Does Yoga Build Bone? What Dr. Fishman's Study Really Shows - and Where It Stops

You heard it from a friend, or a headline, or a hopeful corner of the internet: a doctor proved yoga reverses osteoporosis. And for a moment something in you exhaled - because if that's true, maybe you can skip the heavy lifting, the machines, the medication you've been dreading, and just roll out a mat for twelve minutes a day.

You're not naive for hoping that. The research is real, the doctor is real, and yoga really does more for your bones than walking ever will. So this isn't here to take your hope away. It's here to show you exactly how far that hope can carry you - and where it quietly runs out.

Because there's a villain in this story, and it isn't yoga - and it certainly isn't Dr. Fishman. It's believing a claim because it's worded to sound proven, when the data quietly says otherwise. "A study shows yoga reverses osteoporosis" sounds like settled science. But as you'll see, the study behind it is thin, and the bone change it measured is so small that a DEXA scan can't even tell it apart from its own margin of error. That's the trap: a sentence that is technically defensible yet, in the way that actually matters for your bones, false - comfortable enough to rest on for years while your skeleton, unconvinced, keeps slipping. It's how a lot of women find out, too late, that "doing something" wasn't the same as doing the thing that actually builds bone.

Here is the whole story in a single sentence - the study's own. The "50%" that everyone repeats reads, in full [1]:

"Mean rate of improvement in total hip BMD increased from baseline by 50% after a mean 23 ± 2 months of yoga, but wide variation in individual improvement caused the confidence intervals to overlap, thus precluding statistical significance."

The first half is the headline. The second half - "precluding statistical significance" - is the study itself noting the finding did not hold up. And the 50% was a change in the rate of bone change, not a 50% increase in bone. To his credit, Dr. Fishman put the caveat right there in the same sentence. The misdirection happens later, every time someone quotes the impressive first half and quietly drops the second. Read the sentence to its end and the headline dissolves. The rest of this article is just unpacking that one sentence.

The headline that gives everyone hope

Dr. Loren Fishman is a respected physiatrist (a physician specializing in physical medicine and rehabilitation), and his interest in yoga for bone is serious, not a gimmick. In 2016 he and his colleagues published a study with an irresistible title: "Twelve-Minute Daily Yoga Regimen Reverses Osteoporotic Bone Loss" [1].

The premise is genuinely clever. Certain yoga poses pit one muscle group against another and hold the body in loaded positions - which means the skeleton experiences real mechanical strain, the kind of signal bone responds to. Participants did a 12-pose sequence, and the researchers tracked their bone density with DXA scans over about two years. The reported finding: bone density went up in people who stuck with it.

Give this its full due. It loads bone more than a walk does. It's gentle, accessible, and nearly free. And yoga's balance and flexibility benefits are real and valuable - better balance means fewer falls, and since most fractures begin with a fall, that alone protects you. Dr. Fishman is right about something most people get wrong: yoga is not nothing for your bones.

So the honest question isn't "does it do anything?" It's "how much, for whom, and how long does it keep working?"

What the study actually found

This is where the study itself asks you to read past its own title - and it gives you everything you need to.

Start with the numbers. 741 people enrolled. 227 were compliant enough to count. And just 43 had complete before-and-after DXA data - the subset the strongest claims rest on [1]. And here's the honest headline: across those two years, there was no measurable gain in bone a scan could call real - nothing the data can confirm was rebuilt. On that, the numbers are firm. Whether the group held perfectly steady or kept slipping a little, the scans genuinely can't say - but a real increase in bone density is the one thing the data give no support to.

Why only "no measurable change," and not "improvement"? Because of a number every bone specialist knows but most headlines skip: the Least Significant Change (LSC). Every DEXA machine has one - the smallest difference between two scans that can be trusted as a real change rather than the machine's own measurement wobble. By the international standard, only a change that meets or exceeds the LSC counts as genuine; anything smaller sits inside the margin of error [2]. At the spine and hip that threshold is typically around 3 to 5%. Fishman's reported change worked out to roughly two to three percent across two years - at or below the LSC. In plain terms: on any individual woman's scan, that "improvement" can't be told apart from no change at all. The study surfaced a faint signal only by averaging the results across the 43 participants with complete before-and-after DXA data; for any one person actually being scanned, there was no change you could call real.

And even "maintenance" is the generous reading. Two things are worth being honest about. First, because the change sits inside the margin of error, the data can't separate a small gain from a flat line from a slow continued loss - to the scanner, all three are the same result, so "maintained" is the best case, not a measured fact. Second, the "they were declining before yoga" comparison rests on participants' own earlier scans - retrospective, often on different machines, with no controlled baseline - so even the starting trend is shaky. None of this means yoga did nothing. It means the most you can honestly claim is no change a DEXA could detect - which is a long way from rebuilding bone.

That's also where the eye-catching "50% improvement" comes from - the sentence we opened with. Remember how it ended: "precluding statistical significance." That 50% was a change in the rate of hip density, off a tiny base, and the paper's own clause rules the finding out as non-significant. It is not a 50% increase in bone density. The headline survives only by quoting the first half of that sentence and dropping the second.

Which points straight at that irresistible title. This is our interpretation, not the study's: a title like "Twelve-Minute Yoga Regimen: No Significant Change in Bone Density" would have tracked the findings far more closely than "Reverses Osteoporotic Bone Loss." Even the best-case reading - that committed practitioners held the line - would be worth something on a disease defined by relentless loss. But the data can't confirm even that, and they clearly rule out the one thing the headline sells, and the one thing a woman staring at a worrying DEXA is hoping for: bone actually rebuilt.

