Is Hypermobility Linked to Autism?

About
Latest Posts

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.

Key Takeaways

Autistic people are more likely to be hypermobile than the general population, but the headline 51% figure came from a specialist sample. The pooled estimate from a 2025 meta-analysis of 20 studies sits at roughly 22% across all methods, and around 31% when hypermobility is properly clinically assessed.
People with EDS show substantially higher rates of autism, ADHD and other psychiatric diagnoses compared with matched controls, and unaffected siblings show some of that elevation too, which points at shared biology.
The first hEDS genome wide association study, posted as a preprint in 2025, reported significant genetic correlations between hEDS and autism. That’s the first direct genomic signal alongside the long-standing population-level evidence, but the paper is still in preprint and needs peer review and replication.
Dysautonomia, POTS in particular, sits underneath a huge chunk of the overlap. Around 70% of people with hEDS report it, and it’s well documented in autistic adults too.
If both turn up together, your exercise tolerance, sensory needs and pacing requirements are more involved than either condition on its own would predict, and that has practical consequences for how you train.

What the Research Actually Shows

The number you’ll see chucked around the most is that roughly half of neurodivergent adults are hypermobile. That one comes from a study out of Brighton and Sussex Medical School, where they compared 109 adults who all had confirmed diagnoses, autism, ADHD or Tourette syndrome, against a comparison group^[1]. Using the bog standard Beighton criteria, 51% of the neurodivergent lot were hypermobile, versus 17.5% in that comparison group sitting in the same study. And when you set the neurodivergent figure next to general population data from the ALSPAC birth cohort, which sits around 20%, the gap’s still big^[1]. Now, it’s worth keeping those two comparison numbers apart in your head, because people muddle them all the time. The 17.5% is the study’s own internal comparison group. The 20% is a much larger, separate population reference. They land close to each other, which is partly why they get blurred, but they aren’t the same measurement and they aren’t doing the same job.

The same Brighton paper also looked at how strong that association actually was, not just whether it existed. The headline odds ratio they reported was around four and a half, meaning the neurodivergent group as a whole came out roughly four to five times more likely to be hypermobile than the general population reference^[1]. The autism subgroup on its own moved in the same direction, broadly consistent with that overall figure rather than separate from it^[1]. A different study, this one with 618 adults pitting 199 people with an autism diagnosis against 419 community controls, found much the same shape, and when they looked specifically at symptomatic hypermobility rather than bendiness on its own, the gap got wider still^[4]. The symptomatic part is the bit I’d pay attention to. It tells you something real is going on, not just that autistic people happen to be a touch more flexible at parties.

Here’s where the picture has actually shifted since 2023 though. A 2025 systematic review and meta-analysis pulled together every reasonable study on this question, 20 of them in total, and ran proper pooled estimates^[11]. When you bring the lot together, the overall prevalence of joint hypermobility in autistic samples lands at roughly 22%, climbing to 31% when you only count the studies that used proper clinical assessment rather than self report^[11]. For HSD and EDS specifically, the pooled prevalence in autistic samples comes out at around 28%, rising to 39% with clinical assessment^[11]. So the association absolutely holds, 12 out of 15 individual studies looking at it found something significant, but the population level figure is meaningfully lower than the 51% that’s been getting passed around. The Brighton 51% wasn’t wrong, it just reflects a specialist sample of people who’d already been recognised as neurodivergent. The Baeza-Velasco numbers are closer to what you’d find in the wider autistic population. Both can be true at the same time.

But, before you go sprinting off with those figures, there’s a caveat on that second study that you really do need. 69% of the autistic participants in that cohort also had ADHD, and when the researchers stripped the ADHD out, the association only held for symptomatic hypermobility, not for flexibility alone^[4]. So a sceptic, and there’s always one, could quite reasonably argue that some of what looks like an autism and hypermobility link is partly an ADHD and hypermobility link wearing a different jumper. That doesn’t make it nothing. It just means the clean tidy version of the story is messier underneath than it first looks, which, fair warning, is a theme that runs through this entire topic from top to bottom.

