Hypermobility and Exercise: Why Traditional Strength Training Falls Short (Part 1)

• About
• Latest Posts

Adam Foster

Director at The Fibro Guy

Adam Foster is the founder of The Fibro Guy and one of the leading voices in hypermobility, EDS, and chronic pain rehabilitation. A former Reconnaissance Soldier, he developed chronic pain after an IED blast in Afghanistan and built his approach from the ground up when the medical system had no answers. Since 2009 he has worked with thousands of people across eight countries, drawing on pain neuroscience, motor learning, and sensorimotor rehabilitation. He co-authored "No Pain, No Pain" with trauma psychologist Dr Mairi Harper, directed the documentary "Invisible Monster," and has been featured in Forbes, BBC Radio 4, ITV News, LiveScience, and The Telegraph.

Latest posts by Adam Foster (see all)

• Why ADHD and Hypermobility Keep Turning Up Together - 16 September 2026
• Hypermobility and Early Development: Milestones, Motor Delay, and Predictions - 12 September 2026
• Fibromyalgia and Itching: Why It Happens, What the Science Actually Says, and What You Can Do About It - 9 September 2026

Somewhere between the third physio and the second gym membership, nearly everyone with hypermobility is handed the same sentence: you need to strengthen the muscles around the joint. It isn’t bad advice exactly, it’s just advice that stops about halfway through the full explanation. So, you go and do it, you do it properly. You do the bands, you do the clams, you draw the navel to the spine, and then you hold it for ten, three times a week, for eight months. Then you reach up to a high shelf for a mug, and the shoulder slips out anyway.

Nobody ever explains this bit, and thats what this article is about.

Let’s do the answer first though, rather than making you scroll for it.

Strength training isn’t the villain here, and getting stronger is genuinely worth doing. What strength training doesn’t do on its own though, is teach a joint when to stiffen and when to let go. That’s a timing and information, and it’s the part almost nobody ever touches on, which is unfortunate, because it’s also the part that decides whether you can trust your ankle on the pavement, during the icy months.

Berrie, and Ten Weeks Without a Single Weight

Before any of the science, I want to show you a person, because this is easier to follow if you can see it happening to somebody.

When we first met Berrie she was in a genuinely bad place. She could hardly walk, her joints were subluxing constantly, and she was so tired and so foggy, that she had more or less stopped talking altogether.

For the first ten weeks of working with her, we did not give her a single weight. Not one. Nor a single clam or bridge (which is another conversation)

What we did instead, was work on the quality of the information going in. The feet, the hands, the skin, the breath, the position of the joints she was actually using, in the positions she was actually using them (we are and always have been, big advocates of transferability when it comes exercise). All of it aimed at one thing: making the signal arriving at her brain clearer than it was, so that the movement coming back out of her brain had something better to be built upon. Get that loop going in the right direction, where clearer input produces better output, which produces clearer input again, and you have something to build on. Skip it, and you are adding load to a system that can’t tell you where its own knee is, let alone with external weight going through.

Then we loaded her. Properly, progressively, and with a great deal of thought about the order things went in. That’s the part people are frightened of, and honestly, fair enough, as being told to lift heavy while your shoulder is subluxing twice a day, sounds like somebody who has never had a shoulder sublux. But, it’s needed. It just needs doing after you have some genuine control of the joint, rather than instead of getting it.

And we hammered motor learning principles the whole way through, changing depending on how she was that day, which we’ll come back to properly further down. Same goal, never quite the same solution, over and over, until she had a rule for the movement, rather than one memorised copy of it, until she didn’t have to think about moving.

Where she got to: eating, actually hungry, joints nice and stable, and best of all, I couldn’t get her to shut up. You can watch a bit of it below if you’d like to see what that looked like in practice.

Now, that order of operations, signal first and load second, is our read. It’s built on the mechanisms and on watching hundreds of people go through it in the studios. Nobody has run the head to head study that would prove it beats doing it the other way round. However, good ,solid research, is very much lacking in this area. I do have a feeling there will be studies that come out to show it, but for now, we are left waiting.

Driving in the Fog

Think for a second about what a muscle actually has to do to keep a joint where it belongs. It has to fire at the right moment, in the right order, with the right amount of force, in the right direction, and all of that has to be based on information about where the joint currently is and where it’s heading (you have probably already spotted the problem here).

