Why We Use DEXA Scans, and You Should Too
Losing Weight Is Easy. That's the Problem.
Want to lose weight fast? Stop eating and start running.
Seriously. It works. Cut your food down to nothing, add an hour of cardio a day, and you will drop weight faster than any program on the market. The math is undeniable. The scale will reward you inside of a week.
Nobody recommends it. Ever wonder why?
Because most of what you lose won't be fat. It'll be muscle. You'll end up lighter, weaker, softer, and more metabolically broken than when you started, and the scale will congratulate you the entire way down.
If that mechanism sounds familiar, it should. Suppress appetite, drop calories, watch the number fall. That's the same lever a GLP-1 pulls, just with far better compliance. Bolt resistance training, high protein, and real sleep on top and it's a legitimate tool. Skip those and the tissue you lose starts to rhyme with the guy who stopped eating and started running. We wrote about that here.
That's the problem with weight. It's a number that can't tell you what you lost. And if you can't tell what you lost, you have no idea whether what you're doing is working.
This is why we put a DEXA scan in front of every client.
Muscle is the organ of independence
More muscle and less fat, within reason, is one of the most reliable predictors we have of both how long you live and how well you live while you're doing it.
That shouldn't be surprising. Muscle is the organ of independence.
Here's how it actually plays out. Strength is your ability to produce force. Every activity in your life demands a certain amount of force to complete. When your ability to produce force drops below the threshold that activity requires, you stop doing that activity. Not because you decided to, but because you can't.
Sometimes that's recreational. The pack you need to carry to hunt in the back country. The bike you can no longer lift onto the rack.
Sometimes it isn't recreational at all. Picking up your grandkid. Getting off the toilet without help.
And you don't get a warning shot, which sucks. Without intervention, adults lose somewhere in the range of 3 to 8% of their muscle mass per decade after age 30, and the rate accelerates after 60. Strength goes faster still. The research puts strength loss at two to five times the rate of mass loss. You don't notice the slide down, you just notice hitting the bottom.
Bone comes along for the ride. As muscle mass goes, so goes bone mineral density. Less density means a higher chance that a fall, a stumble, or a missed step off a curb turns into a fracture. And in an older adult, a fracture is rarely just a fracture. It's the first domino: hospitalization, bed rest, accelerated muscle loss, and a permanent step down in independence.
The ordinary mistakes of daily life stop being mistakes and start being catastrophes.
Fat isn't an aesthetic problem
More fat mass drives worse metabolic health. Worse metabolic health drives inflammation. And inflammation plays a causal role in most of the chronic disease that makes Americans miserable before it kills them. Type II diabetes. Heart disease. Cognitive decline. The list is long, expensive, and largely preventable (which you probably won’t hear from a doctor).
For men AND women there's a second issue with too much body fat. Fat tissue converts testosterone into estrogen. The more of it you carry, the more of that conversion happens, and the less testosterone you have left to work with. Then low testosterone makes it harder to build muscle and easier to store fat. It's a loop, and it tightens on itself.
(If you haven't read it: what a "normal" testosterone panel misses.)
So: more muscle, less fat. Simple to say.
The part that's actually hard
Losing weight is easy. Losing fat while keeping (or building) muscle is hard. That's the real objective, and hitting it takes a system:
Resistance training with enough stimulus to justify keeping the tissue you have
Protein high enough, and distributed well enough, to support it
A deficit aggressive enough to work and conservative enough not to eat your legs
Sleep, because you can't recover from what you don't sleep through
Nutrient timing, stress management, and everything else that moves the needle at the margins
Here's the uncomfortable part. You can think you’re doing all of it and still get it wrong. The deficit might be too steep. Protein might be too low for your training volume. You could be dropping a pound and a half a week, feeling fantastic about it, while a third of that is coming out of your quads.
The scale can't tell you. The mirror can't tell you either. It lags weeks behind reality and lies to you based on lighting and mood. Your bathroom smart scale definitely can't tell you.
The only way we know we're getting it right is an accurate body composition analysis: DEXA
What a DEXA scan actually is
DEXA, or dual-energy X-ray absorptiometry, passes two X-ray beams at different energy levels through your body. Bone, fat, and lean tissue each block those beams by measurably different amounts. The machine reads what makes it through, pixel by pixel, and reconstructs a map of what you're made of.
It doesn't estimate. It doesn't feed your height and weight into a population algorithm and hand you an average. It measures.
What comes back:
Total body fat percentage
Fat mass and lean mass in pounds
Regional breakdown. Left arm vs. right arm, left leg vs. right leg, trunk. Asymmetry shows up here that you'd never catch in a mirror.