There’s also a useful set of figures from the other direction. A 2025 Mayo Clinic study pulling on 2,695 people from their specialist EDS clinic looked at how often self reported autism showed up in their patients^[13]. In controls, 5.3% reported autism. In the hEDS group it was 7.3%, in the broader HSD group 7.4%, and in a smaller group with localised or historical hypermobility patterns it climbed to 13.8%^[13]. The differences between that last group and the rest were statistically significant^[13]. Those are self report figures from a specialist clinic, so they’ll be inflated relative to the community, but the pattern is consistent. Whichever way round you slice it, EDS-clinic to autism or autism-clinic to hypermobility, the rates come out above the base population.

Here’s the other big caveat, and it’s a proper one. The people sampled in most of these studies are not community samples, not by a long way. When it comes to recruitment, they tend to get pulled in from specialist referral centres, or through support groups, which means the sample skews towards people whose conditions are more severe, or more recognised, or both. The 2025 meta-analysis tried to control for some of this by separating clinical assessment from self report, which is part of why its pooled figures are lower than the headline 51%^[11]. The direction of effect looks solid. The exact size of it out in the actual world is still genuinely uncertain, and anyone telling you they know it to the decimal point is selling you something.

The Swedish data is worth knowing about too, mainly because it sidesteps a lot of that specialist sampling problem. A nationwide population register study matched 1,771 people with Ehlers-Danlos syndrome against 17,710 comparison individuals, all of it pulled from the national registries^[3]. The Cederlöf paper reported a risk ratio for autism of 7.4 in the EDS group, which is the statistical effect size you’d quote in any psychiatric epidemiology paper. That’s a lot, however you want to dress it up. And the same paper found elevated rates of ADHD, bipolar disorder and depression in the EDS group as well^[3]. But the bit I find genuinely telling, more than any single headline number, is the sibling data. Unaffected siblings of people with EDS, the ones who didn’t have EDS themselves, still showed significant elevations in ADHD, depression and suicide attempt rates^[3]. The autism signal in siblings was clearest in the broader hypermobility-syndrome subgroup rather than reaching statistical significance in EDS-specific siblings on its own, but the overall pattern still points towards shared genetic and early environmental factors rather than the whole thing being down to the lived experience of having a chronic condition^[3]. Which is a different claim entirely, and a more interesting one.

So that’s the association, more or less. Well documented, holding up across different study designs and more than one country, and now backed by a 2025 meta-analysis pooling 20 studies, which is about as good as this kind of evidence tends to get. Two conditions that look completely unrelated on paper, one of them defined by loose joints and dodgy connective tissue, the other by differences in how the brain handles information and social stuff, and yet they keep turning up together at rates well above what chance would give you. Which lands you on the obvious question. Why?

Why the Connection Makes Biological Sense

The connective tissue in your body isn’t just the stuff holding your joints together, even though that’s how most people picture it. It wraps around your blood vessels, it sheaths your nerves, it lines the gut wall, and it forms part of the structure that sensory signals have to travel through on their way up from the body to the brain. So when that tissue has a different quality to it, stretchier, more compliant, less structurally precise, the signals it’s carrying come through noisier^[8]. And noise, really, is the key word for this whole section.

This is where proprioception comes into it. Proprioception is your brain’s sense of where your body is in space, what position your joints are sitting in, how much force your muscles are putting out, all of that, mostly without you ever thinking about it. The brain runs on prediction. It keeps a running model of where your body is and updates that model constantly, based on the signals coming back in. So when those signals are noisy, because the connective tissue carrying them is hypermobile, the brain has to work a good bit harder to resolve the uncertainty, and it doesn’t always get it right^[2].

There’s a model that tries to tie this together. The mismatch between what the brain expects and what the body actually reports back, the proprioceptive surprise if you want to call it that, seems to create a measurable pathway running from neurodivergence through to emotional dysregulation, with hypermobility acting like the switch that turns the path on^[2]. In a study of 182 people, that mismatch correlated strongly with both hypermobility scores and neurodivergent characteristics, and the statistical relationships held up across more than one way of modelling it^[2]. It’s properly interesting work. I do want to be honest about what it actually was though. These were people with complex chronic conditions, screened for neurodivergent traits rather than formally diagnosed, and the sample skewed heavily white and female^[2]. So it’s a strong signal pointing somewhere useful, not the last word on anything.