Strength is but one item on that list.

It’s a bit like driving in thick fog. You’re not driving badly because the engine doesn’t have enough pull, you’re driving badly because you can’t see, and bolting a bigger engine into that car doesn’t improve the situation, it arguably makes it considerably worse. Most hypermobility rehab is engine work. It’s good engine work dont get me wrong, a lot of the time it’s very well designed engine work, it’s just aimed at the one variable that wasn’t the problem.

And you can see that in the numbers, in a roundabout way. Muscle strength does track with how limited someone is in hEDS, but the relationship is tangled up with proprioception, and it’s strength corrected for the sensory side of things that carries the association, rather than strength on its own ^[19]. That’s easy to over read, and we’re not going to over read it: it doesn’t prove that sensory work has to come first. What it does say, pretty clearly, is that measuring one while ignoring the other gives you a misleading answer.

There’s a bigger point sitting underneath all of this though, as disability in hypermobility related disorders is substantial, and it tracks pain, fatigue and distress, rather than tracking how hypermobile anybody is ^[5]. In hypermobile adolescents with chronic musculoskeletal pain, the route into being unable to do things runs through the pain and the distress that comes with it, not through how far the elbows go back ^[6]. Fatigue isn’t merely just a footnote either, as it shows up in how you literally walk. ^[7]. And the pain in hEDS looks generalised rather than local, with pressure pain thresholds lowered in places that don’t even hurt, which points at a nervous system that has turned its dial up, rather than at nineteen separate joint problems ^[8].

So “just get stronger” isn’t wrong, so much as aimed at the wrong target. It’s answering a question about tissue, while most of what’s actually ruining your week is coming from pain, fatigue, fear and timing.

What the Exercise Research Actually Amounts To

Exercise helps, that much is about as settled as anything gets in this field, and it turns up again and again across hEDS and HSD, on pain, on function, on proprioception, and on quality of life ^[1][2].

The trouble starts the moment you ask a harder question, which is: which exercise, in what order, at what dose, and for whom. Nobody has shown any of that (surprise, surprise). What exists is small, short, and mostly not built to compare one approach against another, so what you end up with is a pile of work in which everybody improved a bit and nobody can tell you what did the improving ^[1][2][3]. Where two programmes have been put head to head, the difference between them have generally been unimpressive, which is either reassuring or infuriating, depending on what you wanted out of them.

Now, that’s a real limit, but notice what it doesn’t sayL It doesn’t say strengthening is a failed model. The more recent syntheses land on combining therapeutic exercises with motor function training and individualising how it progresses, rather than just swapping one out for the other ^[4]. Which means a clean hierarchy, signal first and strength second, is ahead of the evidence, and where we work in that order it’s on the strength of what we watch happen rather than because anybody has proven it (for now).

Heavy Lifting Is Not Off Limits, and That Surprised Us Too

This is the part of the old story that has moved the most, and it’s the part where a lot of the internet is still about a decade behind.

For a long time the working assumption in hypermobility rehab was that heavy load was for other people. A lax joint needed careful, light, controlled work, forever, and preferably in pastel colours. That assumption has since been tested, and it didn’t survive.

In adults with hypermobility spectrum disorder whose shoulders had been painful or unstable for months, sixteen weeks of full range, high load, supervised strengthening produced better self reported shoulder function than the low load version of the same programme ^[9]. And read who was in that group, because this is the bit that usually gets lost on people: these were people with genuine shoulder symptoms, instability, and the sort of history that normally gets somebody told to avoid weights entirely. Not comfortable hypermobile gym goers. Nobody came to harm. The heavier group reported more of the ordinary post gym soreness though, and some headaches, and that was about the extent of it ^[9]. Earlier work in the same population found people stuck with it, strength went up nicely, and the complaints were all transient ^[10].

Two caveats though, and both of them matter a great deal.

Pain barely moved in either group, and the gap between heavy and light on pain wasn’t meaningful ^[11]. So if your reason for lifting heavy is that you want the pain gone by Christmas, that is not what was shown. And by a year out, the advantage had disappeared, with the heavy group no longer reporting better overall function than the light one ^[12]. The fair read is a short term functional gain rather than a permanent lead.

