical terms, your InBody reading will likely show you as leaner than you actually are, and the gap can be meaningful.
How InBody Measures Body Fat
InBody machines use a method called bioelectrical impedance analysis, or BIA. You stand barefoot on metal plates and grip handles while the device sends tiny electrical currents through your body at multiple frequencies. The InBody 770, one of the most widely used models, runs six different frequencies (1, 5, 50, 250, 500, and 1,000 kHz) and takes 30 separate impedance measurements across five body segments: right arm, left arm, trunk, right leg, and left leg.
The core principle is simple: lean tissue (muscle, organs, bone) contains a lot of water and conducts electricity well, while fat tissue does not. By measuring how easily current passes through each segment of your body, the device estimates how much of you is fat versus lean mass. Proprietary algorithms then combine these segment-level readings into a full body composition report.
This segmental approach is a meaningful upgrade over cheaper single-frequency devices like basic bathroom scales, which send current through only the lower body and use generic prediction equations. By measuring each limb and the trunk separately, InBody captures a more complete picture of where your mass is distributed.
How It Compares to DEXA Scans
DEXA (dual-energy X-ray absorptiometry) is the reference standard most researchers use to evaluate body composition tools. When the InBody 770 was tested against DEXA in a study of 44 participants, the two devices showed high statistical correlation, meaning they generally moved in the same direction. If DEXA said someone had more body fat, InBody usually agreed. But the absolute numbers told a different story.
The InBody 770 significantly underestimated body fat percentage and total fat mass while overestimating fat-free mass (muscle, bone, water). The researchers described the gap between the two devices as a “large bias,” meaning InBody wasn’t just off by a small rounding error. If DEXA measured your body fat at 25%, the InBody reading could plausibly come back several percentage points lower. This pattern of underestimating fat is consistent across multiple studies and is not unique to one lab or one population.
For someone trying to get a single, precise snapshot of their body fat, this matters. An InBody reading of 18% might correspond to a DEXA reading of 22% or higher. The exact discrepancy varies by individual, but the direction of the error is predictable: InBody leans optimistic.
Where InBody Excels: Tracking Changes
Despite the accuracy gap, InBody devices are remarkably consistent from one test to the next. A study of the InBody 270 found near-perfect test-retest reliability, with intraclass correlation coefficients of 0.98 to 1.00 and technical error of measurement as low as 0.04%. In plain terms, if you step on the same InBody machine twice under the same conditions, you’ll get almost identical results.
This is the real strength of the technology. While the absolute number may be off, the device is reliable enough to detect genuine changes in your body composition over weeks or months. If you start a training program and your InBody body fat reading drops from 24% to 21% over three months, that change is likely real, even if your true starting point was closer to 28%. Think of it less as a precision scale and more as a highly consistent ruler: it may not tell you exactly where the line is, but it can reliably tell you which direction it moved.
Factors That Throw Off Your Reading
Because BIA works by measuring how well your body conducts electricity, anything that changes your hydration level directly affects the result. Dehydration causes the device to overestimate body fat (less water means more resistance, which the algorithm interprets as more fat). Overhydration does the opposite, making you appear leaner than you are. Drinking roughly 800 mL of fluid, which represents about a 2% increase in total body water, was enough to produce statistically significant shifts in readings for intracellular water, extracellular water, skeletal muscle mass, and fat mass in a controlled crossover trial.
To minimize these effects, testing facilities recommend the following preparation:
- Food and exercise: Avoid both for at least 3 hours before testing.
- Alcohol: None for 24 hours prior.
- Caffeine: None on the day of the test (caffeine is a mild diuretic that shifts fluid balance).
- Heavy training: Skip intense workouts before your appointment, as exercise redistributes blood and fluid throughout the body.
Time of day also matters. Most people are mildly dehydrated in the morning and progressively more hydrated throughout the day as they eat and drink. Testing at the same time of day, under the same conditions, is more important than any single reading’s accuracy. If you tested fasted at 7 a.m. last month and then test after lunch this month, the comparison is unreliable regardless of the device’s precision.
Differences Between InBody Models
Not all InBody machines are identical. The InBody 770, which is common in gyms and clinics, uses frequencies up to 1,000 kHz. Newer models like the InBody 970S and BWA 2.0S extend up to 3 MHz, which allows them to differentiate more precisely between water inside cells and water outside cells. All three models take 30 impedance measurements across the same five body segments, so the basic architecture is similar, but the higher-frequency models capture more granular data about fluid distribution.
In practice, the model you encounter at a gym is most likely the InBody 270, 370, or 770. The differences in body fat accuracy between these consumer-facing models are less dramatic than the differences caused by hydration status, meal timing, or recent exercise. Controlling your pre-test conditions will do more for the quality of your results than seeking out a specific model.
How to Get the Most Useful Results
If you’re using InBody to guide a fitness or weight loss plan, treat it as a trend-tracking tool rather than an absolute measurement. Test under identical conditions each time: same time of day, same fasting window, same distance from your last workout. Monthly testing is a reasonable frequency for most people, since body composition changes slowly and testing too often captures noise rather than signal.
Keep in mind that the device will likely report your body fat as lower than it truly is. If you need a precise, one-time measurement for medical or athletic purposes, a DEXA scan is a better choice. But for monitoring your progress over time at a fraction of the cost and inconvenience, InBody’s consistency makes it a practical option, as long as you understand what it can and cannot tell you.