Some earlier brain imaging work came at the same patch of ground from a different angle. Hypermobile people showed heightened reactivity in the insular cortex, which is the brain’s main hub for processing the body’s own internal signals, when they were shown emotionally charged scenes^[9]. Their insulas were firing harder, not quieter, and interoceptive sensitivity statistically mediated the link between hypermobility and anxiety in that one^[9]. Now, the fMRI subsample was tiny, just 19 people, so this is a mechanism candidate rather than a confirmed signal, and I’d hold it loosely. Still, it does give you something concrete to picture sitting underneath all the observational data.

The broad picture you get from this line of research is that connective tissue variants like hypermobility end up creating a body that’s harder to read from the inside. The signals are there, they exist, they’re just ambiguous. And in a neurodivergent nervous system that already processes sensory information a bit differently to start with, that ambiguity stacks on top of itself. If you want the interoception piece spelled out in more detail, our alexithymia post goes through the mechanisms properly, and there’s a fair amount of overlap between the two topics anyway.

On the genetics, this is the bit that’s actually moved the most since the original version of this post. A 2020 review concluded that EDS, hypermobility spectrum disorder and autism share similarities at the molecular and cellular level, including overlapping patterns of autonomic and immune dysregulation^[6]. The sibling data from the Swedish study hinted at shared heritable factors^[3]. And then in late 2025 the first proper hEDS genome wide association study went up as a preprint, pulling 1,815 hEDS cases against 5,008 ancestry matched controls^[12]. Two loci hit genome wide significance, including one near ACKR3, a chemokine receptor with expression effects in tibial nerve tissue and a known role in neuroimmune and pain signalling pathways^[12]. Gene based analyses flagged PSMC3, which is involved in central nervous system development^[12]. And, crucially for this post, the same analysis ran something called LD-score regression to check for shared genetic architecture across conditions, and it found significant genetic correlations between hEDS and autism^[12].

That’s a meaningful step. Up to that point, the genetic link was inferred from the population co-occurrence and the sibling effect. Now there’s a direct genomic signal saying these conditions share some of the same common variants. The caveat, and it matters, is that the paper is still a preprint as I’m writing this, so it hasn’t been through peer review yet^[12]. The sample is also modest by genome wide study standards. So treat it as a lead, a strong one, but not yet a settled finding. Either way, the days of the genetic story being purely speculative are essentially over.

The Sensory Processing Layer

One of the better known features of autism is sensory processing differences. Sounds that are too loud, certain textures that are physically unbearable, fluorescent lighting that genuinely hurts, that sort of thing. And this isn’t some quirky add on, by the way, it’s one of the actual diagnostic criteria for autism, the sensory hypersensitivity or hyposensitivity item^[6]. Which maps pretty naturally onto what we already know goes on in hypermobility.

Hypermobile joints send imprecise proprioceptive signals. That imprecision means the brain’s map of the body is less accurate than it should be, and that can come out as clumsiness, postural instability, difficulty judging force and position, all of it. Autistic people report exactly this kind of motor difficulty a lot, and studies in children with hereditary connective tissue disorders find around 30% of them have developmental coordination disorder, which is the clinical name for what most people just call dyspraxia^[10]. Going the other way round, roughly 17% of children formally diagnosed with DCD turn out to have generalised joint hypermobility^[10]. Small samples though, about 23 children per group, and other cohorts report higher rates than that, so the precise numbers are unsettled and I wouldn’t hang my hat on the exact figure. The direction is the point, not the decimal place.

What you’ll often see in practice is a person who’s hypersensitive to external input, the lights and sounds and touch, while at the very same time struggling to read their own internal signals, things like hunger, where the pain actually is, where their limbs are. On the face of it that sounds contradictory, but it isn’t really. It’s pretty much what you’d expect from a nervous system that’s both noisier than average in its peripheral signalling and working harder than average to make sense of the mess. The overlap between EDS, HSD and autism includes shared peripheral neuropathies and similar patterns of sensory dysregulation^[6].

None of this means hypermobility causes autism, and it doesn’t mean every autistic person has dodgy proprioception either. What it does mean is that when the two of them show up together, the sensory profile gets a good deal more complicated, and the way they interact with each other makes clinical sense rather than being some random pile up of bad luck.