This pattern holds elsewhere too, as two different home based shoulder programmes, both improved things over six months in hEDS and HSD with multidirectional instability, without either of them running away with it ^[13], and supervised heavy resistance training was tolerated perfectly well in women with knee pain and knee hypermobility ^[14]. Pull the whole shoulder literature together and what you get is that conservative work helps, and that we still can’t tell a clinician which version of it to pick ^[15].

Two things not to do with that in mind!

Don’t import general resistance training guidance written for healthy adults and assume it transfers, as guidance on building muscle in people whose joints stay where they’re put, cannot answer what a subluxing shoulder should be doing during a shoulder session at “The Gym group” in town^[16]. And keep the symptomatic and the asymptomatic groups separate in your head, because being hypermobile without symptoms and being hypermobile with pain, poor balance and altered movement, are not the same situation at all, and the second one is where all the burden sits ^[17]. Worth knowing, by the way, that even the groups which look like they were made up of comfortable hypermobile people, often weren’t, as when resistance training was tested in one group of women with generalised joint hypermobility, most of them reported pain at the start ^[18].

So, load is genuinely on the table. What isn’t on the table though, is the idea that load alone finishes the job, which is the other half of the internet, and it’s the half that leaves people doing barbell work with a shoulder that still promptly exits the socket when they put a jumper on.

Your Proprioception, It’s Noisy

Proprioception is just your sense of where your own body is without looking at it. Close your eyes and touch your nose, that’s proprioception, and it’s running constantly in the background whether you’re thinking about it or not. And there is no end to instagram coaches slapping it in their videos as buzzword, with very little understanding of it.

In hypermobility, it’s generally a bit worse. Lower limb joint position sense and the ability to detect that a joint has started moving are both poorer, with the knee findings more consistent than the shoulder ones, so this isn’t uniform across the body and anyone telling you that it is has skipped the details ^[20][21]. Ankle position sense and the sensation on the sole of the foot are both poorer in hypermobile children ^[22], and reduced proprioception alongside altered strength and functional performance turns up again in more recent work ^[23].

Now, you might be reading all that thinking your proprioception feels absolutely fine, and that’s a fair objection, because on a standard position matching test, plenty of those with hypermobility look completely unremarkable.

The interesting part is precision rather than accuracy. Asked to locate their own hand with nothing but the sensation from the limb to go on, those with EDS land in roughly the right place, they’re just a lot more scattered about it than everyone else ^[24]. Which is a genuinely different problem from being wrong. Think of a sat nav with a poor lock on the satellites: it isn’t lying to you, it’s telling you that you’re somewhere within thirty metres of where you are, which is fine on a motorway, but pretty useless in a multi storey car park. Your brain is making predictions off that signal, and a prediction built on a scattered input produces an output that is sometimes too much, sometimes too little, and quite often, just slightly late. And slightly late is how a knee ends up in hyperextension before anything has fired to stop it.

Two pieces of good news sit inside all of that though, and neither gets said nearly often enough.

The first is that proprioceptive recalibration in EDS happens to the same extent as it does in everybody else ^[25]. The system learns from a noisy signal perfectly well, once it’s given something to learn from. Nothing about this is a fixed ceiling, which is worth holding onto if you’ve been told some version of “well, that’s your connective tissue, there you go, have you tried meditating?”.

The second is that in children with hypermobile knees, acuity into the hyperextended range was no worse than in early flexion during weight bearing testing ^[26]. Which quietly kills one of the most common instructions in this whole space, the one where you’re told to train only in the middle of the range and treat the end of it as forbidden territory forever. Train the range you actually live in, because you do live in it, whether your programme acknowledges it or not.

What nobody has shown, and this is the flag we promised earlier, is that cleaning up the signal before adding load produces better outcomes than adding load and letting the signal catch up. That comparison has not been run in a hypermobile population. Our order of operations is a clinical read, built on the mechanism and on what we watch happen in the studios, and it should be held as exactly that.

We’ve always been fairly Switzerland about rehab, and that’s deliberate. We’re not in love with any particular protocol, we don’t have a method with our name stamped on it that has to be defended, and there’s nothing riding on any one approach turning out to be the answer. We just care about helping people. So if some of what we teach now gets overturned in five years because the evidence moved, that’s fine, we’ll teach the new thing and say why we changed our minds.

