What can be mistaken for hypermobility and ehlers Danlos syndrome

• 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

Joint hypermobility is a trait rather than a diagnosis, and pretty much everything that goes wrong in this area starts with somebody forgetting that ^[1][2][3]. Being bendy is a finding on an examination in the same way that being tall is a finding, as it tells you something about the tissue, it doesn’t tell you what’s causing the pain, and it very much doesn’t tell you that nothing else is going on. Hypermobile Ehlers-Danlos syndrome in particular is a clinical diagnosis that depends on other conditions having been actively ruled out first ^[4][5], which means the diagnosis is only ever as good as the ruling out that actually got done on the day.

The conditions that get confused with hypermobility fall into four rough groups: the other heritable connective tissue disorders, inflammatory joint disease, neuromuscular conditions, and the things that share the symptoms without sharing the mechanism, like fibromyalgia and the dysautonomias ^[6][7][8]. It runs in both directions too though, as plenty of people with genuinely symptomatic hypermobility spend years being told they have something else entirely, and both misdiagnosis and long diagnostic delays are common in this population ^[9][10][11].

So, this article is the differential written out in plain English: what the criteria actually ask for, what each of the conditions on the list does differently, and where the whole process tends to fall over.

One thing to say at the front though, as none of this is written to make anybody doubt a diagnosis they’ve fought for years to get, and the far more common experience in this community is being dismissed rather than being over diagnosed. It’s written because the same few conditions keep getting folded into a hypermobility label when they needed treating in their own right, and because a couple of them carry risks that a rehab plan doesn’t address.

Hypermobility Is a Trait Before It Is Anything Else

Joint hypermobility can be asymptomatic, symptomatic, localised to a single joint, peripheral in the hands and feet, or generalised across the body ^[1][2][3]. Those are five different situations, and they get flattened into one word constantly, which is a good part of why the conversation around all of this is such a mess. Somebody who can fold their thumb to their forearm and has never had a painful day in their life, and somebody whose shoulder comes out reaching for a seatbelt, are both correctly described as hypermobile, and the word is doing almost no work in either sentence really.

It’s also common. Generalised joint hypermobility turns up frequently in the general population, and especially in children and in some groups of young adults, so a high Beighton score on its own establishes very little about what’s wrong with anybody ^[5][12][13].

A symptomatic hypermobility disorder is diagnosed when the bendiness comes paired with musculoskeletal consequences, or with broader connective tissue features, rather than when flexibility is simply present and measurable ^[14][15][16]. In adults, hypermobility spectrum disorder is the default descriptive label for clinically relevant joint hypermobility with pain, once the other acquired and single gene causes have been excluded ^[15][7].

Asymptomatic hypermobility, meanwhile, is just a body. It doesn’t need managing, it doesn’t need a programme, and it isn’t a diagnosis waiting to happen, and anybody being told their perfectly comfortable bendy knees are a ticking clock has been handed a problem they didn’t actually have ^[1][2][3]. The error runs hard in both directions in this area, as the same clinical culture that tells one person their flexibility is a serious condition tells the next person, in real pain, that flexibility is harmless and they should go home.

Which makes HSD a genuinely useful label and a slightly awkward one, as a good part of its definition is what it isn’t.

That awkwardness isn’t academic, by the way. If a label is defined partly by exclusion, then the quality of the label depends entirely on how much excluding somebody actually did before writing it down, and there’s no way of telling from the letter afterwards. Two people can walk out of two different appointments with the same three words on a piece of paper, one after a full workup and one after a five minute look at their fingers.

What the Beighton Score Is Actually For

The Beighton score is a screening tool, and a fairly blunt one, with known age effects, a heavy bias towards the upper limb, and a small number of joints in it ^[17][1][18]. It was never built to be the whole assessment, and it doesn’t behave like one either.

The practical problem is coverage: the score misses several major joints entirely, so it can undercall somebody whose clinically important laxity happens to sit outside the nine points that get tested ^[1][19].

It’s especially unreliable in children, as hypermobility is common before biological maturity and some of the hEDS features only emerge later, so a score taken at nine is answering a slightly different question than the same score taken at twenty nine ^[5].

Then there are the presentations that were never going to score at all. Peripheral HSD is hypermobility restricted to the hands and feet with secondary musculoskeletal problems on top, and localised HSD is confined to one joint or one region with symptoms related to that region ^[14][20]. Both are real, both are recognised, and both can be missed by a test that’s busy looking somewhere else.

