DEXA vs InBody vs Scale: What GLP-1 Users Actually Need to Track

You’ve been on semaglutide for three months. The scale says you’re down 12kg. Your clothes fit differently. Your face looks thinner. Everything points in the right direction.

But down 12kg of what?

This is the question that eventually sends every serious GLP-1 user searching for “DEXA scan near me” or “InBody vs DEXA accuracy” at midnight. Because the bathroom scale can tell you one thing and one thing only: your relationship with gravity changed. It cannot tell you whether you lost fat, muscle, water, or some combination of all three. And on GLP-1 medications, where lean mass loss can account for 15 to 39 percent of total weight lost, that distinction isn’t academic. It’s the whole game.

So let’s break down what each method actually measures, where it’s accurate, where it lies, and why the answer to “which one should I get?” is less important than how you use whichever one you pick.

The bathroom scale (and smart scales)

Let’s start here because this is what most people are using. A standard bathroom scale measures total body weight. That’s it. A “smart” scale with bioelectrical impedance analysis (BIA) sends a small electrical current through your body and estimates body fat percentage based on how the current travels through different tissues.

Smart scales are cheap (usually $30 to $80), convenient, and available every morning. The problem is accuracy. The body fat estimates from consumer BIA scales can be off by 5 to 8 percentage points compared to reference methods. Hydration status swings the number dramatically. Drink two glasses of water before stepping on and your body fat percentage drops. Not because you lost fat. Because water conducts electricity well and the algorithm interprets well-hydrated tissue as lean mass.

For GLP-1 users, this matters. Semaglutide commonly causes reduced fluid intake (less thirst signalling) and GI side effects that can dehydrate you. Both of these will make a BIA scale overestimate your body fat percentage, which means it might tell you you’re losing less fat than you actually are. Or it might swing the other direction after a high-sodium meal when you’re retaining water.

The day-to-day readings from a smart scale are essentially noise. The two-week trend, if you measure at the same time under the same conditions, has some directional value. But if you’re relying on a smart scale to tell you whether you’re losing muscle on semaglutide, you’re flying blind.

InBody

InBody scanners use a more sophisticated version of the same bioelectrical impedance technology. Instead of sending one current through your feet (like a bathroom scale), they use multiple frequencies across multiple pathways: both hands and both feet. This segmental, multi-frequency approach is significantly more accurate than a consumer scale.

The InBody 570 and 770 models, which you’ll find at many gyms, physiotherapy clinics, and body composition studios, have been validated against DEXA with reasonably good agreement. Studies show InBody measurements of lean mass and fat mass typically fall within 2 to 4 percent of DEXA values for most populations.

An InBody scan takes about 60 seconds. You stand on the platform, grip the handles, and wait. The output gives you total weight, skeletal muscle mass, body fat mass, body fat percentage, visceral fat level, and segmental lean mass breakdown (each arm, each leg, trunk). Costs run between $30 and $60 per scan at most facilities.

The catch? InBody still relies on impedance, which means hydration status still affects results. If you scan after a heavy training session when you’re dehydrated, your lean mass reading will be lower than reality. If you scan after a sodium-heavy meal when you’re holding water, lean mass will read higher. Consistency of conditions matters. Same time of day, similar hydration, similar recent food intake. Control what you can control and the scan-to-scan trend becomes reliable.

For GLP-1 users, InBody hits a good balance between cost, accessibility, and accuracy. Getting a scan every 8 to 12 weeks gives you a meaningful lean mass trend without spending hundreds on DEXA each time.

DEXA (Dual-Energy X-ray Absorptiometry)

DEXA is the clinical gold standard for body composition measurement. It uses two low-dose X-ray beams at different energy levels to differentiate between bone mineral content, lean soft tissue, and fat tissue. You lie on a table for about 10 minutes while the scanner passes over you.

The precision of DEXA for body fat percentage is typically within 1 to 2 percent of true values. For lean mass, it’s the most accurate commercially available method outside of MRI or cadaver analysis (which, for obvious reasons, is not a practical option for most people).

DEXA also gives you regional data: fat distribution across trunk, arms, legs. This matters for GLP-1 users because semaglutide can affect visceral fat (the dangerous kind around your organs) and subcutaneous fat differently. Knowing where you’re losing from tells you something the total number doesn’t.

The downsides are cost and access. A DEXA scan typically runs $80 to $150 in Australia, $100 to $200 in the US. It involves a small radiation dose (roughly equivalent to a day of natural background radiation, so minimal). And depending on where you live, you might need to book weeks in advance at a specialised clinic.

