DEXA vs InBody: Which Body Composition Scan Is More Accurate?

DEXA and InBody (bioelectrical impedance) both estimate body fat and lean mass, but they measure different things and react differently to water, food and exercise. Here is how they compare on accuracy, repeatability, bone density and cost, and when each one is good enough.

Published

October 5, 2026

Read time

11 min read

Medically reviewed by

Rhonda Collins, FNP-C

Share

Copy link

DEXA is the more accurate and more detailed of the two, and InBody is the cheaper, faster way to track a trend. A DEXA scan uses two low-dose X-ray beams to separate fat, lean tissue and bone, region by region. An InBody scan sends a small electrical current through your body and estimates fat and lean mass from equations. Both are repeatable when you measure under the same conditions. But for an individual, InBody numbers can differ from DEXA and from the research reference method by several percentage points of body fat, and InBody cannot measure bone density.

If you want a precise baseline, regional and visceral fat, or a reliable read on whether weight loss is costing you muscle, DEXA is the better tool. If you want frequent, low-cost check-ins on a trend, InBody can be good enough, as long as you compare it only with itself.

DEXA vs InBody at a glance

What DEXA and InBody actually measure

DEXA measures how tissue absorbs X-rays. The scanner passes two X-ray beams at different energies over your body. Bone, fat and lean soft tissue absorb those energies differently, so DEXA can split each region of the body into bone mineral, fat and lean tissue (Shepherd et al., 2017). Lean tissue here means everything that is not fat or bone, including muscle, organs and the water they hold.

InBody measures electrical resistance, then estimates. A bioelectrical impedance analyzer sends a small, painless current through your body and measures how much the body resists it. Water in lean tissue conducts the current well and fat does not. The device does not see fat or muscle. It calculates fat-free mass from the measured impedance using prediction equations, and a European Society for Clinical Nutrition and Metabolism (ESPEN) review notes that those equations work best in people without significant fluid or electrolyte abnormalities, using equations suited to the person’s population and age (Kyle et al., 2004). InBody analyzers are a brand of multi-frequency, segmental bioimpedance devices, and home body fat scales use the same basic principle.

That difference explains most of what follows. DEXA reads tissue directly, and InBody infers it from body water, so anything that shifts your water shifts the InBody estimate.

How accurate is InBody compared with DEXA?

For a group, InBody and DEXA track each other closely. For an individual, they can disagree by several percentage points of body fat, in a consistent direction. Several independent validation studies show the pattern, each with its own limits:

• InBody reads body fat lower than DEXA. In 1,000 healthy, physically active adults tested on an InBody 770 and DEXA in real-world conditions, with no control over food, hydration, exercise or time of day, InBody body fat percentage was on average 4.2 points lower than DEXA in men and 2.8 points lower in women. It read fat-free mass about 3.4 kg higher in men and 2.0 kg higher in women. The authors called it a moderately accurate method in uncontrolled conditions. The participants were young, fit adults, so the numbers may differ in other groups.
• The bias held across InBody models. In 67 healthy men and women measured on an InBody 230, 720 and 770 against DEXA, all three underestimated body fat and overestimated fat-free mass compared with DEXA. Individual error was small, and the authors concluded InBody can stand in when DEXA is not available, provided you account for the bias. The research-grade 720 and 770 added little over the portable 230 in that study.
• The two are not interchangeable for one person. In 44 healthy women aged 40 to 65, bioimpedance and DEXA showed poor individual-level agreement, with wide limits of agreement. The authors cautioned against applying another device’s norms or cut-points to a bioimpedance result.

Is DEXA the gold standard for body composition?

Not quite. DEXA is the most widely used accurate clinical method, but the research reference for body composition is the four-compartment model, which combines separate measurements of body water, body volume and bone mineral. A review in Obesity found that most studies comparing DEXA with the four-compartment model observed significant differences between them. The four-compartment model takes several instruments and is used in research labs, not offered as a routine scan.

