Flexion, Fear, and the Fine Print

02 Nov Flexion, Fear, and the Fine Print

A new systematic review of yoga injuries has been released, and one line from a summary of it on the website Yoga Research and Beyond (YR&B) is being quoted as evidence for something some respected yoga professionals have argued for years: that older adults with osteoporosis should avoid spinal flexion. The line from the YR&B summary is: “the largest single event was spinal compression fractures from forward bending in older adults with low bone density“.

I’ve challenged that position about avoiding spinal flexion before, and I’m not the only one.

I read this new research paper and the paper suggests something quite different from what it’s being used to affirm.

These case reports do tell us about who may need care. But they do not tell us that flexion is dangerous for everyone with an osteoporosis diagnosis, and I’d argue that teaching it that way can cause its own kind of harm. Ahimsa cuts both ways.

What the paper actually is

The paper is Sharma et al., “Yoga and its risks: A systematic review of adverse events reported in case reports and case series,” published in the Journal of Integrative Medicine in 2026.

The authors searched the medical literature from its beginnings through April 2025 and collected every published case report and case series describing a health problem linked to yoga.

They found 84 adverse events across 70 studies, published between 1971 and 2025. Broken down by body system:

Body system	Cases	Share
Musculoskeletal	34	40%
Ophthalmological (eyes)	16	19%
Cardio-respiratory	12	14%
Neurological	10	12%
Psychiatric	6	7%
Nasopharyngeal / gastrointestinal	4	5%
Dermatological	2	2%

The practices the authors highlight most are headstand (15 cases, mostly eye problems in people who already had glaucoma or other eye conditions) and forceful pranayama (8 cases). Over half of the people affected had a pre-existing vulnerability.

It’s worth sitting with the scale for a moment. They found 84 cases over 54 years, from a practice that the authors estimate around 300 million people worldwide take part in. Which is not to say those 84 people don’t matter. They do. But it frames what kind of evidence we’re looking at.

Where that quote actually came from

That line about “the largest single event…” isn’t in the research paper. It’s from the summary on Yoga Research & Beyond, a website that translates yoga research for teachers.

I suspect most people sharing it have only seen the YR&B summary article (the paper itself is behind a paywall), so they’re taking the summary at its word, which is understandable because most of us don’t have journal access, and journalistic summaries are how research reaches teachers.

And to be fair, the YR&B summary is a good one. It’s careful to preserve the uncertainty shared in the paper. In the very same paragraph as the compression fracture line, it says that case reports can show what serious harms look like and who they tend to affect, but cannot say how often they happen. It also has a whole section warning that the paper could easily be misread, pointing out that a case-report review of running or gardening would produce an equally alarming list.

So the sentence that is getting shared is real, but it was lifted out of a paragraph whose next sentence undercuts the way it’s being used and interpreted.

And the research paper’s own authors never even make that claim! They don’t rank event types at all.

A closer look at the 12 compression fractures

Counted by event type, vertebral compression fractures are indeed the largest single group in the review: 12 of the 84 cases. So the summary isn’t wrong on the arithmetic. But when you actually open Table 2 and read the cases one by one, the picture gets a lot more textured and nuanced.

Source	Age / sex	Pre-existing conditions (as listed in the review)	Movement reported	Fracture
Sinaki 2013	87 F	Degenerative spondylosis, advanced osteopenia	Spinal flexion, halasana (with instructor)	VCF
Sinaki 2013	61 F	Hyperlordosis, spondylolysis, spondylolisthesis, severe disc disease, osteopenia	Spinal flexion, halasana, paschimottanasana	T4
Sinaki 2013	70 F	Osteopenia, spondylosis	Setu bandhasana (bridge) and halasana	T8, T9
Sfeir 2018	53 F	Early menopause, no HRT	Neck and upper back flexion	T8
Sfeir 2018	87 F	Prior wrist fracture, on risedronate	Lifting legs while flexing spine	L2
Sfeir 2018	61 F	On alendronate 10 years	Balance on head while flexing neck and upper back	T4
Sfeir 2018	69 F	Prior T8 fracture, alendronate 8 years	Spinal flexion	T6, L3
Sfeir 2018	68 F	Multiple prior VCFs, hyperparathyroidism	Neck and spinal flexion, bending forward	C5
Sfeir 2018	70 M	None listed	Spinal flexion	L1
Sfeir 2018	66 F	Postsurgical hypoparathyroidism	Thoracic and lumbar flexion and hyperextension	T11
Sfeir 2018	59 F	On prednisone	Spinal flexion	L1
Sfeir 2018	55 F	None listed	Spinal flexion	L1

VCF = vertebral compression fracture. Source: Table 2 of Sharma et al. 2026.