DEXA also has its own quirks. Hydration affects DEXA readings too, though less than BIA. The machine’s calibration and the technician’s positioning of your body matter. Different DEXA machines from different manufacturers can give slightly different results, so ideally you use the same machine and the same clinic each time.

The comparison nobody talks about honestly

Here’s a table that most comparison articles won’t give you this directly:

Method	Body fat accuracy	Lean mass accuracy	Cost	Frequency	Hydration sensitivity
Bathroom scale/BIA	±5-8%	Poor	$30-80 (one-time)	Daily	Very high
InBody (570/770)	±2-4%	Good	$30-60/scan	Every 8-12 weeks	Moderate
DEXA	±1-2%	Excellent	$80-150/scan	Every 12-16 weeks	Low

Every method has error margins. Every method is affected by conditions. The question isn’t “which one gives me the perfect number?” None of them do. The question is “which one gives me a consistent enough signal that I can track the trend over time?”

Why the trend beats the number

Here’s the thing that matters more than which scan you choose: a single body composition measurement is a snapshot. It tells you where you are right now, plus or minus the error margin of the method. It’s useful but limited.

The real value comes from repeated measurements over time using the same method, same conditions, same time relative to your cycle and hydration and training. Because when you do that, the error margin stays relatively constant. The bias might be 3% off from true, but it’s consistently 3% off. Which means the change between scans is accurate even if the absolute number isn’t perfect.

This is why an InBody scan every 8 weeks is more useful than a single DEXA scan in isolation. Three InBody readings over six months tells you the trajectory of your lean mass on semaglutide. One DEXA reading tells you where you were on that particular Tuesday.

For GLP-1 users, I’d suggest this approach: get a baseline scan before or shortly after starting medication (InBody or DEXA, whichever is more accessible and affordable for repeat visits). Then repeat every 8 to 12 weeks, controlling conditions as much as possible. Track the lean mass number specifically, not just body fat percentage. Because body fat percentage can improve even while you’re losing muscle if you’re losing fat faster. The absolute lean mass trend is what tells you whether your nutrition and training strategy is preserving what matters.

The piece that’s still missing

Knowing your lean mass trend is necessary. It’s not sufficient. A DEXA scan every 12 weeks tells you what happened. It doesn’t tell you why. And it definitely doesn’t tell you what to change.

Did your lean mass drop because your protein intake was too low? Because your training stimulus wasn’t adequate? Because your sleep quality tanked during a stressful month at work? Because your GLP-1 dose increased and your appetite crashed, taking your caloric intake below the threshold where muscle preservation is possible?

The scan can’t answer those questions. To answer them, you need to correlate the body comp trend with your nutrition data, your training log, and your recovery metrics from your wearable. You need to look at the eight weeks between scans and understand what happened across all of those dimensions, not just what the final number says.

This is where most GLP-1 users hit a wall. They get the scan. They see the lean mass number. They either feel good or feel worried. But they don’t have a clear path from “I lost 1.2kg of lean mass” to “here’s specifically what to change for the next eight weeks.”

The scan data, the nutrition log, the wearable recovery trends, and the training volume all exist. They’re just in different apps with no way to see the connections between them. Bridging that gap between isolated snapshots and connected, actionable insight is exactly where P247 is focused. Not replacing any of these tools, but making the data they generate actually talk to each other.

---

Suggested X Thread

Tweet 1: DEXA vs InBody vs smart scale: which one should GLP-1 users actually get?

Short answer: whichever one you’ll repeat consistently every 8-12 weeks. The trend beats the number. 🧵

Tweet 2: Smart scales can be off by 5-8% on body fat. InBody is within 2-4% of DEXA. DEXA is ±1-2%. But here’s the thing: even a biased measurement is useful if the bias is consistent. Track the delta, not the absolute.

Tweet 3: For GLP-1 users specifically: hydration shifts from appetite suppression and GI side effects will mess with any impedance-based reading. Control your conditions. Same time, same hydration, same day relative to your last meal.

Tweet 4: The real problem isn’t which scan to get. It’s what to do with the result. A DEXA says you lost 1.2kg of lean mass. But why? Protein too low? Training inadequate? Sleep disrupted? The scan doesn’t connect those dots.

Tweet 5: Body comp scans tell you what happened. Your nutrition log, training data, and wearable metrics explain why. Connecting them is the hard part. New post breaking down what each method actually measures and what’s still missing: [link]