So the fair reading is this: in committed people, yoga was associated with no measurable change in bone over a couple of years - no detectable gain, and possibly a slow continued loss the scan couldn't rule out. That may still beat doing nothing, and it's all the data show. It is not the reversal the headline implies. That's not a knock on Dr. Fishman; it's simply what the numbers say.

Why yoga genuinely helps - and exactly where it stops

Here's the part that ties this article to every other one we've written about movement and bone.

Yoga helps because, for a while, it's a new and real demand on your skeleton - especially if you've been sedentary. Your bones feel poses they've never held, and they respond. That's the same reason walking can briefly build bone in a deconditioned person: almost any new stimulus produces some adaptation.

And then the same thing happens that happens with walking, with vibration plates, with any unchanging routine. Your body adapts to the poses. The sequence that challenged your skeleton in month one is, by month twelve, just your familiar Tuesday. Bone builds when it's loaded harder than it's used to - and a fixed routine, by definition, stops being harder than you're used to. We unpacked this in the adaptation principle and why progress plateaus: your body changes to match whatever you ask of it, and once it has, the osteogenic stimulus that same unchanged load provides is expected to diminish.

There's a sharper edge to this, too. Because your body keeps adapting, the same routine that held the line this year may not even manage that a few years from now - the stimulus that was once just enough to maintain becomes, to an adapted body, not quite enough. That's not a flaw in yoga. It's simply the boundary of what any fixed-stimulus practice can do.

What progressive loading adds - and who each path is for

If yoga's limit is that it can't escalate, the fix has to be a stimulus that does - safely, and for years.

That's osteogenic loading. As we explain in the science of osteogenic loading, the goal is to push your skeleton above what it's already adapted to, then keep raising that load as you get stronger - and to do it in a controlled way that protects your joints and spine. On OsteoStrong's robotic Spectrum devices, you push against a device that doesn't move and produce force in multiples of your body weight from a coached position, in about 15 minutes a week, and the force you generate is measured every single visit - so you can actually see yourself exceeding your habitual load instead of hoping you are. It's the dial that yoga doesn't have.

And here's the difference in how each path proves itself. Dr. Fishman's evidence is one self-selected study with 43 fully-documented participants. Osteogenic loading's evidence is a real-world track record: the method rests on 152 peer-reviewed studies, it's been used by 65,000+ members worldwide, more than 100 Austin-area physicians refer their patients to our centers, and among members who follow the weekly protocol, 8 out of 10 see measurable DEXA gains on their next scan. To be clear about rigor: that last figure isn't from a published trial - it's our own members' follow-up DEXA results, real-world rather than lab-controlled, which is exactly the kind of repeated, ordinary-people proof a single study can't give. You can read some of those in our members' real DEXA stories. Fishman's study is one hopeful signal; OsteoStrong's track record is the same result showing up again and again, in ordinary people, over years.

So who's each path for?

• Keep yoga if you love it - for balance, mobility, calm, and as a genuine adjunct. If you're early, deconditioned, and just starting to move, it's a real first step, and it lowers your fall risk, which matters enormously. See fall prevention that actually works.
• Add progressive loading if your goal is to reverse established osteopenia or osteoporosis - to move your DEXA, not just hold the line. That's the job yoga isn't built for, and it's what women over 50 are actually accomplishing with a weekly osteogenic-loading protocol.

This isn't yoga or bone-building. It's yoga plus the one thing yoga can't do.

Yoga vs. progressive osteogenic loading, at a glance

What's actually at stake while you decide

Here's what the comforting headline never mentions: the cost of waiting. Bone you don't load this year is bone you don't get back next year - the window doesn't reopen. And the sequence that ends independence rarely starts with a diagnosis. It starts with a fall, then a fracture, then a hospital stay, then the slow loss of doing things on your own terms.

So picture two years from now. In one version, you kept your yoga and added a load that climbed week over week - and your next scan finally moves the right direction, and you step off a curb or lift a grandchild without that flicker of fear. In the other, you rolled out the mat faithfully, certain you had it handled, and found out at your next DEXA that "doing something" hadn't been enough. Same effort, same good intentions - very different skeleton.

You get to choose which story your bones tell. Keep the yoga. Add the thing that escalates.

Find out what your bones actually need

You don't have to guess whether yoga alone is moving your numbers. The free 15-minute Bone Health Call is the first step: we look at your DEXA, your current routine, and what progressive loading could add - then measure your baseline at your first visit so you can watch it climb. No pressure, phone or Zoom.

Related reading:

• Does Walking Build Bone? Why It Works Until It Doesn't - the adaptation principle behind this whole article
• Onero vs. OsteoStrong: Why Strong Bones Need More Than a Heavy Barbell
• The Science of Osteogenic Loading, Explained
• What I Told a Room of Biohackers and Longevity Doctors About Bone Health
• Reversing Osteoporosis: What Women Over 50 Actually Accomplish

References

1. Lu Y-H, Rosner B, Chang G, Fishman LM. Twelve-Minute Daily Yoga Regimen Reverses Osteoporotic Bone Loss. Topics in Geriatric Rehabilitation. 2016;32(2):81-87. doi:10.1097/TGR.0000000000000085
2. International Society for Clinical Densitometry (ISCD). Precision Assessment and Least Significant Change. The LSC is calculated as 2.77 × the precision error and is the smallest change in BMD that can be considered statistically significant at 95% confidence; only changes meeting or exceeding the LSC should be interpreted as real when monitoring an individual over time. https://iscd.org/learn/resources/calculators/

This article is for general education and is not medical advice. We are not affiliated with Dr. Fishman or his yoga program, and we've represented his published findings - and their stated limitations - as faithfully as we can. Talk to your own doctor before starting, stopping, or changing any exercise program, especially if you have a diagnosis of osteopenia or osteoporosis.

Frequently asked questions