Shoulders, hips, the jaw, ankles, hands and feet can all be properly lax and properly symptomatic in a person whose Beighton score reads as negative ^[21][1][22]. Recurrent dislocation of the jaw, for one, turns up more often in those with generalised joint hypermobility, and there isn’t a point on the Beighton score that goes anywhere near it ^[22].

There’s also the historical version, meaning somebody who was clearly bendy earlier in life and has since stiffened up, which is extremely common and which a score taken today will read as negative. Those with localised or historical hypermobility report much the same symptoms and much the same associated conditions as those meeting the criteria for hEDS or HSD outright ^[21]. So, the symptom burden doesn’t seem to be politely tracking the number on the form, which is a bit of a problem if the number on the form is what’s deciding who gets seen.

This is easily the most common version of the story we hear in the studios. Somebody arrives with pain, instability, a shoulder that subluxes on a weekly basis, a score of three or four, and a letter telling them they aren’t hypermobile enough to qualify for anything. The score was doing exactly what it was designed to do, it just got asked to do a job it was never built for.

What the hEDS Criteria Actually Ask For

The current hEDS criteria are clinical and exclusion based. They require generalised joint hypermobility, then a set of systemic, family history and musculoskeletal criteria on top of that, and then the exclusion of other heritable and acquired disorders ^[4][5].

That third part is the one that gets skipped, and it isn’t a vague instruction in the slightest. Autoimmune rheumatological disease, neuromuscular disorders, the other EDS types, Marfan syndrome, Loeys-Dietz syndrome and skeletal dysplasias such as osteogenesis imperfecta are all specifically named as things to consider and rule out ^[17][7].

Genetic testing settles a good many of those. It doesn’t settle hEDS though, as there still isn’t a confirmatory molecular test for it ^[17][23][24]. Which is the genuinely odd part of all this, as nearly everything on the differential list can be confirmed or excluded with a test, and the diagnosis people are actually being given is the one that can’t be, so it sits or falls entirely on how thoroughly its neighbours were investigated.

In practice the assessment runs through history, physical examination, Beighton scoring and the five part questionnaire about historical mobility, and then somebody decides whether the pattern in front of them looks like asymptomatic hypermobility, HSD, hEDS, or something else altogether ^[25][20][15]. That’s a reasonable process and a slow one, and misdiagnosis and diagnostic delay remain common anyway, partly because the criteria are still moving, partly because there’s no biomarker to anchor them, and partly because the presentation genuinely overlaps with a long list of other conditions ^[9][10][11].

Now, the fact that the criteria are still moving means something fairly concrete for a person holding a letter. A diagnosis made under one version of the criteria isn’t automatically the diagnosis you’d be given under the next version, and that isn’t anybody being incompetent, it’s just what happens with a condition that’s still being defined ^[9][15]. So somebody assessed a decade ago, under a different framework with a different name attached to it, has a label that was correct when it was written, and that may not map cleanly onto what’s written now.

HSD and hEDS, and the Mild and Severe Problem

HSD and hEDS overlap substantially, and the evidence is honestly mixed on whether they’re two separate disorders or two positions on one spectrum ^[14][15][19].

The formal line between them is this: hEDS needs generalised joint hypermobility plus the additional systemic, family history and musculoskeletal criteria, whereas HSD is what gets used when symptomatic hypermobility is clearly there, the full hEDS criteria aren’t met, and nothing else explains the picture better ^[18][4].

What the evidence doesn’t support is the folk version, the one where HSD is just hEDS with the volume turned down ^[15][19][26]. When the two groups have been compared directly, joint pain, allergy, subluxations, brain fog, headache, anxiety, depression, nausea and constipation all turned up commonly in both, with some symptoms reported more often in HSD and some of the tissue fragility features reported more often in hEDS ^[19]. That is not the pattern you’d expect from a mild version and a severe version of the same thing, and it’s worth knowing if you’ve been handed the HSD label and told it’s the lesser diagnosis.

Older labels add a fair bit to the mess as well, and they’re still very much in circulation, as benign joint hypermobility syndrome and joint hypermobility syndrome were the previous terms, and they overlap strongly with what we’d now call HSD, and in some of the older literature with hEDS as well ^[20][25][27]. Even then, bendiness on its own was never enough, as the old criteria required pain alongside the hypermobility, and the minor criteria sat on considerably weaker ground than the pain component did ^[16].