POTS, Dysautonomia, and Why This Trio Keeps Showing Up

The POTS, EDS and MCAS trifecta is familiar ground for anyone who’s followed our stuff for a while. POTS, postural orthostatic tachycardia syndrome, is a form of dysautonomia, which just means the autonomic nervous system, the part of you that quietly runs your heart rate and blood pressure and temperature and gut without you ever having to think about it, isn’t doing its job properly. And it’s extremely common in hEDS and hypermobility spectrum disorder. The most recent global survey of people with hEDS and HSD, with 3,906 respondents pulled from over 9,000 screened, found that around 71% of them reported dysautonomia, with POTS as the most common subtype^[14]. That paper has now been peer reviewed and published in the Journal of Clinical Medicine, which gives the number a bit more weight than it had when it was sitting as a preprint.

What’s less well known is that dysautonomia is well documented in autism too. In a retrospective review of records from two UK national autonomic referral centres, 71% of the autistic adults who’d been referred had a formally diagnosable autonomic disorder, most commonly POTS^[7]. And of those with autonomic dysfunction in that cohort, 80% also had hypermobile Ehlers-Danlos syndrome^[7]. Now, that’s a small study with an obviously selected population, people who’d already been sent off to a specialist autonomic unit because somebody suspected something, so the exact percentages won’t generalise out to everyone. But the direction of it lines up with the rest of the picture, which is the useful part.

It’s also worth flagging that the prevalence figures bounce around a lot depending on which diagnostic criteria you use, and a 2025 study of POTS in young patients showed this quite cleanly. Looking at one hundred young people meeting POTS criteria, joint hypermobility disorder turned up in anywhere from 13% to 34% of them, depending on whether they used the strict 2017 hEDS criteria or the broader HSD definition^[15]. That’s a big spread, and it’s a reminder that the numbers in this area are always partly a function of which set of rules you happened to use to count them.

The mechanism loops back to the connective tissue argument again. Connective tissue surrounds the blood vessels. So when it’s more compliant than it should be, the veins in the lower body pool blood more easily, and that’s one of the main things driving POTS in the first place. Add in the dysregulated sympathetic nervous system responses you see across both autism and hEDS, and what you’ve got is a body that’s working overtime just to keep its own basic plumbing in order, on top of everything else it’s juggling^[1][7].

For those with this particular combination, the knock on for exercise tolerance is real and it matters. Upright activity, and especially anything that has you standing still for a stretch, gets harder. The brain’s busy managing autonomic instability while at the same time managing sensory input that’s already more difficult to process than usual. Which brings us, finally, to the question most people in this community actually wanted answered when they clicked on this.

ADHD, Dyspraxia, and the Middle of the Venn Diagram

A word on ADHD, because it keeps turning up in the same conversation and it’s earned its own paragraph by now. In the larger studies of hypermobility in neurodivergent adults, roughly half the neurodivergent group had ADHD as either a primary or a comorbid diagnosis^[1]. Swedish paediatric data puts the figure at 46% of 17 to 18 year olds with HSD or hEDS combined carrying confirmed ADHD diagnoses^[5]. And the Cederlöf paper reported a risk ratio for ADHD of 5.6 in the EDS population, which is nearly as big as the autism figure^[3]. So, not a small effect by anyone’s reckoning. The 2025 Mayo Clinic data showed ADHD running at elevated rates across all the hypermobility groups too, in line with the rest of the picture^[13].

The ADHD and hypermobility link is probably running on much the same mechanisms as the autism one. Proprioceptive noise, autonomic dysregulation, the general sensory processing differences that sit underneath both diagnoses. And ADHD and autism overlap heavily anyway, in both their genetics and how they present, so it’s no great shock that both turn up in hypermobility at rates well above the population base rate.

Dyspraxia, developmental coordination disorder, sits right bang in the middle of all this. It’s a neurodevelopmental condition marked by poor motor coordination and trouble learning motor skills, and it’s a lot more common in autistic people than in the general population, and a lot more common again in people with hypermobility. Whether it’s a separate thing in its own right, or a knock on of proprioceptive imprecision, or just another way the same underlying neurology shows itself, is genuinely unresolved^[10]. What is clear, and this is the part that actually matters in practice, is that when somebody has both autism and hypermobility, motor coordination difficulties of this kind are common enough that you should probably be expecting them rather than getting caught out by them.

