[ Image caption: Scene of the crime? Cricks often strike overnight — so pillows get both credit and blame. ]

Sleeping badly and hurting are entangled: poor sleep is linked to worse neck pain, and better sleep to improvement. Anything that helps you sleep is probably part of the neck-pain picture — a bad pillow can interfere with that, and a good one won’t.

But what’s a “good pillow,” other than “comfortable”? Just getting your beauty sleep is not why people buy orthopedic pillows. They get marketed like medical devices — and pillows are especially relevant to neck cricks, because cricks usually strike overnight. Cervical contours! Memory foam! Buckwheat husk! The pitch is always that the right pillow will end your morning neck pain, your tension headaches, maybe even your snoring and your existential dread.

And there’s always something about alignment — the most popular-but-vague concept in all of musculoskeletal medicine.

Orthopedic pillow claims are classic “deepities”: what is most profound about them is untrue, and what is true is rather boring. It’s perfectly reasonable to expect a comfortable, roughly neutral head position to be less likely to trigger neck pain or headaches — but comfort can only do so much, and virtually every other idea about what makes a pillow medically useful has come from the ads, not from research (which barely exists).

Most importantly, the advertised features of orthopedic pillows are meaningless if you do not personally like how the pillow feels! The Venn diagram of “comfort” and “medically beneficial” is basically a circle: all that matters is how comfortable it is.

What a pillow can plausibly do

A pillow does not have a difficult job: fill the space between your mattress and your head so that your neck doesn’t spend the night bent in some unhappy direction. We’ve all regretted an evil pillow — one can ruin a night and aggravate a neck — and any decent pillow prevents that worst-case scenario. Which is why almost everyone is already using a pillow they like: bad pillows are mostly a self-correcting problem.

The harder question is whether you can meaningfully upgrade a pillow that already feels fine. The idea that an okay-feeling pillow can still limit or threaten your neck health is a bit of a reach — and yet that idea must be true for an “orthopedic pillow” to have any value. The sweet spot here is more of a dot than a spot: a problem just big enough to matter, but not obvious, that is actually caused by your head position at night, and that a different pillow would fix. Maybe some people live on that “dot,” but it can’t be common.

What a pillow almost certainly can’t do

A pillow is not a “treatment” for most of the things that cause neck pain and headaches — it cannot reverse osteoarthritis, unpinch an angry nerve root, or untie the knot of risk factors for chronic pain. And no pillow can “realign” anything in a lasting way: the spine doesn’t stay where you put it overnight, only extreme deviations from neutral are a problem in the first place, and it’s wishful thinking that anything short of a vice could stop you from squirming in your sleep. And nobody wants a pillow that feels like a vice! There’s quite a low upper limit on how much “support” a pillow can offer before it becomes a new kind of problem.

[ Image caption: Satire — but only slightly sillier than some real pillow marketing. ]

Almost zero science-based conclusions — with one faint exception

The only way to really know if any kind of pillow can make a difference for neck pain is to test it with good science. Too bad no one ever has. You don’t need good evidence to sell fancy pillows! Hype and hand-waving about alignment are much cheaper.

Pillow trials tend to be small, short, methodologically weak, and funded on the cheap by the pillow pushers themselves. The most-cited pillow trial in the literature — a glowing report about a water-filled pillow — was bankrolled by the pillow’s manufacturer (Lavin et al). Two modern attempts to synthesize this junk (Chun-Yiu et al and Ghosh et al) both found trivial-to-absent effects, no pillow type superior to any other, and sleep quality unmoved … and then both dressed up those nothingburgers as cautiously positive conclusions their own data doesn’t support.

I take all of this apart in properly nerdy detail in my neck pain book, if that’s your idea of a good time.

For the foreseeable future, the evidence remains technically inadequate rather than actually negative. But what we do have shows no clear benefit — despite plenty of it being produced by people who badly wanted to find one. That’s the disappointing outcome you’d expect if pillows just don’t matter much.

There is one shred of good news worth salvaging from the wreckage — as long as we’re clear that it’s really a plausibility claim wearing evidence clothes. Overnight positioning is the only mechanism a pillow actually has, so if pillows help anyone, it’s probably people whose necks are at their worst in the morning. When the pillow industry paid for flattering research, the best it could squeeze out was a faint benefit for waking pain — one measly point on a ten-point scale, and nothing else. Bias that fails to buy an impressive result is mostly more bad news … but it also gives us a rough ceiling on the best case: small and plausible.

If pillows matter at all, that’s where, and that’s about how much.

Personal trial and error is unavoidable — even if we had better evidence

Comfort isn’t a tiebreaker — it’s almost the whole game. Even if there were excellent evidence that Pillow A is the objectively best pillow for the average neck, it might still be the wrong pillow for you — and you wouldn’t stick with it if Pillow B felt better. Good evidence would only narrow the search: it would tell you which neighbourhood to shop in, but it would not save you from testing pillows yourself, over weeks, while trying not to be misled by the natural ups and downs of pain. Did your neck get worse because there’s something wrong with that pillow … or did you just sleep on it funny one time?

If there’s an obvious pain-pillow link in your life, it’s not a mystery and not a problem. If your neck hurts more in the morning than it does the rest of the day, your pillow is a reasonable suspect, and experimenting is cheap relative to most things you might try.

Start with the variable that matters most: height. Side sleepers generally need more, back sleepers less, stomach sleepers almost none — and stomach sleeping, while we’re here, is the position least kind to a neck no matter what’s under your head.

Try one change at a time. Give each pillow more than one night — first impressions are often wrong. But don’t tolerate discomfort for long! Finding a comfy one is the whole point.

And if there is not an obvious pain-pillow link … it’s probably not a problem! If your neck hurts no more in the morning than at any other time, your pillow probably isn’t the villain, and no pillow is going to be the hero. Sorry. Shopping for a special pillow will just cause a pain in the wallet to go along with the one in the neck.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

Steroids won handily, and that’s worth knowing — but I see this study as more of a bad news story for PRP than good news for steroids. Fashionable, fancy PRP couldn’t even outperform saline!

PRP is the injection of your processed blood, a concentrated mixture dominated by platelets, a blood ingredient that releases more growth factors, making it allegedly anti-inflammatory and/or “regenerative.”

Corticosteroids, of course, are powerful anti-inflammatory drugs (synthetic analogues of cortisol, a hormone) often injected into especially “hot” areas. They have a well-understood efficacy and safety profile: often helpful in the short term, though often less than you’d think, and dangerous to keep using. They can be good for helping people get through a crisis, though.

[ Image caption: Frozen shoulder. ]

Some of the nerdy details

Hsieh et al. randomly assigned ninety patients to one of the three injection protocols, with ultrasound-guided injections targeting both the glenohumeral joint (the main, ball-and-socket joint), plus a bursa (friction pad, subacromial-subdeltoid). Everyone got bog standard physio along with their injections. Follow-up ran to six months, with assessments at 1, 2, 4, and 6 months. Both patients and outcome judges were blinded (not the injecting physician).

The results were clear: corticosteroids were substantially superior to PRP on the primary outcome (Shoulder Pain and Disability Index) and most secondaries as well, with an obvious effect by one month and sustained to six months.

But the really disappointing result for PRP is that it did no better than saline on the raw numbers (the paper itself didn’t report that comparison directly, but it’s not subtle if you look at the data).

All three groups improved, which probably reflects natural history as much as treatment effect — frozen shoulder resolves on its own, and there was no untreated control.

Was this actually a fair test of PRP?

I like to see PRP lose. I am biased against PRP, and cranky about it to boot: I’m sick of it being promoted and sold to people way out of proportion to what the evidence can support.

And so I should probably not embrace this result uncritically just because it’s what I want to hear. To be intellectually honest, I must try to see it as a PRP proponent would. And indeed they would have some justification for dismissing it. Anticipating their main nitpicks then, in an abundance of fairness…

• PRP composition was never revealed: no platelet counts, no growth factor analysis, no leukocyte profiling. PRP is not a standardized drug.

• All the injectates — including the PRP — contained 4 mL of 1% lidocaine mixed in. Lidocaine is cytotoxic to platelets and cells at relevant concentrations, and mixing local anaesthetic directly into PRP before injection may have compromised its biological activity. This isn’t discussed in the paper.

• The study also recruited exclusively from the freezing stage, where active synovial inflammation is dominant and corticosteroids should be at their strongest advantage. By excluding the frozen and thawing stages — where PRP’s supposed regenerative mechanisms might be more relevant — the study may put PRP in its worst-case scenario. But PRP wasn’t a random choice of comparator: earlier trials (generally lower-quality) have boosted PRP as a frozen shoulder treatment without regard to the phase of the disease.

Dismissing the result with “poor PRP” is not a frivolous objection, even if it is also a convenient one. These are real limits. But they are not deal-breakers: the result is still a credible negative for PRP, just not nail-in-coffin material. And it’s hardly alone! This is one new discouraging data point among many.

I review the topic thoroughly in my main PRP article, now updated with this citation (as is my frozen shoulder book).

The PRP-vs.-saline equivalence is the provocative finding of this study, but comes with some flaws, none of them critical. Meanwhile, the CS advantage is robust and convincing. We can say, based on this data, that steroids are better than PRP for shoulders in the freezing stage.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

The study reports that a mere six minutes of daily home stretching for six weeks reduced fibromyalgia symptoms by a little over 18%. Not exactly “huge” if true, but it’s good, and cheap, and no major flaws leap out of the abstract, so … should I put that in my pipe and smoke it?

Not today. [YouTube, Game of Thrones scene] Based on the abstract alone, it sure looks like solid good news, but abstracts are like movie trailers: they only show you the good bits.

The honest conclusion — like an “honest trailer” for a bad movie [YouTube again] — is that a six-minute daily routine with weekly researcher contact improved self-reported fibromyalgia symptoms over six weeks compared to — wait for it — being on a waitlist. So the “result” is hardly a surprise, and the reality is a much weaker claim than “stretching works for fibromyalgia.” My bias is safe from this junk.

[ Image caption: Source unknown to me. ]

The critical flaw here is a cliché of amateur science: a failure to adequately “control” major confounders, AKA “the whole point of doing a clinical trial.” The paper title claims it is “controlled,” but it’s just not: the “usual care” waitlist design is a what-were-they-thinking choice.

This trial cannot separate the effect of stretching from the effects of expectancy, novelty, and weekly contact with a caring researcher, yada yada yada — an especially glaring problem with fibromyalgia, where patients are starved for validation. Participants weren’t blinded either, so the waitlisted group results were probably deflated by disappointment (“frustrebo” effect, performance bias), widening the artifactual gap further. All of this has happened before, like with massage for fibromyalgia.