So, if your paperwork says BJHS, it isn’t a different condition you’ve somehow also got, it’s a retired name for roughly the same territory, written by somebody working from the guidance of the day.

What does come through consistently, across both labels, is the size of the problem people are actually reporting. Diagnostic difficulty, a long list of conditions running alongside the joint symptoms, and a set of needs that aren’t being met, all turn up across hEDS and HSD together rather than sorting themselves neatly into the supposedly severe group ^[26]. Which is a reasonable argument that the line between the two labels matters a great deal more to the paperwork than it does to the person living behind it.

Marfan Syndrome

Marfan syndrome overlaps with hypermobility disorders through a long list of shared features: tall stature, long limbs, long fingers, a pectus deformity, scoliosis, flat feet, joint laxity, mitral valve prolapse and aortic root disease ^[28][29][30]. Read that list without the label attached and it could sit under either heading. Which is exactly the problem.

What separates it is where the diagnostic weight sits, rather than how many features get ticked, as the revised Ghent nosology (a nosology being a classification system, in case anybody else was wondering) puts most of the weight on aortic root aneurysm and on ectopia lentis, which is dislocation of the lens of the eye, with FBN1 testing to sort out the uncertain cases ^[28][29][30]. That revision was written partly to stop people being labelled with Marfan prematurely, and relabelling under it isn’t rare, as a meaningful minority of those already known to carry an FBN1 variant ended up with a different diagnosis under the newer nosology than the older criteria had given them ^[29][31].

Now, for anyone reading this with a hypermobility diagnosis, the practical version is short. A dilated aortic root or a lens sitting where it shouldn’t be isn’t a hypermobility spectrum question any more, and it wants a cardiologist and a geneticist rather than a rehab plan.

The opposite also needs saying, as this one generates a fair bit of unnecessary fear. Being tall, long limbed and bendy is not Marfan syndrome, and it doesn’t become Marfan syndrome because somebody on the internet noticed your fingers. The diagnostic weight sits on the eye and the aorta for a reason, and those are things that get measured rather than eyeballed ^[28][29][30].

Loeys-Dietz Syndrome

Loeys-Dietz is the one that gets mistaken for both of the others. It shares aneurysm risk and skeletal findings with Marfan, and it shares easy bruising and sometimes soft, velvety or stretchy skin with EDS, which is precisely the combination that lands somebody with one of the other two labels ^[28][32][33].

Now, the clues that point towards it instead are widely spaced eyes, a bifid uvula or a cleft palate, and arterial aneurysm or tortuosity in vessels beyond the aortic root ^[28][32][33]. Ectopia lentis generally isn’t part of the picture, which is one of the things separating it from Marfan ^[28][32][33].

Reassignment genuinely happens too. People first labelled as having Marfan syndrome or EDS have later been moved to a Loeys-Dietz diagnosis once broader genetic testing was done, particularly where a TGF beta pathway variant turned up, or where arterial findings appeared somewhere other than the aortic root ^[34][35][33].

This one matters more than most of the list, as the vascular risk is real and the management that follows from it is different. It’s also a reasonable argument for broader genetic testing rather than a single targeted test, in anybody whose picture has never quite fitted the label they were given.

Osteogenesis Imperfecta and the COL1 Overlap

Osteogenesis imperfecta can look like a hypermobility disorder because joint laxity, soft tissue findings and mild connective tissue features show up in it too ^[36][37][38]. It’s a collagen problem, the same as the EDS group is, and the features don’t respect the boundary between the two names anywhere near as neatly as the textbooks imply. What points away from a hypermobility disorder and towards OI is a fracture history out of proportion to the injuries that caused it, blue sclerae, dentinogenesis imperfecta, low bone mass, and a COL1A1 or COL1A2 variant ^[36][37][38].

The overlap isn’t theoretical either, as there are people who were first assumed to have EDS or hEDS and who were later found to carry COL1A1 or COL1A2 variants, fitting either an OI and EDS overlap or a broader COL1 related disorder ^[39][38][40]. It runs the other way as well, as some of those being assessed under a general connective tissue disorder banner turn out to have molecularly confirmed OI ^[24].