Against that same reference, InBody also has error. In 82 normal-weight young adults, an InBody 770 had a total error of 4.2 percentage points for body fat compared with the four-compartment model, 2.5 points in men and 4.7 points in women, and it tended to overestimate body fat. Note that this direction is the opposite of the comparisons with DEXA above, which is one more reason not to treat any single device as truth.

How repeatable are DEXA and InBody?

Both are highly repeatable when you measure twice under the same conditions, which is why each works for tracking your own trend:

• In the study of 44 women, duplicate measurements had a precision error of 1.17% for body fat percentage on DEXA and 1.25% on bioimpedance, both within the study’s preset acceptability limits.
• In the study of 67 adults, InBody tests repeated 24 to 72 hours apart needed a change of about 2.1 to 2.7 percentage points of body fat to count as a real difference, depending on the model.
• In the study of 1,000 adults, repeat DEXA and InBody measurements within a week were both highly reliable.

Least significant change. A change smaller than the measurement error may be noise. The least significant change is the smallest difference between two tests that you can trust as real, usually at 95% confidence. The International Society for Clinical Densitometry (ISCD) says each DEXA facility should measure its own precision error and calculate its least significant change, rather than rely on the manufacturer’s figure. It sets the minimum acceptable precision for a DEXA technologist at 3% for total fat mass, 2% for total lean mass and 2% for percent fat (ISCD official positions). For InBody, the 2.1 to 2.7 point figure above plays the same role: a smaller change between two tests may be noise.

Repeatability holds only on the same device. RadiologyInfo advises follow-up DEXA scans at the same institution and ideally on the same machine, because results from different equipment cannot be directly compared (RadiologyInfo). The same logic applies to InBody models.

What changes an InBody reading

Because InBody estimates body composition from body water, its reading moves with anything that moves your water. DEXA is affected too, but less.

• Food, fluid and time of day. In 32 athletic men, testing in the afternoon after normal eating and drinking, rather than fasted in the morning, produced large changes in fat mass and fat-free mass estimates from bioimpedance spectroscopy, a related impedance method. The same conditions produced only small changes in DEXA fat-free mass. The authors found DEXA acceptable if recent food and fluid stayed below about 500 g.
• A normal breakfast may matter less than expected. In 39 healthy adults measured with a single-frequency device, a standardized 400-calorie breakfast had no clinically relevant effect on estimated fat-free mass for 90% of participants. Large meals, big drinks, dehydration and hard exercise are the bigger risks to a consistent reading.
• Real-world testing widens the gap. Researchers in the 1,000-adult study note that impedance devices are most reliable under tightly controlled conditions of food, hydration, recent exercise and time of day.

Does InBody measure bone density?

No. InBody does not image bone, so it cannot measure bone mineral density or diagnose osteoporosis. DEXA of the hip and spine is the established standard for measuring bone mineral density (RadiologyInfo).

A whole-body DEXA scan for body composition also gives a bone reading, but guidelines use a hip and spine scan, such as the DEXA Bone Density Scan, to diagnose osteoporosis. Mito’s guide on how to read DEXA scan results explains T-scores and Z-scores.

Is DEXA safe, and does InBody use radiation?

DEXA uses a very small X-ray dose. RadiologyInfo describes the dose of a DEXA bone density exam as less than one-tenth of a standard chest X-ray and less than a day’s exposure to natural background radiation (RadiologyInfo). DEXA is not done during pregnancy. InBody uses no radiation.

How much DEXA and InBody cost

InBody is usually the cheaper option per test, and some gyms include it with a membership. These are published cash prices from a few providers, checked on October 5, 2026. Prices vary by city and change, so confirm with the provider before you book.

• InBody: $15 a scan at the YMCA of Greater Kalamazoo, which also includes scans in its personal training packages; $30 for members and $40 for nonmembers at LIFT Wellness Center; $45 for a single scan at Creekside Physical Therapy; and $49 at Fitness Formula Clubs.
• DEXA body composition: $75 at Southern Arizona Radiology Associates in Tucson; $99 at DexaFit Denver; and $135 at DexaFit Sacramento.