The authors acknowledge this problem, but they seriously understate how decisively it sabotages the experiment:

“the waitlist design may have induced expectancy asymmetry and limited causal inference compared with an active comparator”

This is like saying that eating potato salad left in the sun for an afternoon may induce facilitated fecal transit. The waitlist design definitely broke the experiment.

I’ve recently reported that some people with fibromyalgia do not think stretching even feels good. This paper’s conclusions are implausibly inflated in favour of stretching, and if anything the study just adds to the pile of accidentally produced evidence that fibromyalgia benefits more from compassion, validation, and any light activity than any other known therapy.

Which is why I will continue my easy daily stretching routine — ideally with a dog.

[ Image caption: Dogs are a superb source of the validation and compassion that were probably the active ingredients in the Støve et al. experiment. A good dog will provide these while you stretch, sometimes while napping. ]

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

I also mentioned that I often lose subscribers because I eventually publish something that triggers them.

Fresh example!

I lost two paying members when I posted that “stretching doesn’t always feel good” … because I was being “negative” about stretching. 🤦🏻‍♂️

That post was about medical conditions that combine poorly with stretching. What kind of person thinks that is so offensively anti-stretching that I no longer deserve their business? More importantly, what kind of so-called healthcare professional would rage-quit my membership program because I wrote about pathologies that complicate stretching?! (While still clearly acknowledging that it does feel great to most people!)

Bonkers. Weird. Sad.

But that is what I was talking about: a pitch perfect example of why my business plan is kind of a ridiculous way to try to earn a living.

On the bright side, the average IQ of my audience went up (hat tip to Drew Baye for that framing). So good riddance and all that. Publishing content that bothers small minds is a good filter. But it can be hard to remember that healthy perspective on a hard day, though … and I don’t think never get used to just how much work the filter has to do! It’s chronically disappointing.

Thanks to the many members with thicker skins and a more sincere and durable interest in the subject matter. And indeed welcome to a few more who joined up when I shared this story on Facebook. A surprise happy ending!

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

“It just hurts too much. Usually too sore!”

“I hit the end range of my joint miles long before I feel any stretch.”

The post did mention both conditions, but only as potential factors in more exotic vulnerability to muscle stimulation — not the more obvious, simple reality for some of these patients: stretch hurts and/or it’s just a nothing burger.

So that’s tuned up now. And I re-wrote everything else under the heading “Unpleasant stretch has many other possible explanations.”

Also, this topic is now properly integrated into my main stretching article.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

“But stretching feels so good!”

And for as long as I’ve been writing about how stretching is not a pillar of fitness (since circa 2005), I have been softening the blow by acknowledging that, yes indeed, it really does feel good. Even great. That is one of the few stretching beliefs I can get behind.

But stretching does not feel good for everyone, and sometimes it can even feel bad. (And I don’t mean stretch injury, although that is also a thing.)

Stretching just feels bland and pointless for some people, like no kind of itch is being scratched. If they bother stretching at all, it must be motivated by an abstract faith in the benefits (despite evidence to the contrary), because it doesn’t feel like it’s doing anything at all, good or bad.

I have heard from some people like that over the years. Not many; they are well outnumbered by the but-it-feels-good faction. But they exist.

And it can get worse than “boring”: imagine feeling queasy and lightheaded just because you stretched! Some people even faint.

Malaise and fainting while stretching

“Stretch syncope” is real but truly rare. Some specific stretches — not just any kind of stretch — can collide with a rare physiological vulnerability to fainting, which may be a complete surprise to the individual (or something they have encountered before in other contexts).

The culprit in most documented cases is probably neck hyperextension, which can compress or stimulate the carotid sinus — a pressure-sensing spot in your neck that helps regulate blood pressure. This won’t happen to just anyone, or no one would ever do a yoga class. The carotid sinus has to be a bit twitchy for some reason — that’s the rare part. But when it is, and you stimulate it just so, it can trigger a reflex blood pressure drop. In many people, this probably just registers as a bit of lightheadedness. In some, it crosses over into nausea. It can freak people out, and I wouldn’t be surprised if it’s responsible for some panic attacks.

In the worst cases — in the unluckiest stretchers — the effect can cascade into presyncope (very woozy) or full syncope (out like a light).

(You can read a bunch about the carotid sinus in my article about the effect of massage on nerves, but it’s for members only because it’s nerdy extracurricular reading.)

This unpleasant nonsense is more common in people with orthostatic intolerance. [Wikipedia] There are many kinds of OI, but it is often associated with chronic unexplained illness and pain, especially the POTS sub-type (postural orthostatic tachycardia syndrome, characterized by a racing heart upon standing up). Although stretch syncope is rare, it’s probably less rare in the people using stretch to try to relieve their aches and pains.

This is less about “stretching can make you faint” and more about a perfect storm: a just-wrong posture combined with an exotic sensitivity, in someone whose autonomic regulation may already be a bit glitchy. It’s a “stupid human trick,” an edge case in a system that mostly works fine.

Unpleasant stretch has many other possible explanations

• Habitually holding your breath while stretching probably won’t cause serious symptoms, but it might make stretching feel less pleasant. And it’s probably common. Just ask a yoga instructor how often they need to remind people to breathe!
• The pleasantness of stretch can be undermined by simple overexertion. Sometimes stretching involves quite strongly held positions, especially in a yoga context, and it is piling more exercise onto a body that is already a bit worn out and run down, maybe quite sore or even borderline injured.
• Any physical activity can collide with the transient vague malaise of mild infections (mostly colds and flus that never develop clear symptoms). This would be a passing, occasional thing, but also probably quite common.
• There are many underdiagnosed health problems that reduce the resilience and increase the sensitivity of muscle and connective tissue or otherwise cause non-obvious vulnerability. For instance, some people are just too sore, always or often, due to common illnesses that cause chronic widespread pain, like fibromyalgia and Ehlers–Danlos syndrome. Many people with these conditions do like stretch, and even find it helpful — but not all.
• Many chronic health conditions also involve malaise and weakness, and stretch or common stretching positions and movements can be a trigger. For instance, POTS seems to be a component of many chronic health conditions, and could be set off just by the getting up and down (no carotid sinus trigger).
• The most serious pathology that can collide with a common stretch position is upper cervical instability, caused by trauma and a variety of diseases of the joints at the top of the spine, which can be subtle yet serious. The only symptom of erratic irritation of the spinal cord and brainstem might be vague episodes of malaise. This is not as rare as you’d think. More dire cases are rare, but they do exist, and — weirdly — still often go undiagnosed, simply because the trigger is not obvious.
• “Exercise intolerance” is exhaustion and malaise triggered by exercise. It comes standard with ME/CFS, and is also linked to several other chronic conditions. I suffer from this: I get bizarrely nauseous when I exercise my muscles strongly (resistance training). Although stretching is not a known trigger for exercise intolerance, it’s not implausible. Stretching is a kind of exercise, and we already know that other kinds of muscle stimulation can go badly …
• Even seemingly healthy people can feel a bit off after a massage: post-massage soreness and malaise, or PMSM, a weird phenomenon. Whatever it is, it could be triggered by any stimulation of muscle, either massage or stretch, making it seem like a kind of exercise intolerance. Muscle tissue is biologically lively and well-integrated into the whole system. For instance, it’s an “endocrine organ” — it produces hormones when stimulated. That’s mostly a good thing (anti-inflammatory), but maybe not in all physiological circumstances. If massage can trigger PMSM without any obvious illness in the picture, then stretching might too — but it seems even likelier if there is a chronic illness, even if it’s undiagnosed and mild. Personally, I would bet that there’s some kind of undetected vulnerability in every case, even if it’s just a weird genetic quirk with no other practical consequences.

There are a lot of ways to hurt and feel bad, and “everyone is different” is an understatement.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

This charmingly quotable wisdom nugget was recently shared by John Gruber, so you may well have come across it already, but it’s worth a look again while thinking about the messy and fiercely disputed “truths” of biology and healthcare:

There are three ways to make a living:

1. Lie to people who want to be lied to, and you’ll get rich.
2. Tell the truth to those who want the truth, and you’ll make a living.
3. Tell the truth to those who want to be lied to, and you’ll go broke.

This quote obviously resonates with a lot of people, but I am going to pull rank: it resonates with me more. As an independent science journalist.

I have been learning this the hard way for two decades now. PainScience.com has always been about trying to make that sweet, adequate “living” by telling the truth to those who want the truth — and it’s a ridiculous business plan, because two thirds of those people only think they want the truth. Or is it four fifths? Anyway, it’s too many, and they all loooove my work right up until I tip over one of their sacred cows, and then suddenly they cannot cancel their memberships fast enough. Sad trombone.

I know it’s possible to make a living this way because I’m still doing it after all these years, and I have the best people in the world to thank for that — my long-term members. But such exceptional people are rare — so hard to find that you can’t make a good living this way, and most cannot manage it at all.

See also: Debunking for a living. See also my origin story. See also a strong fresh example of the problem, a couple days after this post.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

None of the subjects had ever had Achilles tendinitis before. But after a year, there were 23 new cases in their study subjects, 20 in runners, 3 in the non-runners. That’s not many cases — the study’s main weakness. But it’s big compared to other trials like this, it’s an unusually good design, and it mostly just reaffirmed the obvious…

Running volume is the main risk, and exactly how you run is not so important. Many traditional biomechanical scapegoats were exonerated by this study — most notably footfall patterns — because they “do not significantly influence the risk of Achilles tendinopathy, challenging common recommendations to alter running footfall pattern.” Classic physio advice like “avoid forefoot strike,” among other usual suspects.

The risk drivers according to Jandacka et al.

Just how risky is a lot of running for your heel tendons? 67% riskier with every additional 12 kilometres per week. The injured group ran 3.5× more per week than the uninjured group. (Based on fitness-tracker data, not self-report.) To run more is to get hurt more! Duh? Too obvious? You’d think so. But never underestimate the degree to which people like to make up more complicated stories about why we get hurt, so it’s always worth emphasizing the “obvious.”

This next bit is a bit more interesting…

The injured runners were already in trouble when they entered the study! 🤯 They just didn’t know it yet. Although asymptomatic and with no history of Achilles tendinitis, their tendons were visibly flawed on MRI (the VIMATS scoring system). Healthy tendons can easily score 100%. Tendons with symptomatic Achilles tendinitis score more like 50 out of 100. The runners in this study who ended up injured? 95. Just a little compromised: a bit of thickening, a bit rough. The earliest clear signs of degeneration.