So, if bones have broken more easily than the story really explains, that’s a bone density and genetics question rather than a Beighton question, and it’s one of the few things on this list where getting the answer changes what happens next in a fairly urgent way.

The COL1 related overlap disorders are worth knowing about as a category in their own right, rather than as a footnote to either condition. They describe people who genuinely sit between the two, with bone fragility and connective tissue features together, and they exist as a named group precisely because a proportion of people were never going to fit cleanly into either box ^[40][38]. Which is a considerably more honest answer than picking whichever label the person happened to be sent to a specialist for.

Inflammatory Joint Disease

Inflammatory arthritis can be mistaken for hypermobility related pain, and, more awkwardly, the two can sit perfectly happily in the same person ^[41][17].

What should stop a painful joint being quietly filed under HSD or hEDS is inflammation itself: swelling, warmth, synovitis, pain with a genuinely inflammatory pattern to it, morning stiffness that lasts rather than easing in ten minutes, or erosive change turning up on imaging ^[41][17]. None of that is a hypermobility finding, and folding it into a hypermobility diagnosis is how somebody ends up doing careful joint control work, and doing it properly and consistently, for a problem that needed treating medically.

The criteria are explicit about this in adults, as where somebody has an acquired connective tissue disease such as rheumatoid arthritis, hEDS can only be added on top if the non inflammatory features still satisfy the criteria on their own, and pain and instability caused by the inflammatory disease don’t count towards the musculoskeletal criterion ^[17].

In children the same logic runs in both directions. Juvenile idiopathic arthritis is itself a diagnosis of exclusion in a child with arthritis lasting more than six weeks, and imaging is often needed because physical examination isn’t reliable enough on its own for detecting active inflammation, particularly at the jaw, the spine, the sacroiliac joints, the wrists, the hips and the knees ^[42]. And children first treated as having JIA have later turned out to have non inflammatory or single gene conditions instead, especially where the inflammatory markers were normal, the imaging looked atypical, or nothing responded to the usual treatment ^[43].

Which is a fairly good illustration of why any exclusion based diagnosis is only ever as good as the person doing the excluding, in either direction.

And because the two can coexist, this isn’t a case of one answer cancelling the other out ^[41][17]. Somebody can have generalised joint hypermobility, meet the criteria for a hypermobility disorder on the non inflammatory features, and have inflammatory arthritis on top of it that needs treating medically. Deciding it must be one or the other is how one of them goes untreated for years, and, in our experience, it’s usually the inflammatory half that gets missed, because the bendiness was noticed first and it explained enough of the picture to stop anybody looking any further.

Neuromuscular Conditions

This is the group that gets missed most often, in our experience, and it’s the one with the least public awareness attached to it by a long way.

Inherited myopathies, both the congenital ones and the adult onset ones, can present with joint hypermobility alongside weakness that’s only mild to moderate ^[44][25]. From across the room that looks a great deal like a hypermobility disorder, and the bendiness is usually the first thing anybody comments on, so it becomes the headline, and by the time anybody has finished talking about the elbows and the thumbs and how far the fingers bend back, the muscle question never really gets asked at all.

Bethlem myopathy gets named specifically in the hEDS exclusions, which tells you how seriously this part of the differential is meant to be taken ^[7][17]. The flag is a picture where low tone, weakness, delayed motor development or a myopathic pattern is more prominent than any of the skin or tissue features, and that’s the point at which somebody should be looking at muscle rather than at collagen ^[7][17][45].

The childhood version of this is where it does the most damage, as low tone and delayed motor milestones in a bendy child get put down to the bendiness and then nobody revisits it. It’s also where the old “benign” framing came from and where it started falling apart, as joint hypermobility presenting in childhood has been recognised for a long time as carrying multisystem features rather than being a purely orthopaedic curiosity ^[45]. Weakness that’s out of proportion to the laxity is the thing to hold on to though. Loose joints on their own don’t make a child weak, so if the weakness is the loudest feature, it wants explaining rather than absorbing.

Fibromyalgia, POTS and the Overlap Conditions

Fibromyalgia overlaps with hEDS and HSD heavily, in pain, in fatigue, in autonomic symptoms, in neurological complaints and in the effect all of it has on daily life, and holding both diagnoses at once is common rather than unusual ^[46][47]. So, the honest position is that fibromyalgia is sometimes the wrong label sitting on a hypermobility disorder, sometimes the right label sitting alongside one, and sometimes simply the right label on its own ^[46][47]. What tilts it is whether there’s genuine joint laxity and instability on examination, which is a question nobody can answer from a symptom questionnaire, and it’s the question that tends not to get asked.