Insurance rarely covers a body composition scan, because it is elective. A bone density scan for osteoporosis screening is different and is often covered when you meet screening criteria.

When is InBody good enough, and when is DEXA worth it?

InBody is often good enough when:

• you want frequent, low-cost check-ins, such as monthly at your gym;
• you care about the direction of change more than the exact number;
• you can measure on the same device, at the same time of day, in the same fed and hydrated state.

DEXA is worth it when:

• you want a precise baseline to compare against later;
• you want regional fat, left and right balance, or a visceral fat estimate. In the 1,000-adult study, InBody’s visceral fat estimate agreed poorly with DEXA’s, especially in women;
• you are losing weight, through diet, training or a weight-loss medication, and want to know how much of the loss is lean mass. DEXA’s lean mass reading still includes water, so keep scan conditions consistent;
• your hydration is likely to be unusual, for example during a hard training block, which makes impedance estimates less reliable;
• you also want a bone reading, or need a hip and spine bone density test.

Some people use both: a DEXA baseline and follow-up every few months, with InBody check-ins in between to watch the direction of change. If you do, never mix the two numbers in one trend line.

How to get comparable repeat readings from either method

The method matters less than measuring the same way each time:

• Use the same machine. Compare DEXA with DEXA and InBody with InBody, ideally on the same device and at the same location.
• Keep the time of day the same, and arrive in the same state of food and fluid.
• Skip hard exercise, saunas and steam rooms beforehand. Mito’s DEXA scan instructions ask you to skip food for 2 to 4 hours, drink water normally without overdoing it right before, and avoid hard exercise, saunas and steam rooms for 12 hours.
• Keep supplements and routines steady between tests. Anything that changes body water can move lean mass numbers, and starting creatine has been shown to raise DEXA lean mass. Mito’s DEXA body composition guide covers DEXA preparation and repeat scans in more detail.
• Wait for a real change. Rescan when the change you expect is larger than the measurement error, not on a fixed weekly schedule.

Get a DEXA body composition scan through Mito

Mito offers a DEXA body composition scan with no referral needed. Depending on your state, it is booked through Fitnescity Health at a partner facility, such as an imaging center, hospital, or longevity or performance clinic, or at a BodySpec studio in California, Colorado, Illinois, Massachusetts, Texas or Washington. Different machines can give slightly different numbers, so rescan at the same location to track change. Your report lands in your Mito dashboard, next to your lab results, and the product page shows your provider, price and how to book.

Photo by Ginny Rose Stewart on Unsplash.

Frequently asked questions

Are InBody scans actually accurate?

InBody scans are repeatable, but their absolute numbers are estimates that can be off by several percentage points of body fat for an individual. Studies comparing InBody analyzers with DEXA and with the four-compartment reference model found consistent biases, so InBody is most useful for tracking your own trend on the same device under the same conditions.

Which is more accurate, a body fat scale or a DEXA scan?

DEXA is the more accurate and more detailed of the two. A home body fat scale estimates body fat from a small electrical current, the same basic idea as InBody, so its reading moves with hydration. It cannot measure regional fat, visceral fat or bone the way DEXA does.

Does an InBody scan tell you bone density?

No. InBody does not image bone, so it cannot measure bone mineral density or diagnose osteoporosis. Bone density is measured with a DEXA scan of the hip and spine.

Is there anything more accurate than a DEXA scan?

Yes. The four-compartment model, which combines body water, body volume and bone mineral measurements, is the research reference for body composition, and DEXA results often differ from it. It is used in research labs rather than offered as a routine scan.

Can I compare my InBody result with my DEXA result?

Not directly. Studies found the two methods disagree for individuals, and InBody tends to read body fat lower than DEXA. Compare InBody with InBody and DEXA with DEXA, ideally on the same machine each time.

Why does my InBody body fat change from day to day?

InBody estimates body fat from how a small current travels through body water, so changes in hydration, recent meals, exercise and time of day can shift the reading. Measuring at the same time, in the same state, each time reduces that noise.