Visceral adipose tissue. The fat packed around your organs. You can't see it, you can't pinch it, and it's the single fat depot most tightly linked to metabolic disease.
Android/gynoid ratio. How much of your fat is stored around the midsection versus the hips and thighs. One of the most underrated long-term health markers on the sheet.
Bone mineral density, with a T-score and Z-score
What the test is like to get
Underwhelming. It literally feels like nothing is happening.
You lie on a padded, open table, fully clothed, in something without metal. No underwire, no zippers, no belt buckle. An arm passes over you. It never touches you, you're never enclosed, and there's essentially no noise. Six to twelve minutes depending on your size. The whole appointment usually runs 15 to 30 minutes, and most facilities have your results in your inbox before you're back to your car.
Radiation: a whole-body scan runs somewhere around 1 to 5 microsieverts. You absorb roughly 8 to 10 microsieverts a day just from existing on planet Earth. A chest X-ray is 20 to 50. A cross-country flight is about 40. You could scan quarterly for the rest of your life and never accumulate a dose worth thinking about. The one exception is pregnancy. No X-ray of any kind.
Cost: generally $75 to $150 per scan, sometimes less. Mobile and standalone providers are cheapest; hospital-affiliated programs run higher. In Denver you can get one for under a hundred bucks. That's less than most people spend monthly on supplements they've never verified the effect of.
Why not your smart scale
Because it's guessing, and not well.
Bioimpedance sends a small electrical current through your body and estimates composition from how quickly it travels. Current moves faster through water and muscle than through fat. An algorithm built on population averages turns that reading into a number. Which means the number depends on your hydration, your last meal, your last workout, the calluses on your feet, and how closely you resemble the population that algorithm was built on.
DEXA's precision error for body fat percentage sits around 1 to 2%. Consumer bioimpedance runs 3 to 10%. A study in JMIR tested three popular smart scales against DEXA and found fat mass errors in the range of five to ten pounds. The authors' conclusion was blunt: smart scales are not accurate for body composition and shouldn't replace DEXA.
Here's what that looks like in the real world.
I'm working with a client right now who has lost more than 40 pounds over the last several months. He’s done serious work, gotten serious results. His primary goal at the moment is hitting 15% body fat, and his home bioimpedance scale told him he'd arrived. So I said it’s time to confirm… get a DEXA scan.
His DEXA put him at 19.6%.
That's not a rounding error. That's the difference between "we're done here, let's shift to maintenance" and "we have another block of work in front of us." Without the scan, he stops four and a half points short of his own goal, congratulates himself, and then spends six months wondering why he doesn't look the way he expected to.
At his size that error is close to 10lbs of fat.
Kind of explains why he wasn’t as thrilled about his aesthetics as he’d expected. He’s back on the hunt, and making moves.
What testing catches that nothing else does
The 19.6% story is what happens when the scan tells you there's more work to do. Here's the opposite case, where the scan is the only thing telling you the work is succeeding
Jen is 44. She came to us already fit, already lean, already disciplined. In a series of scans, the first read: 18.9% body fat, 99.7 lbs of lean mass, 24.6 lbs of fat mass, 130.3 lbs on the scale
Seven months later she weighed 131.6.
She gained a pound and a half.
By every metric available in a normal person's bathroom, that program failed. Seven months of consistent training and eating, and the number went the wrong direction. That is exactly the point where most people quit, blow up their diet, or go looking for a new coach.
Here's what the DEXA actually said:
| Before | After | Change | |
|---|---|---|---|
| Weight | 130.3 lbs | 131.6 lbs | +1.3 |
| Lean mass | 99.7 lbs | 101.8 lbs | +2.1 |
| Fat mass | 24.6 lbs | 23.7 lbs | -0.9 |
| Body fat | 18.9% | 18.0% | -0.9 pts |
She added two pounds of muscle, dropped a pound of fat, and got leaner, all while the scale went up. That's not a consolation prize. That's the hardest outcome in the sport, and it is completely invisible without a scan.
And the scale number wasn't even the interesting part.
Her visceral fat went to zero. Not "improved." Straight up Zero. The fat depot most tightly linked to metabolic disease, gone.
Her android/gynoid ratio dropped from 0.45 to 0.34. Less fat stored around the abdomen, more favorable distribution. This is one of the strongest long-term health signals DEXA produces and there is no way to see it in a mirror.
Her arm asymmetry closed. She came in carrying a 0.5 to 0.7 lb difference in lean mass between arms. She finished at 0.1 to 0.2. That's not vanity. Asymmetry that size shows up as compensation patterns, inefficient force production, and eventually injury. We only knew to program around it because the regional data flagged it.