POTS sits in a similar spot, as it’s commonly reported in hEDS and HSD groups, and the autonomic symptom burden carried by women with these diagnoses is substantial ^[48].

The proposed triad of hEDS, POTS and mast cell activation syndrome is a different matter though, and it remains controversial, as the strict evidence tying all three together is limited and inconsistent ^[48][49][50]. None of which is us saying the symptoms aren’t there, as they very clearly are, in a great many people. It’s that three conditions turning up in the same group of people is not the same thing as one of them causing the others, and nobody has shown that yet.

Practically, the overlap conditions are less about misdiagnosis and more about premature closure. Once a label is on the notes that explains fatigue, widespread pain and dizziness, everything afterwards gets read through it, and the joint examination that would have changed the answer never gets done ^[46][47]. So, carrying one of these diagnoses isn’t evidence against the others, and a lot of people legitimately hold more than one at once ^[46][47][48].

Working Out Which One It Is

The literature supports a stepwise approach, and it’s a far more useful way to think about your own paperwork than trying to match yourself against a symptom list on the internet. It runs in the same order every time, which is the useful part, as each step narrows what the next one has to consider.

– How old were you when it was measured: Hypermobility is common in children and some of the hEDS features only show up later, so a score taken in childhood and a score taken in adulthood are not interchangeable, and neither is a clean answer on its own ^[5].

– What kind of mobility is it: Generalised, localised to one joint, peripheral in the hands and feet, or historical, meaning it was clearly there once and has since stiffened up. These are genuinely different situations with different implications, and the Beighton score only really speaks to the first of them ^[5][14][15].

– Are the symptoms musculoskeletal only, or broader than that: Pain and instability in joints is one picture. Pain plus skin, gut, autonomic and family history features is a different one, and that split is what the hEDS criteria are built around ^[14][15].

– What has actually been excluded: Inflammatory disease, neuromuscular conditions, skeletal dysplasias and the other heritable connective tissue disorders, in that spirit at least. This is the step that determines whether the label means anything, and it’s the step most likely to have been skipped ^[5][14][15].

– What genetic testing can and can’t settle: It’s central for most EDS subtypes, for Marfan and the Marfan related disorders, for Loeys-Dietz, for osteogenesis imperfecta and for the COL1 related overlap disorders. It cannot confirm hEDS, as there’s still no molecular test for it ^[17][23][24].

If you’re going back to a doctor with this, the useful question isn’t whether you’ve got hEDS or HSD, as they can’t settle that one with anything other than an opinion. The useful question is which of the testable conditions have been excluded and how, and that’s a question with an actual answer attached to it.

What Nobody Knows Yet

The biggest open question in this whole area is whether HSD and hEDS are genuinely two different disorders or two positions on one still poorly defined spectrum, and it is not settled ^[15][19]. Anybody telling you confidently which way round it is has gone beyond the evidence.

There’s no biomarker either, which is the thing sitting underneath most of the rest of it. Work has been going on to find something measurable in blood that would separate these groups, and a shared pattern of extracellular matrix fragmentation has been put forward as a candidate ^[18]. That’s genuinely interesting, and it’s still a long way from being a test that anybody can order.

And the diagnostic delay is well documented without being at all well explained. It’s clear that it happens, and clear that it happens more in this population than in most, which unfortunately doesn’t tell anybody what to change about the appointment itself ^[10][9].

The Fibro Guy

References

[1] Malek, S., Reinhold, E.J. and Pearce, G.S. (2021) ‘The Beighton Score as a measure of generalised joint hypermobility’, Rheumatology International. https://doi.org/10.1007/s00296-021-04832-4

[2] Tinkle, B.T. and Levy, H.P. (2019) ‘Symptomatic Joint Hypermobility’, Medical Clinics of North America. https://doi.org/10.1016/j.mcna.2019.08.002

[3] Tinkle, B.T. (2020) ‘Symptomatic joint hypermobility’, Best Practice & Research Clinical Rheumatology. https://doi.org/10.1016/j.berh.2020.101508