A few things jump out at me.

All 12 come from one place. Both case series come out of Mayo Clinic, and Dr. Mehrsheed Sinaki is an author on both. That’s not a knock on her work (she has spent a career on bone health and exercise), but it does mean the “largest single event” in a worldwide review is really one clinical group’s referral population.

Some patients may be counted twice. The 2013 series has an 87-year-old woman and a 61-year-old woman with a T4 fracture. The 2018 series also has an 87-year-old woman and a 61-year-old woman with a T4 fracture (and Sinaki is an author on both papers). I can’t confirm they’re the same people, and the review doesn’t address it. But if they are, the real number is closer to ten.

It isn’t all forward bending. Halasana (plough) is a loaded inversion with the potential of having the full weight of the legs levering on the upper spine. Setu bandhasana (bridge) is an extension pose. One case was “balancing on the head while flexing the neck and upper back”. Another combined flexion with hyperextension. So when we say “forward bending” it flattens all of that into something that sounds like it could pertain to a gentle seated fold, but that’s not what the evidence suggests.

Most of these people had a lot more going on than low bone density. Prior vertebral fractures. Glucocorticoids. Parathyroid disease. Long-term bisphosphonates. Severe degenerative spine changes. Ages up to 87. And two of the twelve are listed with no pre-existing condition at all.

So what does this cluster actually tell us?

I think it tells us that people with fragile spines, especially those with prior fractures or other big risk factors, can get hurt by loaded, end-range, or inverted flexion. This is worth knowing for sure, but let’s be clear that it’s a much narrower statement than “all older adults with low bone density should avoid spinal flexion.”

Case reports are not facts about risk

The part I really want yoga teachers to understand, because it applies far beyond this one paper is that case reports and case series cannot establish causation. I know these are scary situations to consider but they can only describe a pattern and raise a question worth studying. That’s it.

Why can’t they establish causation?

• There’s no comparison group. We don’t know whether people with osteoporosis who practice those specific flexion poses fracture more often than people with osteoporosis who don’t.
• There’s no denominator. We don’t know how many thousands of people did the same poses with no problem at all.
• Timing isn’t cause. This review’s own inclusion criteria accepted events that were linked to yoga in time or causally. In other words, an incident happening around a yoga session was enough to count.
• Publication bias. The dramatic, unusual cases get written up. A thousand uneventful forward folds never make it into a journal.

This matters especially for compression fractures. In people with osteoporosis, vertebral fractures classically occur during ordinary daily activities like bending, walking or lifting something light. Only about one in five follows a fall, many people can’t recall any specific event, and, here’s a key point: only about a quarter to a third are ever clinically recognized. So, a fracture found after a yoga class could have been caused by the class, aggravated by it, or simply discovered because the person went in for a sore back afterward.

To be fair, case reports aren’t worthless. Loading the front of a fragile vertebra through deep flexion is biomechanically plausible, and a repeating pattern does deserve attention.  In fact, the everyday-fracture research is not a simple story. Flexion under load is really a concern for very fragile spines, and that same loading happens constantly in daily life not just on the mat.

To me, that’s an argument for teaching people in a mindful yoga class how to bend skillfully during yoga rather than to 100% avoid bending. Plausible plus repeating adds up to “worth caution and further study,” not “proven.”

The authors of the research themselves describe their findings as hypothesis-generating rather than definitive, and say that frequency, severity and causality can’t be estimated from this kind of evidence. Their recommendation is pre-practice screening and tailored instruction, not avoiding a whole category of natural movement.