So, who gets injured? The people who had a quiet head start on being injured! Maybe they had a head start because they were also the runniest runners, but we’ll need more research to determine that. Clearly we need to be studying who develops silent tendon degeneration, before they even know anything’s happening.

Interestingly, age did not move the needle — in defiance of every aging athlete’s cynicism about getting old! The odds were against them at any age if they ran a lot. Or had that head start.

Jandacka et al did flag two running form issues that were linked to more Achilles tendinitis: weaker ankle inversion during stance, and not turning the foot out as far. Although those signals were fairly strong in this study, I wouldn’t put too much stock in them: they’re new ideas in the field, might well never be reproduced by another study. They also probably aren’t causes that can be tinkered with. We always want our risks to be “modifiable,” avoidable. Of course, reality doesn’t always cooperate. But an unmodifiable risk is news you cannot use, so even if it’s true it won’t change how you run.

But I suspect it’s not true. I bet that this signal would disappear in a do-over study.

The prospective data dream

Jandacka et al. have given us the kind of evidence about risk factors for injury/illness we’re always waiting for, and can never really get enough of: prospective data from following people over time, which can actually shed some light on causal relationships by showing us what came first. This is far superior to the “just another correlation” conclusions that we get from far cheaper, easier observational studies that only give us a snapshot of a group of people at one time.

“Prospective” is potent. Basically all previous data on this topic was retrospective or cross-sectional, or prospective-but-tiny, making this experiment fundamentally better despite the fairly small group of limping runners they had to analyze after a year.

It’s kind of embarrassing that we have so little science like this, but that’s the hand we’ve been dealt so far. The authors argue their exoneration of footfall pattern is more trustworthy because the retrospective studies likely detected injury-driven gait adaptations, the actual opposite of injurious gait glitches — retrospective studies were seeing the injury's effect on gait, not its cause.

We still need a ten-year study of ten thousand runners. Don’t hold your breath.

•

I’ve updated my guide to Achilles tendinitis with a citation to Jandacka et al.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

Consider this the next the time you hear (or say) “it worked for me.” People once talked that way about their fashionable lobotomies!

Nothing more strikingly betrays the credulity of people than medicine. Quackery is a thing universal and universally successful. In this case, it becomes literally true that no imposition is too great for the credulity of men.

Henry David Thoreau

Here are some colourful historical and current examples of dangerous-but-popular treatments, just updated with that irresistibly apt Thoreau quote, and couple new examples: black salve and vagina steaming.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

Alternative medicine famously never gives up on anything, no matter how grim the evidence. Medicine usually does away with obsolete ideas, but it can be painfully slow, even when the evidence is strong. Doctors may carry on prescribing clearly bad medicine for many years after medical guidelines start discouraging it.

Just how long? About a decade on average, and sometimes much longer. Yikes. 😬

Epidural steroid injection (ESI) is a prime example: it has already been ten years since NICE first discouraged it in 2016, and the writing was on the wall well before that, but about 9 million of these are still being done every year in the US, costing billions. And that’s just the sticker price in one country — not counting all the indirect harm done.

“There is no better de-implementation candidate in pain management,” write Ballengee et al. in a new paper for the Journal of Pain (not a “study” but a “focus article” — expert perspective and patient advocacy). [1]

“De-implementation” is a polite way of saying that ESIs need to be booted out of the medical establishment, because they just don’t work well enough. It’s an appealing, “common sense” treatment in theory, but the benefits are barely there in practice — such poor value that 9 million annual prescriptions is alarmingly excessive. They need to mostly stop.

But bad medical practices don’t de-implement themselves! Ballengee et al. argue that they must be shown the door: “successful transformation requires not only having scientific evidence but also developing means to proactively and systematically phase out entrenched practices.”

In this big new post, I explain and explore ESIs for members only. You can get the whole thing by becoming a member — even briefly, though of course I encourage you to stay awhile. The full post covers:

• Epidural steroid injection overview: what exactly is it?
• Ballengee et al. don’t think ESI works very well — and for good reason
• Slightly less bad news: short-term relief is real, and the odds are better with sciatica
• Why don’t spinal steroid injections work better than they do? Five possible reasons ESI fails
• Five harms of spinal steroid injections (beyond just being a waste)
• Why do steroid injections persist despite the very discouraging evidence? It’s not just the money
• How to solve the problem, maybe, someday, in just 378 easy steps

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

Nothing that lived and breathed was truly objective — even in a vacuum, even if all that possessed the brain was a self-immolating desire for the truth.

Area X Trilogy, Jeff VanderMeer

I was going to just drop that quote into an obvious place — my article “Objectivity is Overrated” — and get on with my day, five minutes tops. But my creative attention was hijacked! Two hours later I’d substantially rewritten and expanded that little article. So much for whatever else I was supposed to be doing this morning.

I have a thousand more pressing publishing priorities, but sometimes I still do my job like a cat chasing a laser pointer.

The Area X novels (AKA the Southern Reach Series) are fascinating, but an acquired taste, surreal and unsettling. Worth a shot if you love wilderness, though.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

But it’s especially odd the way pain is routinely ranked without any curiosity about duration — which is arguably more important than the intensity at any given moment. This is a common complaint about pain care:

“I don’t think I have ever been asked in 20 years about whether my pain is constant, or how long severe pain episodes last for. Only ever asked to score it ‘now’, or a ‘best’ and ‘worst’ in a 2 week period.”

Pain isn’t actually “mild” if it never ends. Persistence can turn even a small pain into a nightmare.

Everyone knows that a toe stub is surprisingly intense, but we also know that the misery is brief. If you had a pain just like that in your noggin, you’d think your head was exploding (an aneurysm). Most headache pain is way less intense than a toe stub, but we also know it’s probably going to last the rest of the day, and most people would prefer the toe stub. Get it over with!

So it’s not that we don’t get this. We talk about it imprecisely and indirectly all the time.

The concept exists in the science. It’s an old tool in analgesic trial methodology — the summed pain intensity difference, or SPID — precisely because trialists realized a snapshot pain score might not capture treatment benefit over time. Used … but only in certain kinds of trials, and debated even there: “SPID does not appear to add anything” (Tfelt-Hansen et al). Just last year Chukka et al set out to “validate SPID’s reliability and clinical utility in orthopaedic surgical cohorts” (and they say they did). And then there are some “competitors” of various types … all barely used. [1]

So the literature acknowledges the problem and researchers have proposed solutions, but pain-over-time is still mostly missing from most pain research, clinical assessment, and patient communication. [2] It’s a good example of what I mean about pain medicine being surprisingly primitive and improvisational.

And it’s weird because describing pain well so clearly involves both intensity and time (among other things, but let’s stay focussed); you truly can’t express the experience of chronic pain when the time thing has been downgraded to a footnote!

For an awful lot of chronic pain patients, being asked mainly about pain intensity is a little awkward — because the reality is that a lot of chronic pain is actually not very intense, and it doesn’t sound so “impressive” when the real story is buried in an addendum: but it never lets up.

[ Image caption: This chart shows pain intensity over time for three types of pain: toe stub (intense but brief), tension headache (moderate for a day), and chronic low back pain (a little milder, but much longer). Shaded areas represent total pain burden (intensity × time). Note that the x-axis uses a compressed time scale to fit seconds through days in one view, a design choice with pros and cons. Data visualization is hard! ]

Even when the burden over time is substantial, many people won't seek help if the intensity never gets beyond a certain point. Beware of ignoring persistent pain just because it’s mild in any give minute of your day. You might avoid it because you aren't taking it seriously yourself; or you might avoid it because you fear that you won't be taken seriously by healthcare professionals.

The persistence of pain should never be an “asterisk” on your report. As a pain patient, you should not feel like you need a qualifying “but” after giving your pain a number. If the pain intensity alone doesn’t tell the story — and it often doesn’t — then emphasize what does: intensity times duration. When talking to professionals, try to explain how “much” pain there is, and lead with the “area under the curve.”[Wikipedia] Tell them how bothersome it is overall.

If you’re a professional, please make it easy for people: ask about the persistence and pattern of the pain, as well as its intensity. Remember your mnemonics (SOCRATES and OPQRST), and more: take T-for-time seriously! (See footnote #2.)

This post is part of a developing series on pain rating. Previously:

• What are the worst kinds of chronic pain?
• There will never be a pain-o-meter — and maybe that’s a good thing

Notes

1. In 2014, Salamon et al developed a score to "capture pain intensity along with frequency and duration" — the descriptively named pain frequency-severity-duration scale (PFSD).

   In 2016, the National Institute on Drug Abuse developed the Pain Frequency, Intensity, and Burden Scale (P-FIBS).

   In 2020, anaesthesiologists Lang-Illievich et al tackled another method, an "area under the curve" analysis of the graph of pain over time.

   A softer alternative: in 2025, Edwards et al validated single-item "pain bothersomeness" scales — simply asking patients how much their pain bothers them — and showed they outperform intensity-only measures as predictors of functioning, depression, and social outcomes. Bothersomeness doesn't measure time explicitly; instead it asks patients to do the integration themselves. A reasonable workaround, though it obscures rather than illuminates the duration component specifically.

   None of these have seen much (any?) use.

2. Sharp clinicians may be wondering why I would say such a thing when there are well-established mnemonic frameworks that explicitly cover a bunch of factors in assessing pain, including T for time:

   SOCRATES: Site, Onset, Character, Radiation, Associations, Time, Exacerbating/relieving factors, Severity. The T covers timing, duration, pattern, and whether it’s constant or intermittent.

   OPQRST: Onset, Provocation/Palliation, Quality, Region/Radiation, Severity, Time. Again, time is explicitly there.

   But there is a huge gap between the ideal and reality, as there so often is in medicine. Many aspirational clinical tools like this are routinely ignored, oversimplified, and sometimes even outright abused in practice. In this case, it’s mostly just neglect. Either clinicians don’t use SOCRATES or OPQRST at all, or it’s token, or muddled, and doesn’t translate into actually weighing time against intensity. A clinician can dutifully note “constant pain for 3 years” and still anchor their severity judgment entirely on an intensity score. To some extent these frameworks may even de-emphasize the time element by reducing it to one checkbox among many. All those factors matter — but intensity×duration burden may be in a class by itself.

   Consider this reader report, demonstrating what I mean:

   “I remember going to a doctor when I had had a headache for months, and they asked me to rate it 0-10 and I said 2. They completely lost interest.” That shouldn’t happen! But it does. And that’s my point.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

The Mayo Clinic’s Pain Rehabilitation Center (PRC) program is a well-known, decades-old interdisciplinary pain rehab program that officially aims to help people “cope” with their serious chronic pain, to function and live better in spite of it, primarily with cognitive behavioural therapy (CBT), acceptance and commitment therapy (ACT), and physical therapy (largely for fitness).