And when we do step outside the usual box, we say so, and we explain the reasoning.

That last part is where this space falls down though, as transparency is in remarkably short supply, and you can see what it produces. Pilates is the best, no, yoga is the best, no, my method is the best, and round it goes forever. Everybody plants a flag and then spends their time defending it, which is a great deal easier than saying the honest thing, which is usually “it depends”, followed by a fairly long explanation of what it depends on.

Meanwhile the person in the middle of all that, who only wants to know what to actually do this week, ends up having to pick a team.

Honestly, we couldn’t care less which camp a useful idea arrived from, and we’re not against any of it (Pilates and yoga both have their place, they just aren’t the whole answer for everybody). What we care about is that the person reading it gets somewhere they belong, gets the correct information at the time, gets the education to understand why it works, and ends up on a path that’s genuinely worth walking down.

If that means changing our minds in public every so often, we’re fine with that.

Motor Learning, In Plain English

Motor learning is how something goes from needing every bit of your attention, to happening on its own while you think about what to cook for dinner. It’s the difference between being able to do a movement and actually owning it, and it is, in our opinion, the single most underused idea in hypermobility rehab. I have not once, ever seen it mentioned in a blog or YouTube video, or anything for that matter (which is nuts)

The basic mechanism isn’t complicated: you sense what’s happening, you predict what should happen next, you compare the two, you update, and then you do it again, slightly differently. The old formulation of it is repetition without repetition, meaning the goal stays the same while the solution never quite does, because a body is not a machine reproducing an identical output and it never was ^[27]. There’s rarely one correct way to do anything either (which we have already showed 50 times already in this article), as you have more joints and muscles available than the task strictly needs, and that surplus is a resource, rather than a nuisance ^[28].

Sitting alongside that, is the idea that the brain stores a flexible rule rather than a fixed copy of a movement. Which is why practising a spread of versions should generalise better, than just drilling one of them ^[29]. And if that’s right, then doing the same exercise, at the same tempo, on the same mat, teaches your nervous system that exact situation, and not a great deal else. It’s the difference between learning a route and learning to read a map. The route works beautifully until somebody closes the road and you get diverted. Think of it like walking down the street in summer versus the icy winter, you’re still doing “walking” but external factors, like Ice, cause you to deviate a little and adapt how you walk.

That’s most of how we programme, and it’s what we did with Berrie the whole way through. But the evidence for it is softer than it usually gets presented, and we’d rather tell you that than sell you an absolute (after all, only Sith deal in absolutes).

When the varied practice prediction was examined across decades of experiments, the support came out as limited rather than emphatic, and a fair few of those experiments weren’t really testing it properly anyway (I know, rather annoying) ^[30]. More variation isn’t automatically better either, as low to moderate amounts have outperformed heavy amounts, with the right level depending on the person and the task ^[31]. Mixing practice up does tend to help what you retain, though the advantage is at its most impressive in a laboratory and can shrink to almost nothing once you’re out in the real world ^[32]. The broad pattern across learning generally is that variability makes the early going harder and the later generalising easier, which is a trade rather than a free lunch ^[33].

So, the honest version is this: vary things, but vary them deliberately and moderately, and stop well short of turning every session into a different circus. Nobody has directly compared varied against repetitive practice in a hypermobile population, so all of the above is imported from elsewhere, and it’s a reason to suspect rather than a reason to believe.

Where You Point Your Attention Changes the Movement

There’s a long standing finding in motor learning that where you aim your attention changes the movement itself. Focus internally, on the muscle, feel your glute switch on, and both performance and learning tend to come out worse than if you focus externally, on the outcome or on something in the environment ^[34].

That got treated as gospel for about twenty years and it probably shouldn’t have been, as a careful reanalysis of the whole area found substantial reporting bias and a much smaller average effect than the textbook version implies ^[35]. So the instruction isn’t “external focus always”. It’s that the cue is a variable you can change, and internal cues are not free.

Somebody has actually run this in generalised joint hypermobility (crazy right?), which is rare enough in this field to be worth stopping on. But, keep in mind “generalise joint hypemrobility”…….better than nothing I suppose.