The study behind the warning

If you trace the “avoid flexion” advice back far enough, almost all of it leads to one paper: Sinaki and Mikkelsen’s 1984 study in the Archives of Physical Medicine and Rehabilitation. It’s cited constantly and, I suspect, rarely read. I have a copy and I’ve written about its flaws many times. Let’s look at the details again here.

The study looked at 59 women aged 49 to 60, seen at Mayo Clinic between 1969 and 1981, with back pain and postmenopausal spinal osteoporosis diagnosed from spine X-rays (this was before DXA), and many, though not all, already had collapsed vertebrae.

Each had been given one of four exercise programs, and follow-up X-rays were compared with the first ones for any “further” wedging or vertebral compression fractures (VCF). That’s the paper’s word, and its table simply labels each patient “same” or “worse.” Since many of these women already had fractures when they started, “further” can include existing fractures getting worse, not only brand-new ones:

Group	Patients	Further fractures or wedging
Extension exercises	25	16%
Flexion exercises	9	89%
Extension and flexion	19	53%
No exercise	6	67%

That 89% is a striking number, and I understand why it strikes fear. But look closer.

• It wasn’t a trial. It was a retrospective review, and patients weren’t randomized. Each physiatrist simply prescribed their usual program. We have no way of knowing whether the women given flexion were frailer, more fractured, or in more pain to begin with.
• The flexion group was nine women. The 89% is eight of them.
• “Flexion” meant sit-ups and a slumped, seated toe-reaching stretch. This is repetitive loaded trunk flexion in women with significant osteoporosis (many with existing VCFs). These are not a hip-hinged gentle forward fold or even a breath-controlled fold a la Viniyoga.
• The no-exercise group did almost as badly, at 67%. This is the detail almost nobody mentions. The paper’s significant comparisons are all against the extension group. So you could read the same data as “extension strengthening protected people” just as easily as “flexion hurt them.”
• Follow-up varied from one to six years, with the no-exercise group followed longest, and compliance wasn’t measured.
• Everyone got the same daily-life advice: avoid lifting more than 10 pounds (another outdated warning. More on the LIFTMOR trial below), and bend with the knees rather than the spine, which is “how to bend” advice, not “never bend.” But daily life activities that include bending are automatic, so compliance on this advice would be difficult to measure.

To its credit, the X-rays were compared without knowing which program each woman had followed. And I don’t think the paper should be thrown out. It actually tells us that giving sit-ups to women with significant osteoporosis is a bad idea. And again, that’s simply a much narrower statement than “spinal flexion is dangerous.”

What about the 2002 follow-up?

The paper usually offered as confirmation is Sinaki and colleagues’ 2002 study in Bone. Women who had done two years of prone back-extensor strengthening (lifting a weighted backpack while lying face down) had fewer vertebral fractures ten years later than women who hadn’t: 6 fractures in 378 vertebrae versus 14 in 322.

It’s an encouraging finding. But it doesn’t confirm the 1984 flexion result, because nobody in it did flexion. The control group simply didn’t do the exercises. It also studied healthy women aged 48 to 65 without osteoporosis or vertebral wedging at the start. And when fractures are counted by woman rather than by vertebra, the difference (7 of 23 versus 3 of 27) wasn’t statistically significant.

So what do the two papers support together? That strengthening the back extensors appears protective. I’m completely on board with that, and it fits beautifully with a “teach people how to move” approach. What they don’t establish is that flexion itself causes fractures.

It’s also worth noticing that the 1984 study, the 2002 study, both yoga case series in the Sharma review, and, as we’ll see, the 2019 Mayo records review, all come from the same research program. That’s a testament to Dr. Sinaki’s dedication to this question.

It also means these aren’t independent confirmations of one another, even though they’re often cited as if they were.

Wouldn’t studying flexion be unethical?

When I’ve suggested actually testing yoga forward bending in people with low bone density, I’ve been told it would be unethical, given the 1984 results. I understand the instinct. But research ethics rests on genuine uncertainty, and nine women doing sit-ups isn’t strong enough evidence to settle the question.

More to the point, in a more controlled study on yoga forward bending, we would not study sit-ups, plough, shoulder stand or the other loaded and inverted positions implicated in the case series. I’d keep those out of any trial. What we could study is simple, slow forward bending, the breath-supported kind, taught by trained Viniyoga teachers to people with low bone density who’ve been screened for the higher-risk factors, like prior vertebral fractures. That’s a very different question from the one the 1984 study asked.