Not one of those five findings is detectable by a scale, a mirror, a tape measure, or how her clothes fit. Without the scan, Jen's seven months of excellent work reads as "gained a pound." With it, we could see exactly which levers were working and keep pulling them.
Read the full breakdown: Leaner, Stronger, and More Balanced: An Optimization Success Story
Your body doesn't fall apart evenly
Jen's arm gap closed. Mine opened.
Some context first. Between my two scans I herniated my L5 disc badly enough that I couldn't train for close to a year. Back, hip, and leg pain running down my left side. So my totals over those seventeen months read exactly the way a year off reads:
| Aug 2024 | Jan 2026 | Change | |
|---|---|---|---|
| Body fat | 16.0% | 17.4% | +1.4 pts |
| Fat mass | 33.0 lbs | 35.8 lbs | +2.8 |
| Lean mass | 165.0 lbs | 162.6 lbs | -2.4 |
I'm the coach, and I lost two and a half pounds of muscle and put on nearly three pounds of fat. I'm publishing it anyway, because that's the entire point of the tool. It doesn't grade you on effort or circumstances or how good your reasons were. It tells you what happened.
But the totals only tell you how much. They don't tell you where, and where turned out to be the more interesting question.
Two numbers moved in the right direction during the worst training year I've had since my twenties.
Visceral fat went down. I gained 2.8 lbs of fat and none of it landed in the compartment that matters. My VAT volume dropped from 24.17 to 22.82 cubic inches. That's the fat wrapped around the organs, the depot actually driving metabolic disease, and it shrank while my total fat mass climbed.
My android/gynoid ratio improved. It went from 1.41 to 1.30. That's the split between fat stored around the abdomen and fat stored at the hips and thighs. Lower is better and the target is under 1.0, so no, I'm not there. But I moved toward it during a year I couldn't train.
I can't run a controlled experiment on myself, so treat the mechanism as a reasonable inference rather than proof. But it tracks with what we know about where the body puts fat. Poor sleep, chronic stress, alcohol, and a garbage diet all push storage toward the visceral compartment. Protein-forward whole food, real sleep, and managed stress don't.
Exercise was off the table for a year. Nutrition and sleep weren't. So I held those, hard. And the scan says the fat I gained went to the least dangerous place available to it
That's the case for reading more than one number. If I'd tracked body fat percentage alone, that year reads as pure failure. The full report says something more useful. I paid the muscle and fat cost that a year of immobility charges everyone, and the systems I could still control kept the bill from landing where it would have done real damage.
The real problem showed up somewhere else entirely.
I've had a complete patellar tendon rupture on my left side, surgically repaired. Same side the disc lit up. I am not confused about whether I have a bad leg. What I had no way of knowing was how much ground it had given up.
| Left leg lean | Right leg lean | Gap | |
|---|---|---|---|
| Aug 2024 | 26.9 lbs | 26.7 lbs | 0.2 lbs |
| Jan 2026 | 25.1 lbs | 27.5 lbs | 2.4 lbs |
In August 2024 my legs were functionally even. Seventeen months later there's a 2.4 lb, 9.5% gap, and it isn't because the right leg got bigger. The right added 0.8 lbs. The left lost 1.8. My surgical, symptomatic side atrophied while the fat percentage in that same leg climbed.
For context, my report notes that legs commonly carry imbalances of up to 1.5 lbs. Mine is well past that.
This is what an unconscious compensation pattern looks like on paper. A year of pain down one side, and my body quietly rerouted the work. Every stair, every rep, every time I pick up my kid, load goes to the side that doesn't hurt. That's not a stable arrangement. That's how the other side becomes the next problem.
Nothing else would have caught it. Not the mirror. Not the scale. Not how I feel in a session, because compensation is specifically the thing that makes you not feel it. It took a scan to turn "my left leg is a little off" into a measured 1.8 lb deficit with a cause and a programming fix.
The metric nobody asks for
Nobody books a body composition scan to learn about their bones. Then they read the report and find the most important page in it.
Bone density is what DEXA was originally built to measure. Body composition came later. So every body comp scan hands you bone mineral density along with everything else: a T-score comparing you to a healthy 30-year-old, and a Z-score comparing you to your own age group.
One honest caveat. Total body BMD from a composition scan isn't the same exam as a diagnostic bone density scan, which images the femoral neck and lumbar spine specifically. That's the test used to diagnose osteopenia and osteoporosis, and it's typically ordered by a physician. Total body BMD is a monitoring tool, not a diagnosis.
But monitoring is the entire point.