In practice, their methods may sometimes reach past that defensible “live better” goal to the much more dubious premise that pain is actually driven by poor coping, and therefore can be cured by getting people to think and act differently — in other words, treating patients like pain is a psychosomatic problem, caused or perpetuated by their hang-ups and “attitude” about their pain. Consider this review by a young participant, Reddit user WhickenBicken, who characterized his or her experience at the PRC as “abuse”:

I went to the Mayo Clinic PRC program for teens around 4 months ago. It was the worst medical trauma that I’ve ever experienced. First off, the program is $50,000. Way over priced for the abuse that I, and the other teens experienced. It’s a three week program where you attend lectures and physical therapy. When I was looking into going there, I was told “we’ve had teens who arrive in wheelchairs, and walk out on their own.” That sounds pretty impressive.

Fifty grand? Even fifty bucks would be too much to pay for “abuse.” (The $50,000 figure is roughly corroborated by other PRC patients on Mayo’s own forums, who report out-of-pocket costs in that range, but those with insurance paid far less.) It’s a troubling story no matter the price. If true, it was textbook gaslighting:

I was told to stop asking questions, or leave. Now, what kind of questions got me kicked out? Probably something horribly disrespectful right? Nope. There is a term they use called “pain behaviors” which they vaguely described as unhelpful behaviors that one does because of their chronic symptoms. Seems pretty straight forward. And it was until they said that if we are caught doing a pain behavior, our parents are supposed to give us a consequence.

I cannot know if this characterization of the PRC is fair. It might be a gross distortion; WhickenBicken may not be a reliable source, and the Mayo Clinic might well have a reasonable explanation, and there are positive reviews out there too. That said, accounts are mixed even on Mayo’s own patient forums, and the program’s approach has attracted criticism: Dr. Chris Centeno argues that it ignores the mechanical and chemical complexity of pain and reduces everything to a nerve-sensitization story. Centeno runs a competing regenerative medicine practice, so weigh that accordingly, but the critique is recognizable — I have made that very point myself. From his blog post, “Chronic Pain Programs Rise from the Ashes: This Was Silly in the ’90s and It’s Still Silly Today”:

These pricey three-week outpatient affairs touted that all these patients needed was to learn physical and mental coping skills and they would be fine. The problem was that the basic thesis of the program — that patients were in pain simply because they believed themselves to be and learned bad habits because of those beliefs — was not in keeping with the science on pain as it evolved.

Stories about this kind of care aren’t implausible or even unfamiliar in the field of chronic pain management. Both patients and professionals have many reasons to suspect the power of the mind over pain — reasons I have documented extensively. Fierce debates about this are common. The goals of biopsychosocial care are laudable in theory but often missed in practice, and what we see instead is a simplistic, obnoxious overemphasis on the importance of “psychosocial factors.”

Believing that pain is actually caused by social and psychological forces is the beginning of a slippery slope to making patients at least partially responsible for it, and it’s all too easy to actually blaming people for their pain if they don’t get with the program — figuratively or literally (in WhickenBicken’s experience). This works exactly like the more familiar problem of “toxic positivity” (and indeed it is just another form of it): you can’t credit people with the power to make things better with a good attitude without also implying that a bad attitude caused their problems in the first place.

Whether this bad review of the Mayo Clinic’s PRC is fair or not, it resonates with much that I do know to be true, and what most chronic pain patients have seen all too much of: the mind gets the blame. (Sorry, not the mind per se, but nerves that are sensitized by anxiety, fear, and pessimism. There, now it’s a “neurological” story rather than psychological!)

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

“Sure, it’s not going to work any miracles! But … every little bit helps, amiright?”

The first problem (explored in 2022) is that the “little bit” part may just be wrong: it’s not a little bit of help, it’s zero help. That trial didn’t actually show a “promising” small benefit, but just an illusion or exaggeration — which is practically standard for supposedly promising trials.

The second problem (and today’s focus) is that even a genuine small benefit is not necessarily a win. Some examples from this category:

• A little bit of benefit from hacking your posture.
• A tiny window of opportunity opened by vibration therapy.
• A modest, fleeting bit of relief from strapping on a TENS machine from the drugstore.

Not every little bit helps. An awful lot of things like this are technically helpful, but poor value. They just aren’t worth the money, the time, the energy. It’s “penny wise and pound foolish,” something we’re supposedly better at avoiding in other areas of life. And yet plenty of people will still burn $10 of gas to save $3! Not everyone can math.

And healthcare decisions are trickier than gas prices. Barely-effective treatments and over-optimizing for “wellness” often have a significant but subtle opportunity cost [Wikipedia] — that is, the hidden expense of not doing other things. The sneaky price of the neglected options.

That time you spent stretching? Walking would have been a better investment. Or making a healthier dinner. Something with a much clearer return on investment, probably something basic. Or something that lifts your mood, like friends or making music.

But what if the up-front cost is just five minutes? What if it’s truly cheap and easy?

Surely those little bits help! Maybe, yes. If the opportunity cost is small, and the benefit is real, that does change the calculus. A trivial effort for even a modest benefit is decent value, or at least not a terrible one.

Those cheap-but-good unicorns are rare, though! Far from “every” little bit.

And even when a single-serving really is easy, hardly anyone ever has just one five-minute intervention. They repeat and accumulate: a posture habit here, a stretch there, a gadget, a supplement, a breathing exercise. Suddenly you have an hour-long wellness routine that could have been an earlier bedtime. Individually defensible little bits can easily add up to an indefensible whole lot of bits.

The important question isn’t just whether something works at least a little bit — that's not enough on its own. What counts is whether it’s the best use of your resources. And a lot of people in pain are running on fumes. Opportunity cost is the underestimated non-physical harm of many of the tame snake oils — not just the poor financial investment, but the poor investment of time and attention that people actually need for other things in life.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

If nerdy fun is a non-starter for you, here’s the zero-nuance bottom line: surprise surprise, cannabis does not work well for neuropathic pain, if at all. There’s about seven asterisks on that, but that’s the reasonable broad strokes conclusion. Discouraging evidence never stops anyone from shouting “it works for me,” of course. Mostly not about anything, but cannabis is in another evidence-denial league. Maybe it works for you and maybe it doesn’t, but I can guarantee — based on this evidence — that it doesn’t work very well for very many.

(And no, there probably are no juicy outliers missed by the studies. A good benefit for a lucky few mostly cannot “hide” in weak average results. If it were good, it would boost that average.)

[ Image caption: Perhaps the most interesting and controversial plant in the world. Photo by Jose Luis Sanchez Pereyra on Unsplash ]

REVIEW #1: Ateş et al was a meta-analysis from the Cochrane Collaboration — which has a reputation for high standards and getting this kind of thing at least less wrong. They delivered a low to very low certainty conclusion: barely a conclusion at all! Really just the classic “garbage in, garbage out” pattern for so many meta-analyses of pain treatments. What the data (barely) shows is a minor bump in some softer secondary outcomes (and only for a THC/CBD combo, not for either THC or CBD on their own), which the reviewers themselves flagged as not clinically relevant. Plus an entirely predictable increase in adverse events. This is how they frame their bottom line:

“There is no clear evidence for an effect of [cannabis] medicines on pain relief of 50% or greater …”

REVIEW #2: Mitchell et al is another meta-analysis of mostly the same studies, with another barely-positive result with a much more positive-sounding interpretation:

Cannabis-based therapies offer modest analgesic benefit and should be reserved as an adjunctive option for select patients with refractory symptoms, guided by a shared decision-making process that considers individual risks and preferences.

But “modest analgesic benefit” isn’t justified by their data … and “should be reserved” is just editorializing loosely based on that unjustified conclusion. We can recommend practically anything that isn’t blatantly a poison as “an adjunctive option” for patients with tough cases; “it’s worth a shot” doesn’t even really require good evidence of anything other than a lack of harm.

Of course, there is evidence of harm in this case. From both studies. THC has never been harmless, and opinions to the contrary have never really been anything but propaganda. (See my full cannabis article for more information.)

So were Mitchell et al. just making lemonade out of lemons?

Not entirely. There is a legit case for their more positive conclusion, even if it’s still unimpressive. Standard mean differences (SMDs) are arguably the more meaningful measurement than the Ateş headline outcome that focused on a threshold (not counting any pain relief that wasn’t at least 50%); they focused on an interpretation that was more sensitive to small effects.

But sensitivity to small effects is only valuable if small effects are big enough, and SMD −0.30 in pain research mostly isn’t.

And what’s more damning is that the “modest analgesic benefit” disappears with another analysis (change-from-baseline). Mitchell’s headline conclusion leans on one of two pre-registered analyses, the one that produced a positive result, with no reason given for preferring it — so we’re left to assume the obvious, namely that the authors just liked one answer better than the other. When two legit interpretations disagree, “modest analgesic benefit” is not a fair summary unless it also acknowledges “but it depends on how you squint.”

Meanwhile, the harms were the same no matter how they crunched the numbers. The downsides of cannabis are confirmed more robustly in both reviews. So that’s not a great look.

This is what we get from the mundane optimism bias that pervades pain research. We are constantly hearing the best news that can possibly be justified by the data, or tortured out of it, and even that is unimpressive. The cynical salamander’s rule of thumb: supposedly “promising” research almost never actually is. And in this case we’ve got a second review for the reality check, instead of just my skepticism.

Read more about cannabis for pain.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

[ Image caption: “Thinking at Hell’s gate”, by innoxiuss, CC BY 2.0 ]

• touch fire 👉🏻 get burned
• throw rock 👉🏻 break window
• eat too much 👉🏻 get sick

There are countless simple correlations like this that we master effortlessly before we can even tie our shoes. Infants can actually see simple causality in billiard-ball animations! (Probably.) We see B follow A and we just kinda get it that A caused B.

Unfortunately, we also get it wrong a lot. Like when we believe that a treatment worked because our pain seemed to ease right after it. Maybe it did, and maybe it didn’t.

We especially fumble causality detection in health and medicine. Like when we are trying to judge the effect of a treatment for pain. Or interpret the meaning of gua sha bruises. Or grok the correlation between stretching and reduced mortality. Or decide if a tendon tore because of a drug we’re taking. Or whether noisy joints are doomed to fail from arthritis. To name a few thorny examples from just the last couple months on this blog.

This problem underlies practically everything on PainScience.com. Posts like this dig down to the bedrock of how we figure out what actually helps with pain and injury. Take us to Nerd Factor 9, Mr. Sulu!