And, it’s important to highlight, the case reports rarely tell us anything about the teaching itself like whether anyone was supervising, or how experienced the teacher was.

There’s precedent for this. Heavy lifting was long considered too risky for people with osteoporosis (note the advice given in Sinaki, 1984, to not lift more than 10 lbs), until the LIFTMOR trial tested it. Postmenopausal women with low to very low bone mass did eight months of supervised high-intensity deadlifts, squats and overhead presses. Their bone density and physical function improved, and the only adverse event was one minor lower back spasm.

Careful, supervised research overturned an assumption that had kept people away from something that helped them. I’d love to see the same done for forward bending.

Until then, some of the best information we have comes from experienced teachers.

What experienced teachers see

Peter Blackaby is one example of someone who has taught yoga for over forty years and whose students are now mostly over fifty, and he makes a point I find compelling. If an 89% fracture rate applied to people with osteoporosis who flex their spines in yoga, teachers like him would have seen a great many students suffer compression fractures. He hasn’t.

I know this is anecdotal and not reported as a case study. His students weren’t X-rayed, and most vertebral fractures go unnoticed (something he acknowledges). So it’s a plausibility check rather than counter-evidence. But as a plausibility check, it’s a strong one.

The 89% figure came from nine women doing sit-ups. It was never a measure of what happens in a yoga room nor with a thoughtful, bone-informed, experienced teacher.

Two more studies

Two more recent studies usually come up in this conversation. They’re often presented as evidence on opposite sides, but they aren’t really the same kind of evidence at all.

The Mayo Clinic records review (Lee, Huntoon and Sinaki, 2019). Researchers looked back at the records of 89 patients seen at Mayo Clinic between 2006 and 2018 for injuries they attributed to yoga. Thirteen (about 15%) had compression fractures, and the poses most often implicated involved extreme spinal flexion and extension. Their conclusion was that people with osteopenia or osteoporosis would benefit from avoiding extreme spinal flexion. Notice that word, “extreme”.

But this is less new evidence than a wider look at the same clinic’s patients. The 2018 Sfeir case series was also a records review of Mayo patients whose back pain began after yoga, from the same institution, with Dr. Sinaki again as senior author, and its nine fracture cases fall within Lee’s 2006 to 2018 window. Neither paper says whether the patients overlap, so I can’t confirm it, but it would be surprising if they didn’t. If they do, the 13 fractures in 2019 may include most of the 12 cases in the Sharma review, and the number of distinct people behind all three papers could be quite small.

It also shares the same limits. The patients came from the same specialist musculoskeletal clinic, there’s no denominator, and the authors themselves noted the results are hard to generalize. They had follow-up on only 22 patients. Interestingly, those who modified their poses reduced their pain. (Note “modified”, not quit.)

Dr. Loren Fishman’s 12-pose study (published 2015 and 2016). This one is different. It comes from a separate research group, and rather than collecting injuries after they happened, it looked at a whole group of people with low bone density who practiced yoga. Fishman, a Columbia physiatrist who trained with B.K.S. Iyengar, followed 741 people, most with osteoporosis or osteopenia, who practiced a 12-minute daily sequence. With more than 90,000 hours of yoga logged, there were no reported or X-ray detected fractures or serious injuries related to the practice. That’s remarkable. But I’ll be honest about the study’s weaknesses too: it was retrospective and non-randomized, fewer than a third of participants stuck with the routine, and the 12 poses (tree, triangle, warrior II, side angle, revolved triangle, locust, bridge, supine hand-to-foot, twists and savasana) largely avoid loaded flexion. So it suggests that “yoga can be safe for people with osteoporosis,” not “any flexion is fine.”

So where does that leave us?

Not with two studies pulling in opposite directions, but with one research program’s patients reported several times, and one independent study that found no fractures with a carefully chosen sequence.

I think the evidence still supports caution with loaded, end-range, repetitive or inverted flexion in people at high fracture risk.

It does not support telling everyone with the diagnosis to fear bending forward.