What This Means in Practice

Right. So what do you actually do with all of this?

If you’re autistic and hypermobile, or you reckon you might be one or the other, or both, there are a few things worth getting straight in your head. Firstly, the sensory difficulties you get around movement, in exercise settings, in response to touch and manual work, are not imaginary, and they are not you being anxious or difficult or precious about it. They reflect a nervous system that’s genuinely receiving ambiguous signals and working overtime to make sense of them. And when it comes to exercise and rehab, that changes the approach quite a bit, more than most people appreciate.

Mostly it means progression has to be slower. The usual push through and adapt model of programming just doesn’t work well when the sensory system is already maxed out before you’ve even started. Load progression needs to follow what you can actually handle on the day, rather than some external timeline scribbled on a spreadsheet six weeks ago. And pacing here isn’t only an energy thing, it’s a sensory one too, because when you’ve overdone it you’re not just physically knackered, you’re neurologically knackered, and that, in our experience, takes a good deal longer to come back from.

The proprioceptive side matters for what kind of exercise is actually worth doing, as well. Movements that lean hard on accurate joint position sense, the balance work, the closed chain loading, anything that puts precise demands on proprioception, are genuinely harder for those with hypermobility, and harder still when autistic sensory processing piles more noise on top. The answer isn’t to dodge them though, that’s the mistake people make. It’s to build them slowly, in environments that keep the external sensory overload down, with more repetition and less variation than a standard training model would normally throw at you. Motor learning principles really matter here, and there’s a lot more on that in our hypermobility hub.

And if dysautonomia and POTS are in the picture too, then upright cardiovascular work needs bringing in carefully, with a proper eye on how the body responds to changes in position. Starting your early aerobic conditioning lying down, or reclined, isn’t a workaround and it isn’t a cop out, it’s just a sensible reality for a lot of people, and there’s no shame in it.

Second thing. The diagnostic picture in this group is usually a bit of a mess, and you should go in expecting that. A lot of autistic people, particularly those assigned female at birth, got missed or misdiagnosed for years on end. A lot of people with hEDS racked up diagnoses of anxiety, depression and functional illness long before anyone could be bothered to actually look at their joints. The most recent global hEDS and HSD survey, now peer reviewed and published in the Journal of Clinical Medicine, put the average diagnostic delay at around twenty two years^[14], which is grim reading but probably won’t surprise anyone reading this. And when both conditions are present at once, symptoms from one can mask or amplify symptoms from the other, in ways that are genuinely hard to pull apart even for people who do this for a living. That’s not a reason to throw your hands up and give in. It’s a reason to walk into any assessment with that complexity already in mind, and to not assume that sorting one condition out will magically clear up everything that actually belongs to the other.

The Bottom Line

The connection between hypermobility and autism is well evidenced at the population level. Autistic people are noticeably more likely to be hypermobile than the general population, and people with EDS show substantially higher rates of autism and ADHD compared with matched controls. And that holds across different study designs and different countries, and as of 2025 it’s now backed by a proper meta-analysis^[11], which is about as good as this sort of evidence ever gets.

The mechanisms, on the other hand, are emerging rather than nailed down. The most compelling model has proprioceptive imprecision, coming off the back of hypermobile connective tissue, feeding into predictive processing problems in a neurodivergent nervous system, which then spills out as emotional dysregulation, sensory difficulties and autonomic instability^[2]^[8]. The genetic link, which used to be statistical inference from co-occurrence data, now has its first direct genomic evidence in the form of a 2025 hEDS GWAS preprint reporting significant genetic correlations with autism^[12]. Still preprint, still needs replication and peer review, but the picture is firming up.

And honestly, the data are still patchier than the confident social media summaries would have you believe. Sample sizes are often modest, the populations are often specialist rather than community, and the prevalence numbers move about depending on how you measure hypermobility, who you sample, and how you define autism in the first place. The 2025 meta-analysis nailed down the headline pooled figures, but the underlying heterogeneity across studies is real^[11]. That’s where the science actually sits right now, and it’s worth knowing that, rather than being sold a certainty the evidence doesn’t quite back up.