Peak bone mass arrives in your mid-to-late thirties. Everything after that is managed decline. If your first bone density number shows up at 68 because you fractured a wrist, you have a diagnosis and zero trajectory. If it shows up at 40, you have a trajectory and twenty-five years to change it.
And it does change. We've walked multiple clients from osteopenia back into normal range without medication. The protocol isn't exotic:
Protein high enough to actually support the tissue
Calcium, vitamin D, and the supporting nutrients dialed intentionally rather than hoped for
Resistance training with real progressive overload, because bone remodels in response to load
Then we retest. And the density number moves. Not a feeling, not a hope. A number on a report that reads differently than it did a year ago.
For whatever it's worth, this is the one page of my own report I'm happy with. My total body BMD came back at a T-score of 2.1 and a Z-score of 1.4. That's meaningfully above the average healthy 30-year-old, not just above my own age group.
A surgeon once told me my leg bones were among the densest he'd ever worked on. He wasn't paying me a compliment at the time. He was explaining why they'd destroyed several drill bits and anchors trying to get hardware into my femur.
Fifteen years of loading them did that. Bone responds to what you ask of it.
So what's a good number?
There's far more individual variation here than the internet admits. Humans come in a lot of shapes and a great many of them are healthy.
That said, here's roughly where we aim:
Men: 12 to 17%
Women: 18 to 22%
And lower is not automatically better. Push below those ranges and you generally start paying for it: reduced capacity to handle training load, and compromises to hormonal health. There's a floor past which "leaner" stops being a health win and becomes a performance and endocrine tax. Chasing a number below it is how people end up looking better in photos and worse in every metric that actually matters.
A scan is only useful if you scan the same way every time
One DEXA is a snapshot. What we actually care about is the trend, and a trend only means something when the conditions are controlled.
So we standardize:
Same facility, same machine. Absolute body fat percentage can differ between manufacturers. Comparing a scan from one system to a scan from another is comparing two different rulers.
Same time of day, fasted.
Consistent hydration. Not depleted, not loaded.
No training beforehand.
Do that and the number is trustworthy enough to make real decisions with. Skip it and you're paying to generate noise.
We scan at the start of the program and every six months after. That cadence exists for a specific reason. It catches a change in trajectory while the correction is still a small tweak. Wait two years between scans and you're not adjusting anymore. You're doing a rebuild.
We work with clients across the country and, occasionally, outside it. Finding a facility isn't your problem to solve alone. Tell us where you are and we'll help you find somewhere to test.
Where DEXA stops
DEXA is the best practical tool we have for measuring body composition. It is not the only thing we measure, and it's worth being clear about why.
DEXA tells you how much muscle you have. It doesn't tell you what that muscle can do. Function is what we're actually after. And remember, strength declines two to five times faster than mass, and in the research, loss of strength predicts disability and death more consistently than loss of muscle does.
So the scan is one input. Blood work is another. Performance testing is another. A client who added four pounds of lean mass but didn't add a pound to any lift is telling us something very different from one who added two pounds and put 40 on their deadlift.
The scan keeps us honest about the tissue. The rest keeps us honest about the function.
The number you don't want to see
Let's name the actual objection, because I hear it constantly.
A lot of clients don't want to scan at the beginning. Not because they doubt it works. Because they already know what it's going to say, and there's shame attached to seeing it written down. It's one thing to suspect you're in rough shape. It's another to hold a five-page report that quantifies it.
I understand it completely. I published my own bad scan earlier in this post. A year of lost muscle and gained fat, in a piece I wrote to sell coaching. It's not a fun number to put in writing.
And I'll tell you what those same clients tell me on the other end of coaching, almost word for word:
I wish you'd made me do it.
Because when you've done the work, six months or twelve months of real change in how you look and move and sleep, you want the receipts. And the clients who skipped the baseline can't have them. They've got an ending with no beginning. They know something changed. They can't tell you what, or how much, or where.
That first number isn't a verdict. It has no opinion about you. It's a starting coordinate, and it's the only thing that makes the second number mean anything at all.
Take the bad scan. It's the price of admission for the good one.
The point
You can't manage what you don't measure. And you can't measure body composition with a scale, a mirror, or a good feeling about how the week went
Effort is almost never the problem. Our clients work hard. What's missing is precision. Knowing whether the work is producing the specific result you're after, and having someone who can adjust the plan before you spend a month heading the wrong direction.
A DEXA scan costs about what you'd spend on a decent dinner and takes less time to complete. It's the cheapest insurance in fitness against spending six months getting smaller instead of getting better.
If you want to know what you're actually made of, and what to do about it, let's talk.