So is yoga safe? You can’t have it both ways

The people sharing this review to argue against flexion often start by saying yoga is incredibly safe in general. I agree, actually. But that conclusion doesn’t come from this paper, because a pile of case reports can’t show that anything is safe any more than it can show something is dangerous. It comes from studies that are more quantified. Those studies are reassuring, and also point to the variability in approaches to yoga.

When Holger Cramer and colleagues pooled randomized controlled trials in 2015, yoga’s adverse event rates matched usual care and other forms of exercise. Also, in a 2019 national survey of 1,702 German yoga practitioners, Cramer’s group found about 0.6 injuries per 1,000 hours of practice, mostly minor strains and sprains, with Power yoga highest at 1.5. The studies they compare against put running at about 2.5 injuries per 1,000 hours, soccer at 3.7 and tennis at 5. Their conclusion: yoga appears to be as safe or safer than other exercise. (They did not specifically measure people with osteoporosis.)

Here’s my problem, though. If we accept the counted, denominator-based data to say yoga is safe overall, we can’t then turn around and use uncounted case reports to justify a blanket rule for millions of people. Why trust the denominator when it reassures us and ignore it when it doesn’t? Why is a cluster of twelve cases (maybe ten) from one clinic enough to warn everyone with a diagnosis, but 90,000 incident-free hours in Fishman’s cohort isn’t enough to soften that warning?

Applied consistently, the evidence leads to where the Sharma authors land: screen, stratify, and tailor.

Ahimsa cuts both ways

The argument, as I understand it, is that ahimsa obligates us to warn everyone with osteoporosis about the dangers of flexion. I do take that seriously. Non-harm is the first of the yamas for a reason, and I’d never argue for ignoring real risk.

But I think non-harm also includes not handing someone a generic fear story that may not apply to their spine.

Osteoporosis is a spectrum, not a single condition. A fit, active 60-year-old whose DXA scan just tipped over the osteoporosis threshold and a frail 87-year-old with three prior vertebral fractures on long-term prednisone share a diagnosis. They do not share a risk profile. Telling them the same thing is a bit like telling someone to “take medication” without saying which one, what dose, or for what. Look back at that table of twelve cases. Those are mostly the second person, not the first.

Words land in the body. The nocebo effect (the flip side of placebo) is well documented, especially in back pain, where fear-avoidance beliefs predict more pain and disability. I’m not aware of a study testing nocebo in osteoporosis messaging in yoga specifically, so I’ll call this a reasonable inference rather than a proven fact. But when a student hears “flexion will break your spine,” what do we imagine happens the next time she bends to tie her shoe?

Fear of falling is itself a fall risk. This one is well established in the research on older adults. People who fear falling tend to restrict their movement, lose strength and balance, and become more likely to fall. And for someone with osteoporosis, a fall is the real danger. A hip fracture is a far bigger threat to an older person’s life and independence than a forward fold in a supervised class.

In the Pañcamaya model I teach from, this is where the vijnanamaya dimension comes in, the layer of identity, behavior, discernment, and wisdom. A danger signal doesn’t stay in the annamaya (physical) dimension. It shapes how a student understands her own body, what she believes she’s capable of, and how she moves through her whole life. That’s not a small thing to get wrong.

How to move, not never move

So what would I actually suggest?

Not “flexion is fine for everyone,” and not “never flex.”

Something in between that respects both the evidence and the student.

Stratify by risk, not by diagnosis. The cases in this review point to who genuinely needs extra caution:

• A prior vertebral fracture (the single strongest signal in these cases)
• Long-term glucocorticoid use, like prednisone
• Very low bone density, or a high fracture-risk score from their doctor
• Other bone or metabolic conditions, like parathyroid disease
• Advanced age combined with frailty

For those students, I’d be very cautious, especially with loaded, end-range, rapid, repetitive or inverted flexion like plough, shoulder stand, deep seated folds with someone pressing on the back, and the like. For a fit student with osteopenia and no other risk factors, I’d teach intelligent movement, not a list of forbidden shapes.