What isn’t in any doubt is this. If you’ve got this combination, then your exercise tolerance, your sensory needs and your pacing requirements are more complicated than either condition on its own would have predicted. And that deserves proper attention, and proper adjustment, rather than being waved away as you being awkward.

FAQ

Does hypermobility cause autism?

No, and it’s worth being clear on that one. Hypermobility doesn’t cause autism. The two co-occur at higher rates than chance would predict, which points at shared biological factors, likely genetic, likely involving connective tissue and nervous system development, but neither one causes the other. The 2025 hEDS genome wide study reported significant genetic correlations between the two, which strengthens the shared biology story, but a correlation in genomic architecture is still a different thing from one condition causing the other.

What percentage of autistic people are hypermobile?

Depends who you’re counting and how. The widely shared 51% figure comes from a specialist neurodivergent sample, not the general autistic population. Pooled estimates from the 2025 meta-analysis land somewhere between a fifth and two fifths depending on how you measure it, and that’s still a meaningful elevation above base rate. Full breakdown is in the research section above.

Should I get checked for autism if I’m hypermobile?

Not on the basis of hypermobility alone, no. The co-occurrence rate is raised at the population level, but the majority of those with hypermobility are not autistic. If you recognise in yourself a consistent pattern of sensory processing differences, difficulty with social communication, a strong pull towards routine and predictability, or significant sensory hypersensitivity or hyposensitivity, then it’s worth raising with a GP or a clinical psychologist, whether or not your hypermobility is part of the picture. Autistic adults, and especially those who weren’t picked up in childhood, very often say that a late diagnosis explained a whole lot of things that had felt confusing for years.

Are autism and EDS the same condition?

No, they’re not the same condition. Autism is a neurodevelopmental condition affecting how the brain processes information, social communication and sensory input. EDS is a connective tissue disorder, with a whole range of subtypes involving defects in collagen structure and function. They’re distinct diagnoses with distinct criteria. What the research suggests is that they share some overlapping biological features, including, on current preprint evidence, some common genetic variants, and that they turn up together more often than you’d expect by chance, not that they’re two faces of the same disease.

Does autism affect joint pain?

Directly, no. But autism affects how sensory signals get processed, and that includes pain signals. The interoceptive processing differences common in autism mean some autistic people experience pain differently, sometimes reporting it less clearly, struggling to locate it or describe it, and sometimes experiencing it as more intense because of the sensory hypersensitivity. And when hypermobility is also in the mix, the pain experience gets complicated further again by proprioceptive noise and autonomic instability. The combination tends to produce pain that’s harder to communicate and harder to manage with the standard off the shelf approaches.

Can hypermobility training help if I’m autistic?

Yes, broadly, but it needs adapting. The standard hypermobility rehab principles, the proprioceptive training, the stability work, the pacing, the graduated loading, are all still the right direction to be heading in. What changes is the pace of the progression, the sensory environment you train in, and the communication style of whoever’s supporting you through it. Autistic people with hypermobility often tell us that loud, unpredictable, high stimulus exercise settings are genuinely counterproductive, and not because they’re unwilling, but because the sensory load is competing for the same attention needed to learn and execute a movement accurately. So quieter, more predictable environments, with clearer verbal and tactile cues, tend to work a lot better. This isn’t a special accommodation. It’s just a sensible adjustment to a real neurological reality.