Cueing attention externally improved postural stability and balance, more than internal or neutral instructions did, in hypermobile people specifically ^[36]. And in the same work, both internal and external cues beat neutral instructions for quadriceps peak torque ^[36]. Read both halves of that, because they point very different ways. For balance and control, you point the attention outwards. For producing force, telling somebody to think about something specific beat telling them nothing at all, whichever direction you happened to point it.

Which is a considerably more useful conclusion than the version that used to be given out, as it means internal cues aren’t poison, they’re just a poor default for the thing most of those with hypermobility are actually trying to fix. If you want a shoulder that behaves when you reach for a mug, then “reach for the mug” is a better instruction than “set your scapula”, and it’s also a lot easier to say to somebody who is already halfway out of a chair.

The Pilates Question, Handled Honestly

We get asked about this constantly, usually by somebody who has done three years of it and wants to know why their shoulder still subluxes. Our position on it has historically been sharper than the evidence justifies, so it’s worth going through properly rather than repeating the line.

For a start, Pilates isn’t one thing. It’s an umbrella covering an extremely wide spread of practices, formats, equipment, dosages and teaching styles, and recent work is explicit that treating it as a single intervention is a mistake ^[37][38]. A reformer session with a very experienced teacher, and a Wednesday night mat class in a village hall, share a name and not much else.

The direct evidence in hypermobility also isn’t negative. An online modified Pilates programme for those with hypermobility produced modest improvements in the impact of hypermobility, in body awareness, and in kinesiophobia, which is the fear of moving that keeps so many people parked at the starting line ^[39]. Talk to the people who did it and you get reported physical and psychological benefit alongside some very real barriers, mainly pain and fatigue ^[40]. In hypermobile children with chronic musculoskeletal pain the picture is thinner and more mixed, with some gains in fatigue, strength and quality of life but no clear improvement in pain or postural control ^[41]. So, there’s genuinely something there, and a blanket claim that Pilates keeps you stuck isn’t supportable.

Now, the variability question, which is where our actual argument lives. How movement variability is used in Pilates as a learning principle has barely been examined at all, so the honest statement is that nobody has studied it, rather than that it’s been studied and found wanting ^[42]. Meanwhile the one serious attempt at bringing motor learning into Pilates argued for using those principles to improve the method, with feedback, contextual interference and transfer built in deliberately, rather than for abandoning the whole thing ^[43]. Which is more or less what we’d say too.

What follows is our own position rather than anybody’s finding, and we want to be nice about this, because it isn’t a dig. Over a lot of years, we have yet to meet a Pilates teacher whose training covered motor learning in enough depth to teach with it, and that’s a comment on what the qualification contains rather than on the teachers, who are generally excellent at the thing they were actually taught to do.

There’s also a structural problem in there that genuinely isn’t anybody’s fault. If your class is full of people in pain, whose joints sublux, the teacher is going to go slow. Very slow. They’ll keep the range conservative, keep the tempo even, keep the setup predictable, and they are right to do all of it, because that is a sensible response to the room they are standing in and nobody wants to be the reason somebody’s hip went out. It just also happens to remove almost every condition under which an adaptable movement strategy gets built (a bit awkward).

So, our read is that Pilates is often very good at getting somebody out of the rut, the one where nothing feels safe and nothing gets moved, and the whole body has become a list of things to avoid. Getting out of that rut matters enormously, and if that’s where you are, it’s a reasonable place to start. It’s the step after that, building a pattern that still holds when you’re exhausted, on gravel, carrying a toddler and reaching behind you for a car door, where the format starts to run out of road. That’s our clinical read rather than a finding, and the direct evidence above doesn’t test it either way.

If Pilates leaves you calmer, more aware of your body and more willing to move, that’s a real gain and you should keep going. Just don’t expect it to be the whole programme, and if you’ve been at it for three years and the joint still goes, more of it harder is unlikely to be the thing that changes it.

And also the more important bit. If you enjoy, do it. For so many people rehab becomes this big monster that their focusing their attention on everyday…sometimes its just nice to do something you enjoy.

The Core Matters, and “Engage It” Is Still the Wrong Instruction

The core genuinely does matter in hypermobility, arguably more than in most groups. What’s gone wrong is the instruction, not the target.