Causality inference is a double-edged sword

We are bad at understanding complex causality for the same reason we are good at detecting simple causality — we’re “sensitive” to it. So we get causality right constantly when the variables are simple and readily observable. But we rarely get it right in health care, or any other complex endeavour, where there are many variables and many are subjective or otherwise murky. What’s really going on in a causal relationship almost always turns out to be different and waaaaay more complicated than we thought.

Our failures in this department are legion and disastrous. By far the most important thing anyone needs to understand about the relationship between correlation and causation is that A did not necessarily cause B just because B followed A, and making this mistake is one of the Greatest Hits of human thinking glitches.

That should be emphasized every time correlation is discussed, because, as Barker Bausell put it (Snake Oil Science), we have a problem with “confusion between correlation and cause on an industrial scale.” Many amusing examples of spurious correlations have been mined from data. This problem has been trumpeted ad nauseam by so many smart people for so long that it seems like an unassailable edifice.

And yet …

Actually, correlation kinda does imply causation

The famous rule — “correlation does not imply causation” — is in fact a misleading oversimplification. At the very least it’s missing a word, and it should be “correlation does not necessarily imply causation.” (Although “equal” sometimes replaces “imply,” the more common usage is definitely imply.)

Or maybe we should just rephrase it entirely. Edward Tufte, an American statistician who made the same point quite a while ago, suggested that a good informal re-wording would be:

“Correlation is not causation but it sure is a hint.”

Edward Tufte

Because correlation actually does “imply” causation, and many (if not most) events that occur in sequence that appear to be causally related are in fact causally related. Their correlation is not a coincidence. Clapping makes noise, braking stops cars, coffee makes you poop.

Our brains are tuned to detect those relationships, and that mental superpower served us well as we grew up as a species. The problem is that we’re so good at it that we overdo it. Our brains are reckless with it. We see causality everywhere, even when it isn’t there — “illusory correlation” (Chapman, 1967), yet another form of apophenia (or “patternicity,” Shermer, 2008), “the tendency to perceive meaningful connections between unrelated things.” We aggressively over-detect and over-interpret all kinds of things: faces, creepy crawlies, kinship cues, symmetry, emotionality in voices, heights and edges! And many, many more. Brains are stuffed with these biases.

So we think we understand causality in many situations where it is actually much too difficult to know. And very few situations are as bewildering and counter-intuitive as physiology and medicine.

The difference between general and specific causality inference

One of the things that trips up our causality inference superpower the most is the tricky difference between inference of causality and the attribution of mechanism. General versus specific causes, basically. We can and routinely do correctly detect causes when correlation gives us a strong enough hint, but we routinely screw up exactly what caused what.

In other words, we over-detect correlations and we over-interpret the ones we do find.

For instance, most people will assume that when a very stubborn old pain goes away during a one-hour acupuncture session that the experience must have caused the relief, because the relief followed the experience. And that assumption is probably correct. The appearance of relief probably isn’t a coincidence, probably not just regression to the mean (too quick). Something about the experience was probably helpful in some sense.

But what and how exactly? Most people will then (carelessly or self-servingly) move on to the next “logical” assumption: that the treatment caused the relief because acupuncture works as advertised. Alas, it almost certainly does not, despite seemingly credible evidence of efficacy.

We can be right about the causality in a wide view, but still be hopelessly wrong about what specifically caused what. Most people will simply ignore the possibility that the true mechanism of relief was not the efficacy of acupuncture, but the power of a caring professional promising aid and performing fascinating rituals that reek of implied potency. And the power of “surely no one would do this if it didn’t work!” These factors are wildly underestimated by most acupuncture patients. And acupuncturists.

The hint is real enough; what it’s hinting at is often not what you think.

This blog post is a slightly adjusted excerpt from one of my weirder full articles, The Power of Barking: Correlation, causation, and how we decide what treatments work. Among other things, it goes on to answer the question, If a treatment works, who cares if it’s not for the reason we think? Which maybe the most common objection to skepticism about any kind of medicine people want to believe in — and one of the hardest to respond to.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

Exercise for depression and anxiety just got a resounding endorsement from an unusually large new meta-meta-analysis — yes, that’s two metas, meta². I don’t normally write about pure psychology, but this study is truly huge, big enough to swerve out of my lane for.

It’s also informative to compare those exercise results from Munro et al. to the SSI results reported by Ziadni et al., which were distinctly underwhelming — and, by the way, about the same as what people get from bog-standard cognitive behavioural therapy (CBT). Small but significant benefits, the authors said. Standardized mean differences in the 0.25–0.37 range. Fine. Technically real, as reported. Not exactly a revolution. Not even if you believe them.

And then along comes Munro et al., covering 81 meta-analyses of over 1,000 component studies, with nearly 80K participants, all examining exercise for depression and anxiety. The effect sizes? SMD was 0.61 for depression and 0.47 for anxiety.

Quite a bit more than for SSI. And that is what I am always hoping to see in an effect size. Not necessarily huge, but … respectable.

For reference, 0.2 is considered “small,” 0.5 is “medium,” and 0.8 is “large.”

This is a casual apples-to-pears comparison I’m making between these studies: Munro was about exercise for people without diseases, while SSI had the handicap of focusing on people in chronic pain, who often struggle with exercise. A direct comparison might narrow the gap. Or … not? Because exercise for anxiety can work just as well for people with pain: SMD .63! Exercise for depression with pain is a different story: it didn’t work nearly so well, just .21 (much like SSI). See Amiri.

Exercise can truly affect psychological states, regardless of whether there is also pain involved (high confidence) … but SSI barely does so in people with pain (and quite possibility not even that, because the data is so much weaker and fishier). If your mind and mood are relevant to your pain at all, you should certainly be taking exercise more seriously than SSI/CBT. And that’s without even considering the other general health benefits of exercise.

I’ve collected a bunch of other highlights from Munro et al. for the PainSci bibliography.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

So efficient!

That’s the enticing goal of Empowered Relief, “a two-hour intervention that provides individuals with essential pain relief skills.” It’s a branded, pain-focussed version of the generic concept of a “single-session intervention,” (SSI) an “emerging and dynamic” approach to mental health. [1] Empowered Relief was developed by Stanford pain psychologist Beth Darnall.

A new scientific review by Ziadni et al. — with Darnall as senior author — attempts to make some evidence lemons into lemonade. [2] The results are framed as good news, but I think the data is a reality check if you read the numbers instead of the narrative. These methods might be helpful, but the trials just aren’t persuasive for any self-respecting skeptic. [3]

So what actually happens in that one mighty session?

Empowered Relief is a mind-over-pain presentation delivered by a psychologist to a small group, a crash course in the highlights of CBT and pain neuroscience education, plus a few other ingredients and seasonings.

PNE is based on the claim that pain can be fought by learning how it supposedly works: namely that it’s a complex, brain-generated experience tuned by thoughts and emotions, especially the perception of threat. The hope here is that knowledge and confidence are actually analgesic. This is a controversial idea!

The active CBT ingredient is learning to identify distressing thoughts about pain (“I’m doomed!”), and then reframe them with the classic CBT move: swap in a less dire one (“think of it as an adventure!”). [4]

There’s also some guided relaxation, mindfulness training, and participants take away a personalized self-soothing action plan and a 20-minute relaxation audio file featuring binaural tones. (And, yes, the binaural tones are a red flag. More on this below.)

Teaching coping skills to people in pain is not a bad idea, but it is easily abused. And this is a whole casserole of trendy mind-over-pain ideas, and we need really good evidence to establish that it can achieve even that goal, let alone easing pain itself, let alone quickly. So … do we have that evidence? Place your bets!

[ Image caption: I’ll say this much for the idea of “single session” therapy: one and done is certainly cheaper and more accessible than many sessions of more traditional talk therapy! But neither is a good value if the job doesn’t actually get “done.” Comic by Loren Fishman, HumoresqueCartoons.com ]

The meta-analysis

Ziadni et al identified 17 RCTs of psychological single-session interventions for chronic pain, and reported improved pain intensity with a standardized mean difference (SMD) of 0.17, less pain interference (0.27), less pain catastrophizing (0.37), and less depression/anxiety (0.25/0.29).

These are small effect sizes. And size matters.

An SMD of 0.17 for pain intensity is borderline negligible, the kind of number that would vanish with a stiff breeze (or a mildly skeptical sensitivity analysis). Even the largest effect (0.37 for catastrophizing) is modest by any standard. [5] (Want to see an effect size for psychological outcomes that actually does impress? See my next post.)

The review reports “good” internal validity for 76% of included trials — which seems like trying to find something nice to say about studies with deeper, more glaring problems: expectancy bias from unblinded participants, self-reported outcomes, and the question of what a “control condition” actually means across 17 different studies.

Full disclosure: I don’t have the full text yet, so I can’t assess how heterogeneity and publication bias were handled, and those are not small gaps. Until I can, the headline numbers deserve a raised eyebrow — not a standing ovation. I do have some of the other papers to look at though …

The flagship study: Empowered Relief vs. CBT

The crown jewel of the SSI evidence is about Darnall’s own Empowered Relief program, as reported in two papers derived from the same trial: the original 3-month results [6] and a 6-month follow-up. [7]

The trial randomized 263 adults with chronic low back pain to one of three arms: a 2-hour Empowered Relief class, a 2-hour health education class (control), or 16 hours of group CBT over 8 sessions. The primary outcome was pain catastrophizing at 3 months, for what little that’s worth. The statistical goal was noninferiority — showing that Empowered Relief was “not meaningfully worse” than standard CBT.

It met that bar, for whatever it’s worth to be no worse than CBT.

Noninferiority trials are easier to “win,” because you’re not trying to show your treatment works better than nothing — just that it isn’t worse than something else. Combine that with a small sample, unblinded participants reporting subjective outcomes, and a primary outcome (catastrophizing) that doubled as an inclusion criterion (PCS ≥20 to enroll) — and you have a study that was much likelier to find what it found.

The 6-month follow-up reports that Empowered Relief actually surpassed CBT on pain catastrophizing, pain bothersomeness, and anxiety — a two-hour class outperforming sixteen hours of structured therapy. If that doesn’t trigger your “too good to be true” reflex, you should probably recalibrate.

That result was also accompanied by a delayed disclosure of a scoring error on the physical function measure, which flipped that result from “inferior” to “noninferior” when it was discovered and corrected. These things happen, but the direction of the correction was awfully good news for SSI.

There are a lot of problems with these papers — too many to be credible on their own. They need support from the new meta-analysis … but that data just damns with faint praise, even if we accept its conclusions.