Teach the how. It’s more than just the mechanics of “what”. Exhaling to begin the fold. Hinging from the hips may assist the process but continue through the whole spine, honoring each segment’s mobility not locking the motion in one section. Moving slowly and with control to disperse the energy through the whole spine and the whole body. Bending the knees if necessary. Supporting with props and hands. Coming out of a pose with the breath supporting the spine’s extension. Everyday life is full of flexion (gardening, picking up grandchildren, getting in and out of the car), so the goal is to make students skillful at it, not afraid of it.

Use svadhyaya and viveka. Teach Yoga. Self-study and discernment are the tools yoga already gives us for exactly this. Can a student notice where she’s moving from? Can she feel the difference between a long spine and a collapsing one? That’s empowerment, and it travels with her off the mat.

Use the breath to support the spine. Exhale to flex, inhale to extend. This is the piece I think gets the least attention. Coordinating movement with breath, especially using the exhale to engage the deep abdominal and pelvic floor support before bending, gives the spine a natural internal brace. Using the inhale to lift and extend to come out of a bend. Some lineages, like Viniyoga, emphasize this far more than others, and I think it’s one of the most useful (and least fear-inducing) things we can teach.

There’s some survey evidence that style and teaching do matter. In that same German survey, people who practiced Viniyoga had about half the odds of reporting an acute injury compared with other styles, and practicing without any teacher raised the odds by about 1.7 times. To be clear, a survey can’t prove cause. Still, it fits the idea that how yoga is taught matters as much as which shapes are in it.

And talk with their medical team. Screening is literally what the paper recommends. Asking a student about prior fractures, medications and what her doctor has said is far more useful than a generic warning. But in general, when you don’t have the opportunity to ask for this information, you can know that there are only a handful of poses that are implicated in these case studies. Take note and revise your sequencing accordingly for all students.

I know it’s easier to screen, stratify, and tailor in a private session than in a group class. And it’s fair to say yoga teachers can’t be expected to know how to work with every condition. But if there’s one condition worth our time, it’s low bone density.

Most 200-hour trainings already teach basic modifications for prenatal students, and learning the essentials of osteoporosis would take about the same amount of time. Yet a pregnant student in a general class is relatively rare, and not every woman will be pregnant. But every one of our students has a skeleton, and every one of them is aging.

The numbers make the case. By 2030, more than 20% of the US population will be over 65, and women make up the majority of that group, just as they make up the majority of yoga students. Bone loss typically accelerates after menopause. About 10 million Americans have osteoporosis, roughly 80% of them women, and another 44 million have low bone density that puts them at increased risk. Whether or not they know it yet, many of the people in our classes are among them.

And just as yoga isn’t one-size-fits-all, neither is osteoporosis. The fit 60-year-old and the frail 87-year-old share a diagnosis, not a risk profile. A blanket “don’t flex your spine” cue can’t serve them both. What serves them is cueing that teaches each of them how to move for their own body.

Some will argue that because osteoporosis is a “silent” disease, students can’t know what’s appropriate for themselves. It’s true that no one can feel their bone density. But that’s exactly why we teach interoception. A student who has learned to notice where she’s moving from, how much effort she’s using, when she’s approaching end range, and whether her breath is steady or strained has a far better safety system than one who has simply been told to fear a shape. Interoception doesn’t replace screening. It’s what makes good cueing work in a body we can’t see inside.

Reading past the headline

I believe we all want older students to keep practicing yoga, and to practice safely. Where some of us differ is on what the evidence can bear, and on whether a blanket warning is the kind version or just the cautious-sounding one.

The Sharma, et al. paper, read in full, doesn’t suggest avoiding flexion across the board. Neither does the YR&B summary, if you read the next sentence. What both point to is something more demanding of us as teachers: to know our students, know their real risk factors, and teach them how to use the tools of yoga to move well rather than to fear moving at all.

That asks more of us than a rule does. I think it’s also what ahimsa actually asks.

 

---

 

Want to learn more? Get on the waitlist for the next “How to Teach Bone-Informed Yoga“.

 

---

 

References

Blackaby, P. (n.d.). Considerations when teaching spinal flexion to people with osteoporosis.