Most of “engage your core” grew out of a finding from the nineties, that the deep abdominal wall switched on late in those with low back pain during quick arm and leg movements ^[44]. It was a good observation. What got built on top of it though, an entire industry of isolating one muscle and holding it on before you’re allowed to move, went a long way past what that observation could actually carry. Later work showed that muscle fires direction specifically and asymmetrically rather than tightening evenly like a corset ^[45][46], and the argument that the assumptions stacked on top of the original finding were overstated has been made at length, and has largely stood up ^[47].

There’s an overcorrection available here though, and we’re not taking it, because it’s just as wrong in the other direction. Core training does sometimes improve the activation and timing of that deep abdominal wall, so the feed forward story hasn’t collapsed entirely ^[48]. And going the other way, symptoms have improved in people whose abdominal onset timing didn’t change at all, which is awkward for anyone claiming the timing is the mechanism ^[49]. Core stability work generally beats doing very little, and generally doesn’t beat other kinds of exercise, which is the least exciting result imaginable and is probably the true one ^[50]. The diaphragm, meanwhile, contributes to trunk stabilisation and gets badly underplayed in most core programmes ^[51].

So, what’s the actual objection to the standard advice.

The core works as a pressure system, with the diaphragm at the top, pelvic floor at the bottom, abdominal wall around the sides, and its job is managing pressure as you breathe and move. Inhale, and the diaphragm drops, the floor gives a little, the abdominals lengthen. Exhale, and the whole thing recoils. It only works as a system, and the parts of it don’t get hired and fired independently, however much a cue implies they do.

Now, drop a noisy proprioceptive signal into that system and something fairly predictable happens: the brain hedges by stiffening. It co contracts, holds several groups on at once, and substitutes mechanical rigidity for the sensory clarity it can’t get. It’s the same thing you do driving on ice, both hands locked on the wheel at ten to two, and it works, in the sense that you don’t crash, and it is also a terrible long term strategy. The pelvic floor often gets recruited into that and then stays recruited, and the diaphragm can’t descend properly into an abdomen that’s already braced, so the breath drifts up into the chest and the neck and shoulders start doing work they were never designed for.

The pelvic floor half of that is not hypothetical either, at least not the burden. Pelvic floor symptoms in women with EDS are extremely common, with stress incontinence, urgency, prolapse and pelvic pain all reported at rates nobody should be calling rare, and pelvic pain running at around seven in ten ^[52]. Greater pelvic floor distress and worse quality of life in EDS than in others, shows up again in more recent comparison work ^[53].

What is not established though, is the mechanism we’ve just described. Whether the dominant problem is chronic overactivity, rather than weakness or the tissue itself, hasn’t been settled, and what exists suggests pelvic floor dysfunction can contribute without showing that overactivity is the main driver across hEDS ^[54]. So, the burden is a fact, and the over bracing explanation is a working model, ours and a fairly common one, and not proven.

Either way, the instruction lands badly. Telling a system that is already holding on too tight to hold on harder, brace before you move, pull the navel in, squeeze for ten, is a bit like asking somebody with a clenched jaw to bite down. The diaphragm gets less room, the floor gets less ability to lengthen, the breath gets shallower, and the exact strategy that was causing the problem in the first place gets rehearsed another thirty times. Then people walk away thinking they failed the exercise, which they didn’t.

What We’d Actually Change

None of this is an argument for doing less. It’s just an argument for a different shape of work, and a harder one at that, as it asks for attention rather than just sweating it out.

– Train actions, not muscles: The nervous system does not have a folder labelled with a muscle name and a button to press. It has reach, step, carry, stand up, and catch yourself, and those are the units it actually works in. A programme that reads as a list of muscles is organised by anatomy rather than by how movement is assembled, and in a hypermobile body that tends to mean a lot of work for not much carry over into the reaching and the carrying you actually do.

– Vary the conditions, not the exercise list: This is the one nearly everybody gets backwards. Variation doesn’t mean a new exercise every week, it means the same fundamental action performed at different speeds, through different ranges, from different starting positions, on different surfaces, with your attention in different places. Keep it deliberate and keep it moderate though, because more variation is not automatically better learning ^[31].