The elephant: conflicts of interest

Darnall created Empowered Relief. Stanford receives revenue from its certification training. Darnall is the principal investigator on multiple grants studying it. She is also senior author on the meta-analysis that synthesizes the broader evidence.

All of this is disclosed, but it means the same research group is building, testing, promoting, and now systematically reviewing its own product. [8] As I have often written, COIs are not necessarily a deal-breaker … but you want to know about them. They’re not a big deal when the research is otherwise solid. When it’s not, and the only conclusions we have are from the people who stand to benefit from them the most … well red flags don’t get any bigger or redder.

And if there was any remaining doubt that the COI is significant here, just browse the Empowered Relief website for a minute, or sample any of Darnall’s popular writing about SSIs. The confidence of the claims consistently outpaces the evidence behind them, in ways that are hard to miss. For instance, she claims that the binaural tones in the Empowered Relief relaxation recording “stimulate each hemisphere of the brain in rapid alternating succession” and “enhance the listener’s relaxation response” — pretty sketchy stuff. [9]

[ Image caption: A single session of CBT-ish stuff being “no worse than CBT” is like being “no worse than a placebo.” The good reputation of CBT for pain is mystifying, because the evidence for it is extremely weak sauce. ]

The real takeaway: no worse than something that isn’t very good

Single-session interventions almost certainly do something. A well-designed two-hour class that teaches people about pain neuroscience and coping skills probably isn’t useless. Of course we should try to help people with pain suffer less and live better.

But this evidence for ER is based on the assumption that CBT is some kind of impressive gold standard to match — and that is highly debatable. The full-dose CBT that Empowered Relief claims to rival is not exactly saving the world from chronic pain! Conventional CBT has distinctly underwhelming benefits for pain patients — so underwhelming that they might even be an illusion.

Claiming equivalence with that is not much of a flex. “Non-useless” is an awfully low bar for Empowered Relief to clear.

The meta-analytic effect sizes show only small, fragile benefits associated with pain rather than for pain itself, and all based on experiments with major methodological limitations and off-the-charts risk of bias. They tell us less about the power of a couple hours of mind-over-pain education, and more about the poverty of the standard of comparison.

Notes

1. Here’s how Mental Health Academy defines general SSI:

   “Single session therapy is a structured therapeutic encounter designed to achieve measurable outcomes within one session. While follow-up sessions may be offered if necessary, the guiding philosophy is to treat each session as potentially complete in itself. SST focuses on the client’s presenting issues, offering immediate solutions or strategies rather than engaging in prolonged exploration. Central to SST is the belief that meaningful change can occur within a short time frame. The session is highly collaborative, goal-oriented, and underpinned by evidence-based techniques drawn from broader therapeutic modalities such as cognitive-behavioural therapy (CBT), solution-focused brief therapy (SFBT), and motivational interviewing.”

2. Ziadni MS, Dildine TC, Edwards KA, et al. Single Session Behavioral Interventions for Chronic Pain A Systematic Review and Meta Analysis of Randomized Controlled Trials. J Pain. 2026 Mar:106220. PubMed 41833852 ❐

3. Are trials ever persuasive to me? Of course! Just not often, not even across all of medicine, and even less in physical and pain medicine particularly, which remains relatively primitive and is still shockingly dominated by marketing and branding and where gurus fill knowledge gaps with speculation. What does it does take to convince me? The short version:

   • Three good quality trials with clearly positive results that don’t damn with faint praise.
   • Including at least one replication from relatively unbiased researchers.
   • And double that if the hypothesis is implausible or self-serving, or if the intervention is expensive or risky.

   The long version: Most Pain Treatments Damned With Faint Praise: Most controversial and alternative therapies are fighting over scraps of “positive” scientific evidence that damn them with the faint praise of small effect sizes that cannot impress.

4. Officially CBT is only ever supposed to be trying to help people cope with pain, to reduce pain-related suffering, to be more functional in spite of pain … but in practice many patients and professionals believe that there are feedback effects on pain itself. I discuss this in much more detail in my guide to psychotherapy for pain.
5. And catastrophizing is a psychological surrogate measure of dubious value anyway — a proxy for what patients actually care about, not a direct measure of it. Catastrophizing scores may correlate loosely with pain-related suffering, but they don’t answer the questions patients most want answered: Does it still hurt? Can I still work and play? Treating a psychological construct as the primary outcome of a pain trial is also a rather flattering choice, because a psychological construct is the easiest kind of outcome to move with a psychological intervention.
6. Darnall BD, Roy A, Chen AL, et al. Comparison of a Single-Session Pain Management Skills Intervention With a Single-Session Health Education Intervention and 8 Sessions of Cognitive Behavioral Therapy in Adults With Chronic Low Back Pain: A Randomized Clinical Trial. JAMA Netw Open. 2021 Aug;4(8):e2113401. PubMed 34398206 ❐ PainSci Bibliography 49214 ❐
7. Darnall BD, Burns JW, Hong J, et al. Empowered Relief, cognitive behavioral therapy, and health education for people with chronic pain: a comparison of outcomes at 6-month Follow-up for a randomized controlled trial. Pain Rep. 2024 Jan;9(1):e1116. PubMed 38288134 ❐ PainSci Bibliography 49209 ❐
8. To be fair, this isn’t weird for modern academic research: it’s more or less business-as-usual these days to try to study stuff that can be sold, licensed, or otherwise monetized. Universities actively encourage it, under labels like “translational research” and “technology transfer.” But normalizing a COI isn’t the same thing as neutralizing it. If anything, the ubiquity of the practice just means we need to be more alert and cautious.
9. The binaural beats literature consists mostly of small, poorly blinded trials. Multiple systematic reviews have found only small, fragile effects on anxiety and cognition — not clearly distinguishable from placebo or from simply listening to any pleasant ambient sound. The “hemisphere stimulation” framing is an oversimplification with no strong neuroimaging support. This goes beyond just being promotional — those are specific, unsourced neuroscience claims about a component of her product.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

Steven Novella, in a recent post for ScienceBasedMedicine.org, made it clear that “the data is in, exercise is good for you.” He particularly emphasized that it offers great bang-for-buck, and even small improvements in fitness are disproportionately beneficial. So far, so good, and amen to all that. But then later in the post he adds this seemingly evidence-based bonus prize:

Stretching is often neglected, but has an independent correlation with longer lives and greater functionality. Just doing a few minutes of basic stretching can be helpful.

That was probably … overstated. Ironically, I’ve never heard anyone explain the foibles of correlations better than Steve!  He really gets it, and I do too thanks to him.  In this case, I suspect he was in a hurry and just didn’t give the sub-topic enough attention.

So allow me. 😜

The word “independent” is doing a lot of heavy lifting there. The implication is that all other possible explanations for the correlation have been factored out, leaving “stretching” as the only plausible explanation standing. But life and research are messy, and that’s extremely hard to establish. Whatever science it’s based on, I predicted that it wouldn’t actually do the job.

So what’s the source? He links to a Harvard blog post about the study that leads with the strong implication that boosting flexibility is an active ingredient in longevity, and not just something that came along for the ride:

“Can increasing your joint flexibility help you live longer? … it might.”

And the post is in turn about a 2024 study it does not link to (sigh), Araújo et al:

Araújo CGS, de Souza E Silva CG, Kunutsor SK, et al. Reduced Body Flexibility Is Associated With Poor Survival in Middle-Aged Men and Women: A Prospective Cohort Study. Scand J Med Sci Sports. 2024 Aug;34(8):e14708. PubMed 39165228 ❐

They followed 3,139 middle-aged subjects over 13 years, scoring their flexibility across 20 movements in 7 joints, reporting that subjects with poor flexibility had shorter lives. But this was absolutely not designed to show that the lives were shorter because of the poor flexibility.

So how about all the potential confounders? All the other things that could explain the link? Did they control for all of those? Hell no! They barely scratched the surface. They didn’t even control for diet or physical activity — the most obvious possibility. And the sample was primarily affluent white Brazilians. So reduced flexibility could easily have been a marker of poor lifestyle habits rather than a causal factor in reduced mortality. The authors acknowledge this limitation; it’s a glaring gap, and they know it. But they also dismiss it with a poor argument, and overstate the case for causality and its clinical implications.

There was nothing “independent” about this association.

The flexible subjects were probably not blessed with longer lives because they were flexible; they were probably more flexible because they had been active and health-focused, and in turn that’s likely thanks in large part to genetic gifts and luck.

This study was indeed a textbook example of “correlation is not causation.” Sometimes a correlation is indeed a valuable hint about causation, but only in the absence of more plausible explanations for the link.

I am a Dr. Novella fan, and have been for a long time, and I feel disloyal writing about something I think he got wrong when he gets so much right. But I couldn’t let this pass! There is (still) no evidence that stretching is a pillar of health and fitness, or therefore of longevity, or therefore an inoculation from the aches and pains of aging.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

Many horror fans hate this film for being “boring,” but it’s exactly what I want out of the genre: genuine, emotionally mature human drama in a crazy context.

And then there was the physiotherapy scene. 🙂

Daisy Ridley plays a physical therapist … someone begs her for shoulder help … she reluctantly goes to work, and … massage. And hurts-so-good massage, of course. “It’s horrible, don’t stop.”

Interesting.

Not that I expected the plot to pause for an interview and physical exam, of course, but I thought was a fascinating oversimplification. Sore body? Rub it! Dig into those sore spots! Duh!

Great movie overall.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

[ Image caption: “Thinking at Hell’s gate”, by innoxiuss, CC BY 2.0 ]

“It worked for me.” Almost everyone, sooner or later, feels like they got some blessed relief from a remedy of dubious value — something that’s not supposed to work according to the science, something “alternative,” maybe something straight out of the folk medicine of people from the other side of the world. Perhaps it was gua sha, or Epsom salts or a homeopathic arnica cream. Maybe it was something a bit more modern-seeming but damned with faint praise by the studies, like laser or shockwave therapy.

How could such a thing happen? Could the science just be wrong?

Could the studies have missed people like you?

It’s conceivable.

There are, of course, many ways for treatments to seem effective when something else actually deserves the credit. But most people don’t know about most of those, and even card-carrying skeptics often feel certain that the treatment worked, somehow, and the science be damned!

The #1 pretentious rationalization for these apparent successes is this classic half-truth about the limitations of science:

“But scientific trials show us the average response to a treatment. It may still work well for some people! It worked for me, after all.”

Or “it works for some of my patients” — because this idea is more often floated by professionals rationalizing their clinical choices.

This is the outlier gambit. It is often (suspiciously) extended into a rant against evidence-based medicine itself, scoffing at the futility of testing treatments at all. It’s not entirely wrong, but most people try to put much more weight on it than it can support.