Bone Health & Osteoporosis Foundation. (n.d.). Osteoporosis fast facts [Fact sheet]. https://www.bonehealthandosteoporosis.org/wp-content/uploads/Osteoporosis-Fast-Facts-2.pdf

Cramer, H., Quinker, D., Schumann, D., Wardle, J., Dobos, G., & Lauche, R. (2019). Adverse effects of yoga: A national cross-sectional survey. BMC Complementary and Alternative Medicine, 19, Article 190. https://doi.org/10.1186/s12906-019-2612-7

Cramer, H., Ward, L., Saper, R., Fishbein, D., Dobos, G., & Lauche, R. (2015). The safety of yoga: A systematic review and meta-analysis of randomized controlled trials. American Journal of Epidemiology, 182(4), 281–293.

Griffith, J. F. (2015). Identifying osteoporotic vertebral fracture. Quantitative Imaging in Medicine and Surgery, 5(4), 592–602. https://doi.org/10.3978/j.issn.2223-4292.2015.08.01

Lee, M., Huntoon, E. A., & Sinaki, M. (2019). Soft tissue and bony injuries attributed to the practice of yoga: A biomechanical analysis and implications for management. Mayo Clinic Proceedings, 94(3), 424–431. https://doi.org/10.1016/j.mayocp.2018.09.024

Lu, Y. H., Rosner, B., Chang, G., & Fishman, L. M. (2016). Twelve-minute daily yoga regimen reverses osteoporotic bone loss. Topics in Geriatric Rehabilitation, 32(2), 81–87. https://doi.org/10.1097/TGR.0000000000000085

Mayo Clinic. (2019). People with osteoporosis should avoid certain spinal poses in yoga, Mayo Clinic study says [Press release]. Newswise.

Rabinovitch, A. (n.d.). Is yoga risky? A new review in context of a decade of safety research. Yoga Research & Beyond.

Schousboe, J. T. (2016). Epidemiology of vertebral fractures. Journal of Clinical Densitometry, 19(1), 8–22. https://doi.org/10.1016/j.jocd.2015.08.004

Sfeir, J. G., Drake, M. T., Sonawane, V. J., & Sinaki, M. (2018). Vertebral compression fractures associated with yoga: A case series. European Journal of Physical and Rehabilitation Medicine, 54(6), 947–951. https://doi.org/10.23736/S1973-9087.18.05034-7

Sharma, S., Bhandari, S. S., Malhotra, V., Kuppusamy, M., Kumar, S., & Dhakal, M. (2026). Yoga and its risks: A systematic review of adverse events reported in case reports and case series. Journal of Integrative Medicine, 24(5), 667–682. https://doi.org/10.1016/j.joim.2026.07.001

Sinaki, M. (2013). Yoga spinal flexion positions and vertebral compression fracture in osteopenia or osteoporosis of spine: Case series. Pain Practice, 13(1), 68–75.

Sinaki, M., Itoi, E., Wahner, H. W., Wollan, P., Gelzcer, R., Mullan, B. P., Collins, D. A., & Hodgson, S. F. (2002). Stronger back muscles reduce the incidence of vertebral fractures: A prospective 10 year follow-up of postmenopausal women. Bone, 30(6), 836–841. https://doi.org/10.1016/S8756-3282(02)00739-1

Sinaki, M., & Mikkelsen, B. A. (1984). Postmenopausal spinal osteoporosis: Flexion versus extension exercises. Archives of Physical Medicine and Rehabilitation, 65(10), 593–596.

University of Bristol. (n.d.). Osteoporotic vertebral fractures: A guide to diagnosis for healthcare professionals in primary care.

Watson, S. L., Weeks, B. K., Weis, L. J., Harding, A. T., Horan, S. A., & Beck, B. R. (2018). High-intensity resistance and impact training improves bone mineral density and physical function in postmenopausal women with osteopenia and osteoporosis: The LIFTMOR randomized controlled trial. Journal of Bone and Mineral Research, 33(2), 211–220. https://doi.org/10.1002/jbmr.3284

Weldon Cooper Center for Public Service. (2024, July 22) National and 50-state population projections: 2030, 2040, 2050. University of Virginia. https://www.coopercenter.org/research/national-50-state-population-projections-2030-2040-2050

World Health Organization (WHO). (2025, October 1). Ageing and health [Fact sheet]. https://www.who.int/news-room/fact-sheets/detail/ageing-and-health