– Load it, and stop being frightened of the weight: Where a joint is symptomatic and the work is supervised, heavier training has been tolerated and has helped, so a permanent diet of pink dumbbells is not the cautious choice it looks like ^[9][14]. Progress it by how you respond rather than by a number written on a plan six weeks ago, and expect the benefit to show up in function before it shows up in pain ^[11].

– Cue outwards for balance and control: Point the attention at the outcome, or at something in the environment, rather than at the muscle, as that’s the direction which improved postural stability and balance in hypermobile people specifically ^[36]. For producing force, a clear instruction of either kind beat no instruction at all, so don’t mistake vagueness for freedom ^[36].

– Use sensory adjuncts, with modest expectations: Compressive garments and insoles improved postural control in hEDS in early work, particularly once vision was removed, which is exactly when you’d expect a sensory aid to earn its keep ^[55]. Taping has been described improving gait mechanics in hEDS at the level of individual cases, which is a hint rather than a finding ^[56], and in the broader taping literature the direction you apply it in changes what happens to sensation and control, so the detail genuinely matters ^[57]. Treat all of it as a way of making the signal easier to read while you train, not as a treatment in itself.

– Pace it around the autonomic side: If dysautonomia or POTS is in the picture, slower progression and individualised adaptation is what the available guidance supports, and it is not an optional extra ^[58]. A programme that ignores what your heart rate is doing when you stand up is a programme you’ll bail from by week three.

– Give it months, and expect the first stretch to look like nothing: Learning a proprioceptive skill has its own arc, acquisition, then retention, then transfer, and the transfer part is both the slowest bit, and the entire point of the exercise ^[59]. The first six to eight weeks of a well built programme often look completely uneventful from the outside, because what’s changing is the quality of the information rather than the size of anything. This is the stage where most people quit, and it’s the stage Berrie was in for ten weeks.

If that list looks like a lot to organise on your own, it is, and that’s genuinely the hardest part of doing this without somebody watching you do it. We teach the whole thing, the sensory work, the tone side, the motor learning and the loading, in the Hypermobility Fundamentals Bundle, and we go through it live with people at the workshop linked at the bottom of this page. But the principles above are the principles regardless of who you do them with, so take them to whoever you’re working with.

What Nobody Has Settled Yet

The biggest open question is the one you’d most want answered, and it’s which combination of strengthening, sensorimotor work, cueing style, pacing and adjunct support works best, at what dose, for which group of those with hypermobility. The available work is too small and too different from itself to settle it, and everybody working in the area says so ^[2][4][15].

What can be said is that one dimensional advice is the thing to leave behind. By that we don’t mean strength training, which earns its place and then some, and we don’t mean Pilates, which helps a lot of people and isn’t the whole answer for anyone. The thing to leave behind is the idea that any single ingredient, applied hard enough for long enough, will make a hypermobile joint reliable, when the problem is made of tissue, information, timing, fatigue, pain and what you believe your body will tolerate, all at the same time.

Berrie didn’t get better because we found the right exercise. She got better because we spent ten weeks giving her nervous system something clearer to work with, and then, once it had that, we loaded her properly and taught her a pattern rather than a routine.

Part two picks up the sensory side of it in detail: how a joint gets information to the brain in the first place, what that information is actually made of, and what happens to it when the tissue holding the sensors is more compliant than average.

The Fibro Guy

References

[1] Reychler, G., De Backer, M., Piraux, E., Poncin, W. and Caty, G. (2021) ‘Physical therapy treatment of hypermobile Ehlers–Danlos syndrome: A systematic review’, American Journal of Medical Genetics Part A. https://doi.org/10.1002/ajmg.a.62393

[2] Buryk-Iggers, S., Mittal, N., Santa Mina, D., Adams, S.C., Englesakis, M., Rachinsky, M. et al. (2022) ‘Exercise and Rehabilitation in People With Ehlers-Danlos Syndrome: A Systematic Review’, Archives of Rehabilitation Research and Clinical Translation. https://doi.org/10.1016/j.arrct.2022.100189

[3] Palmer, S., Davey, I., Oliver, L., Preece, A., Sowerby, L. and House, S. (2020) ‘The effectiveness of conservative interventions for the management of syndromic hypermobility: a systematic literature review’, Clinical Rheumatology. https://doi.org/10.1007/s10067-020-05284-0