The idea of “strong responders” is not crazy

The results of trials can indeed fail to reveal significant benefits for a few people who actually do well with a treatment that is unimpressive for most others. And I should probably acknowledge this more often than I do. Trial conclusions mostly focus on the typical response to a treatment, and, yes, sometimes the distribution of results shows high variance, and that spread does include a few genuine “strong responders” in the statistical Good Place.

The clearest cases of important “subgroups” of people who respond much better than average are pharmacological, where known biology makes their existence more plausible (if not almost a certainty). One of the all-time best examples comes from Yarnitsky et al, who showed in 2012 that patients with measurably poor endogenous pain inhibition — a testable glitch, not a guess — responded better to pain meds than patients without that problem. The response varied enormously depending on their capacity for inhibition, a critical clinical reality that was just invisible in the average effect.

That subgroup is real, the predictor is measurable, and the mechanism makes sense. (And no, unfortunately we don’t know why people have poor inhibition.)

Examples like that are a potent seed of truth for the outlier gambit.

Outliers may exist, but there are a lot of very big “buts” here (“and I cannot lie”)

Variance cuts both ways. Variance itself varies, and sometime the range of responses to a treatment isn’t always wide, and sometimes it just doesn’t reach far enough from the mediocre mean to include any strong responders — only “meh” or “slightly better than nothing” responders. It can also hide strong negative responders.

Many, if not most, outliers in small trials are noise, not signal. That noise is already routinely mined for “positive” results that are actually an illusion. They are routinely better explained by regression to the mean, natural recovery, reporting bias, co-interventions (and many other possible confounders) rather than a lucky treatment effect for a special few. Remember that trials routinely fail to effectively control for exactly the kinds of things that will produce meaningless outliers.

A genuine benefit hiding in weak average results is usually rare or modest at best…or it would pull up the average! Unless it was confined to a mysterious subgroup, and such groups are rarer in physical medicine and manual therapy. Good luck trying to find a single good example. One of the best candidates was a “clinical prediction rule” to identify which low-back-pain patients would respond best to manipulation — a treatment with modest average effects (at best). This effort generated some real excitement, precisely because of the common optimism about the power of subgrouping, and how it might validate a kajillion "but it worked for me" anecdotes for an extremely popular treatment.

And then an independent team tested it (Hancock et al) and … sad trombone. No replication. The hope of an important subgroup, a signal strong enough for clear replication, evaporated on contact with fresh data. Bummer. (No really: it is a bummer. I’m not happy about this! I sincerely wish it had worked. Just like I truly wish there was a dinosaur hiding in a Scottish loch, because that would be really cool.)

There are quite a few other notable examples of failing to clearly identify clinically significant subgroups, which is why Rabey et al wrote, “while unidimensional subgrouping has been thought useful to guide treatment, it is unlikely to capture the full complexity of chronic low back pain.” It’s why Saragiotto et al concluded the quest for outliers is “far from optimal and not yet ready to be implemented in clinical practice.”

And so the widespread enthusiasm for special cases is a kind of special pleading. [Wikipedia] It’s what people propose in the conspicuous absence of validated subgroups.

You probably aren’t an outlier

Or, if you’re a clinician, your patients probably aren’t outliers.

If trial evidence is negative — or merely damns a treatment with faint praise, failing to impress — it's probably genuinely ineffective for most people, most of the time. The exceptions do exist, and maybe enough of them to justify cautiously trying iffy treatments once in a while, as long as they're relatively safe and cheap. But the exceptions mostly aren’t important exceptions.

Robust treatment benefits rarely lurk in the statistics, and bad clinical trial news is mostly as bad as it seems. The outlier hope is no more of a genuine win for healthcare than lotteries are a solution to poverty: yes, some people do win, but not many, not much, and not you.

Further reading

• The Power of Barking: Correlation, causation, and how we decide what treatments work — A silly metaphor for a serious point about the confounding power of coincidental and inevitable healing, and why we struggle to interpret our own recovery experiences
• Extraordinary Health Claims — A guide to critical thinking, skepticism, and smart Internet reading about health care
• Studying the Pain Studies — Tips and musings about how to understand (and write about) the extremely flawed science of pain and musculoskeletal medicine

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

That article was junk. 🗑️

It was basically correct junk, but a poor piece of science communication — not thorough, meticulous, or classy. I was mostly preaching to a small choir of skeptics and enraging professionals invested in that technique. It was also a reckless full-frontal assault on a specific brand, a legal threat magnet. I was young and pissed off. 😜

So I kicked that trash to the curb ages ago. Rewriting was too big a job, so I dropped the topic in my maybe someday hole — a place many writing goals never escape from.

But that someday has arrived!

This is the second coming of that old article: a substantive and open-minded critical analysis of scraping massage, AKA instrument-assisted soft tissue mobilization (IASTM), or its Chinese ancestor gua sha.

[ Image caption: Care for a little scraping massage? The steel massage tools of Graston Technique®. GTindy, CC BY-SA 3.0 | upscaled, cropped ]

This post is pointedly not about any specific modality empire. Brands are named but not singled out. This is a tour of the pros and cons of the generic concepts that are the basis for those brands. A distinction without a difference, perhaps? But there is a legal difference, and this is how I do it these days. There is still snarky skepticism (I yam what I yam), but also balance and diplomacy.

And this reincarnated article is members only temporarily — a relatively “safe space” for it for now, but I’ll release it into the wild later this year. Topics in the post:

• Scraping massage big picture
• Gua sha roots: scraping massage as Chinese folk medicine
• Gua sha is not necessarily rough … or superstitious.
• Supposedly advanced western “instrument-assisted soft tissue mobilization”
• What’s it good for? The goals and methods of IASTM
• Safety of scraping massage, especially IASTM
• Can you scrape your pains away? The science report
• 96.5% effective! Success rate claims (and insurance claims)
• Is gua sha a waste of time and money if you like how it feels?

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

Ilio-what now? A quick primer: IT band syndrome (ITBS) is one of the two common kinds of runners’ knee (with patellofemoral pain). It’s a tendinitis-y overuse injury of the huge iliotibial band, a massive connective tissue structure on the side of the thigh, and surprisingly weird and notoriously stubborn. I started writing about it after I got a nasty case of it myself on an “exciting” hiking trip (grizzly included, one of my best stories).

I’ve been writing about ITBS for about twenty years, so you’d think I’d have the topic licked by now, but my work is literally never done. Due to lack of any obvious updates, the book hadn’t been updated in quite a while, probably the biggest maintenance lull in its history. But it only took a few minutes of review to reveal plenty of work to do. So I dove in, and here are just the highlights from the changes to the book this year so far. (Although the full updates are all in a paywalled book, in many cases here you can get free detailed notes on the citations.)

• I rewrote the chapter “IT band syndrome may be more about physiological vulnerability than knee stress,” which directly addresses the “surprisingly weird and notoriously stubborn” nature of the beast. Before revision: a minor chapter musing about some of the more mysterious cases. After: a substantial, evidence-based chapter that sheds some genuine light on the problem.
• Another update to that chapter, complementary but separate, was a drop-in citation based on a lot of analysis of a low-quality genetic study of the causes of ITBS … which I dug into anyway, because it inadvertently undermines the authors’ pet theory that it’s all about hip weakness. That analysis is free to anyone in the PainSci bibliography, see Yao et al.
• One study, Sanchez-Gomez et al., was good for two seemingly unrelated updates:
  • Justification for a tepid recommendation to at least consider orthotics to limit supination (not pronation). Kind of a long shot, and at odds with one of my biases, but it would have been intellectually dishonest not to include it, so … included.
  • A good fresh example of very weak science about the role of the TFL in IT band syndrome.
• Cited He et al. on the state and history of ITBS research, and the most cited papers on this topic, which was a nice find … although not as nice as it might have been, because fully half of them have almost nothing to do with ITBS. Weirdly and annoyingly, they are mostly about using pieces of the tough IT band like a patch in ACL injuries. But I did make sure that all the others are cited in the book, which was easy because most were already there.
• One of the few most-cited papers I’d never cited before myself was a 2010 paper by Powers. That should have been in the book since at least 2015 just because it was so popular — and now it is, FWIW. Not a good paper, though: just a single author’s opinion that “motion impairments at the hip may underlie [knee] injuries” based on his read of the literature.
• Speaking of hip weakness, I also added a much fresher and more substantive citation to Leppänen et al., which is actually more important, and it challenges my bias against hip weakness as a cause of IT band syndrome. One of the only studies that does. 😜
• And still more hip-is-the-knee-key stuff, and another entertainingly bad study: Jahanshahi et al., a trial of functional motor control exercises for wrestlers with ITBS. It had suspiciously huge effects and was unblinded, among other critical flaws — barely better than propaganda for a belief. (Why do I cite low-value studies? We must make do with scientific scraps in this business.)
• I updated the introduction with prevalence data from Taunton et al. and Marais et al.
• I added a fascinating discussion of ITBS induced by surgery. Takagi et al. is a very illuminating report on four cases of ITBS after knee replacement, all involving obstructions under the IT band.
• Manon et al.. is a cool study of IT band microanatomy and its layering and fibre directions, which was a nice addition to a good collection of other anatomical “fun facts” about the IT band. And more anatomy: cited Claes on the re-discovery of the “anterolateral ligament” in 2013.
• Cited and discussed two experiments, Heiderscheit et al. and Baggaley et al. showing reduced knee loading with shorter, quicker steps — a good addition to the chapter on the role of tempo/pace.

Sooooo many updates! A lot of work, but worth it for a large batch of new readers.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

[ Image caption: I wonder what AI prompt was used to generate this image? Make one of them look like a burn victim? 🙄 ]

It is, of course, horrible nonsense, pure marketing/propaganda. It inflates the perceived value of magic hands, and tries to make not getting regular massage seem dangerous. It’s a flagrant visual lie that makes the profession of massage therapy look ridiculous to anyone with an ounce of sense or knowledge. Anyone who wants to promote this idea is probably not going to be moved by any criticism of it, but people who don’t have “skin in the game” (ha ha) might benefit from seeing more knowledgeable people rolling our eyes at it.

As ridiculous as this image is, though … it actually does represent the spirit of what a great many massage and manual therapists truly do believe, even if they know better than to think it literally looks like this under your skin. They might argue that it feels like that, or that it’s just an exaggeration of what it looks like on a smaller scale. The whole basis for most flavours of “fascial therapy” is the incredibly prevalent idea of soft tissue “distortions.” And this image is basically “what distortions supposedly look like.” From my guide to fascia pseudoscience:

“The idea that fascia is medically important has its most important roots in the 1990s, when physician Stephen Typaldos formulated the Fascial Distortion Model (FDM) in 1991, proposing that most musculoskeletal complaints are caused by deformations of fascia. Typaldos thought deformed fascia could be fixed by skilful force, kind of like banging the dents out of a car.”

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

Many people, including a lot of experts, would guess the knee of the intact leg, “particularly affecting the medial compartment of the intact-side knee due to increased joint contact forces possibly resulting from altered gait biomechanics.” That is the basis for the conventional wisdom.

But a new study by Watson et al shows the opposite: “significantly worse [radiographic] osteoarthritis on the amputation-side knee 11-years post-amputation.”

A couple caveats before we interpret that: the sample is modest (36) and drawn from a modern military population with good rehabilitation and prosthetic access, and it just might not go like this for civilian amputees. And the study reports a difference seen on x-ray only. Actual symptoms were the same despite the fact that amputation-side knee cartilage was objectively more degraded. (But… care to bet on how they’ll all feel in another decade?)

So, why would the amputation side degenerate faster?

Many people are now aware that osteoarthritis has never been a simple biomechanical wear-and-tear disease, but a complex physiological/inflammatory one. Some other posts and articles on this theme:

• Bone on Bone
• Preventing arthritis and tendinitis is easier than you think (with patience)
• Ozempic works for arthritis, and probably other pain (Member Post)
• Running does not wreck knees (probably)

Load is involved, but it’s not the essence of the thing. This modern view of arthritis has taken root enough now that a lot of people will claim that they are not surprised by this result. Instead, their assumption is that it is explained by the lack of loading on the amputation side.

Not so fast! As the authors explain, the intensity of loading of the amputation side knee is known to be either similar or just a little bit less. It’s probably variable, and in some cases it may even be more, as reported by one PainSci reader, a below-knee amputee: “I load bear more on my amputated side.”

Awkward loading of the amputation side knee, possibly worsened by janky prosthetics in some cases, is another possible explanation.

But, again, loading is just generally not the major driver of arthritis. It might combine poorly with the physiology; it isn't irrelevant. But it’s probably not what really matters.

What (probably) really explains the result

All load-based explanations — under, over, and/or awkward loading — are probably overshadowed by the most plausible explanation: long-term inflammation and physiological dysfunction associated with major trauma, both the original injury and the amputation. Specifically and fascinatingly, it could also reflect previously under-appreciated consequences of high-energy injuries, lots of microtrauma throughout the leg:

“We hypothesise that blast injury could instigate degenerative processes in the cartilage (42) within the amputation-side knee leading to early increased radiographic knee OA. These high-energy injuries cause all the reported predictors of poor outcome: primary high-energy insult to the articular cartilage, comminuted joint surface, articular bone loss, soft-tissue damage, and open and contaminated wounds susceptible to infectious complications.”

But it’s complicated and there are caveats every which way. This paper is well-written, and there's a good and thorough “Discussion” section with much more nuance than I’ve provided here, and the full text is available.

I was particularly interested in this paper because my father, although not an amputee, is a veteran with a major gunshot wound, a shattered femur. He is quite keen on evidence about the effects of high-energy trauma! You can read his story here.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

(It’s not useful or neat for the rats! As always, we are grateful to the rats for their sacrifice. Thank you, rats: you are lovely little critters, despite your naked tails. We truly regret that you suffer so that primates can try to understand biology better.)

Micro microglia primer

Rat brain changes were observed in microglial cells. “Glial” cells are all the brain’s support cells collectively, like the astrocytes, oligodendrocytes, and ependymal cells. Micro-gial cells are the immune-derived glial cells. Microglia are glia, but not all glia are microglia. They largely stay put, but have long and dynamic arms. Creepy.

[ Image caption: Artistic representation of a typical microglial cell. They are very hard to image — not particularly photogenic! Mostly just low-resolution blobs, like this. [Wikimedia] Those arms are always moving, and can extend up to several times the width of the cell body, over a few minutes. This is a modified version of an image by Microsome, CC BY 3.0, via Wikimedia Commons. ]

Microglial cells patrol neural tissue like cop-janitor hybrids, and there are billions of them in the brain, roughly one for every 10 neurons — like if 10% of people were heavily armed sanitation professionals. They aren’t neurons, but they work with them extremely closely, and probably affect many aspects of brain function. They’re sensitive to signals from both brain and body, which may explain why systemic inflammation can fog your thinking. They have many jobs: pruning unnecessary synapses during development, clearing cellular debris, responding to infections and traumas.

Like most immune cells, microglia come with trade-offs. You don’t want them to be too active for too long. Which is why this research was a little troubling!

Rat brain changes with sciatica mapped in detail

And now back to today’s science. Here’s the full citation:

Cazuza RA, Laoumtzis M, Lacagnina MJ, et al. Brain-wide changes in microglial morphology following peripheral nerve injury. J Pain. 2026 Jan;38:105582. PubMed 41109469 ❐

Cazuza et al. mapped how microglia in the brain change their shape after nerve injury, which is a proxy for how “activated” or inflamed they are. They induced sciatica with squeezing, and then examined 52 different brain regions at 7 and 28 days afterward, using automated software to measure microglial shape changes at scale. This was quite ambitious.

Microglial activation wasn’t confined to pain-sensing areas: it showed up across brain regions involved in emotion, cognition, motivation, and stress regulation. This fits with the idea that chronic pain is a whole-brain deal, not just a sensory one.

This isn’t especially surprising: we’ve known for more than 20 years that microglia react to injuries out in the body, although the majority of that research has focused on the spine. What’s new here is the high-resolution wide-angle view of the brain over time, producing the big takeaway: the pattern of microglial activation shifted as the pain dragged on.

At day 7, the activation action was in the motor cortex.

But by day 28? The shape-shifting shifted to the hypothalamus and reward circuitry — regions associated with stress, sleep, and motivation. All of which are commonly disrupted in chronic pain patients.

Neurology “news you can use”? Reaching for why microglia might matter

But … “in rats.” Animal models aren’t great. Shape-shifting rat microglia don’t necessarily mean functional change; activated-looking cells aren’t necessarily doing what we assume, and we know that rodent and human microglia are not identical. Microglial research is one of the best examples of pain science that’s intriguing for nerds, but does little for people in pain. But it is neat to know that the brain measurably changes as pain drags on, and here are a few possible practical ramifications:

• If microglial botheration can spread, it might explain why chronic pain often comes with depression, insomnia, and brain fog — and suggests the need to treat those downstream effects of pain, if the pain itself can’t be relieved.
• It punctuates the value of timely and competent diagnosis, leading to any possible relief from the pain — because pain isn’t “just” pain. It has complications!
• It hints that it might be important not to wait for a new pain to become chronic before taking it seriously (“come back and see me again if this doesn’t go away” might be bad advice).
• It suggests that even “ineffective” and “temporary” treatments that “mask” pain might actually have value: they could prevent the spread of microglial arousal. Just because pain causes brain changes doesn’t mean analgesia can prevent or reverse them, but it’s a thread worth pulling on.
• It might be a particularly good reason not to push your luck with “powering through” painful problems — which is especially straightforward in the case of overuse injuries.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

A new meta-analysis by Saez-Berlanga et al. pooled data from 23 trials and reported that elastic‑band resistance training (EBRT) produces meaningful reductions in body fat, fat mass, total and LDL cholesterol, triglycerides, fasting glucose and blood pressure … and a boost for HDL. Yahtzee.

And metabolic health upgrades are highly relevant to musculoskeletal health and pain. Seriously.

The subjects were older, mostly sedentary adults at risk of metabolic syndrome. The authors pooled 23 trials including 1,134 participants (average age 70.4 years; 92% women) and searched literature through September 2024.

Compared with non‑active controls, EBRT produced modest but consistent improvements in body composition and cardiometabolic risk factors. Body fat percentage fell (~1.44 percentage points), and fat mass fell by ~1.6 kg. Total cholesterol dropped by 18.2 mg/dL and LDL by 11.0 mg/dL. Triglycerides dropped ~12 mg/dL. HDL rose by 2.4 mg/dL. Fasting glucose decreased ~9 mg/dL. Blood pressure fell, systolic −7.7 mmHg, diastolic −5.5 mmHg.

The authors found minimal evidence of publication bias. This is relatively straightforward science, and completely unsurprising — we've known for a long time that resistance training is generally just as good as aerobic workouts. Exercise is exercise. Muscles are endocrine organs that, when stimulated, produce all kinds of healthy signals. And you can stimulate muscles perfectly well by pumping rubber instead of iron.

Elastics aren't for everyone, of course. This is just another option, and it’s nice to have those, but there are many others — that's really part of the point. Always aim to be active in whatever way suits you best! Perhaps you prefer a nice music dance experience, for instance. But for many, a basket of big rubber bands in the TV room is a way nicer option than a rack of iron at the gym.

This is the FREE RSS feed, which delivers the full text of most posts. But PainSci members get their own feed with every post, plus audio versions by podcast, and more. And of course this nagging blurb goes away. 🙂 JOIN NOW for USD $5/month (or $50/year).

Web version of this post.

I guess I should have stayed put.

I’m hardly alone. I’ve heard many stories of tendons that ruptured with little provocation: Stepped off a curb. Reached into the back seat. Caught a falling coffee mug. Sneezed hard.

Partial ruptures like mine are probably even more common, but often undiagnosed. It was less dramatic than a full rupture, of course, but I could feel it tearing, an awful sensation: a wet, shuddering giving way. It felt just like the last time I definitively tore some connective tissue (coracoclavicular ligament rupture, a sports accident in the mid 2000s). A memorable sensation.

[ Image caption: Like rope fibers, tendon collagen can fray and fail under stress. But tendon fibers are a lot more complex than rope fibers, and fail for much more complicated reasons. ]

I didn’t curse. I just groaned and rolled my eyes and sighed the sigh of the defeated. I have already endured so many insults like this, what’s one more? But perhaps I should have cursed, because it turned out that this injury was probably the tip of a much more disturbing iceberg. This tendon tear was — drum roll please — quite possibly a drug side effect.

Say what now? Yes, you heard that right: tendon rupture as the side effect of a medication. A side effect which I had never heard of before, despite my expertise — despite even knowing about other drugs that do this. This post continues a new theme of “hidden sources of tissue fragility” and other surprising ways to hurt, and digs deep into tendinopathy induced by drugs, especially the statins for high cholesterol. How does that even work? And is it truly happening, or are the drugs innocent? The goal here is practical awareness without fear-mongering: when tendons suffer and fail, meds should be on the shortlist of suspects.