— The Fibro Guy Team —

References

Csecs, J.L.L., Iodice, V., Rae, C.L., Brooke, A., Simmons, R., Quadt, L., Savage, G.K., Dowell, N.G., Prowse, F., Themelis, K., Mathias, C.J., Critchley, H.D. and Eccles, J.A. (2022) ‘Joint hypermobility links neurodivergence to dysautonomia and pain’, Frontiers in Psychiatry, 12, p. 786916. doi: 10.3389/fpsyt.2021.786916
Eccles, J.A., Quadt, L., Garfinkel, S.N. and Critchley, H.D. (2024) ‘A model linking emotional dysregulation in neurodivergent people to the proprioceptive impact of joint hypermobility’, Philosophical Transactions of the Royal Society B: Biological Sciences, 379(1908), p. 20230247. doi: 10.1098/rstb.2023.0247
Cederlöf, M., Larsson, H., Lichtenstein, P., Almqvist, C., Serlachius, E. and Ludvigsson, J.F. (2016) ‘Nationwide population-based cohort study of psychiatric disorders in individuals with Ehlers-Danlos syndrome or hypermobility syndrome and their siblings’, BMC Psychiatry, 16, p. 207. doi: 10.1186/s12888-016-0922-6
Glans, M.R., Thelin, N., Humble, M.B., Elwin, M. and Bejerot, S. (2022) ‘The relationship between generalised joint hypermobility and autism spectrum disorder in adults: a large, cross-sectional, case control comparison’, Frontiers in Psychiatry, 12, p. 803334. doi: 10.3389/fpsyt.2021.803334
Kindgren, E., Quiñones Perez, A. and Knez, R. (2021) ‘Prevalence of ADHD and autism spectrum disorder in children with hypermobility spectrum disorders or hypermobile Ehlers-Danlos syndrome: a retrospective study’, Neuropsychiatric Disease and Treatment, 17, pp. 379-388. doi: 10.2147/NDT.S290494
Casanova, E.L., Baeza-Velasco, C., Buchanan, C.B. and Casanova, M.F. (2020) ‘The relationship between autism and Ehlers-Danlos syndromes/hypermobility spectrum disorders’, Journal of Personalized Medicine, 10(4), p. 260. doi: 10.3390/jpm10040260
Owens, A.P., Mathias, C.J. and Iodice, V. (2021) ‘Autonomic dysfunction in autism spectrum disorder’, Frontiers in Integrative Neuroscience, 15, p. 787037. doi: 10.3389/fnint.2021.787037
Sharp, H.E.C., Critchley, H.D. and Eccles, J.A. (2021) ‘Connecting brain and body: transdiagnostic relevance of connective tissue variants to neuropsychiatric symptom expression’, World Journal of Psychiatry, 11(10), pp. 805-820. doi: 10.5498/wjp.v11.i10.805
Mallorquí-Bagué, N., Garfinkel, S.N., Engels, M., Eccles, J.A., Pailhez, G., Bulbena, A. and Critchley, H.D. (2014) ‘Neuroimaging and psychophysiological investigation of the link between anxiety, enhanced affective reactivity and interoception in people with joint hypermobility’, Frontiers in Psychology, 5, p. 1162. doi: 10.3389/fpsyg.2014.01162
Piedimonte, C., Penge, R., Morlino, S., Sperduti, I., Terzani, A., Giannini, M.T., Colombi, M., Grammatico, P., Cardona, F. and Castori, M. (2018) ‘Exploring relationships between joint hypermobility and neurodevelopment in children (4 to 13 years) with hereditary connective tissue disorders and developmental coordination disorder’, American Journal of Medical Genetics Part B: Neuropsychiatric Genetics, 177(6), pp. 546-556. doi: 10.1002/ajmg.b.32646
Baeza-Velasco, C., Vergne, J., Poli, M., Kalisch, L. and Calati, R. (2025) ‘Autism in the context of joint hypermobility, hypermobility spectrum disorders, and Ehlers-Danlos syndromes: a systematic review and prevalence meta-analyses’, Autism, 29(8), pp. 1939-1958. doi: 10.1177/13623613251328059
Petrucci-Nelson, T., Guilhaumou, S., Berrandou, T.E., Gensemer, C., Georges, A., Huff, M., Fustier, M-A., Esmael, A., et al. (2025) ‘Complex genetics and regulatory drivers of hypermobile Ehlers-Danlos syndrome: insights from genome-wide association study meta-analysis’, medRxiv preprint. doi: 10.1101/2025.09.19.25336146
Fairweather, D., Bruno, K.A., Darakjian, A.A., Wilson, F.C., Fliess, J.J., Murphy, E.F., Kocsis, S.C., Strandes, M.W., et al. (2025) ‘Localized and historical hypermobile spectrum disorders share self-reported symptoms and comorbidities with hEDS and HSD’, Frontiers in Medicine, 12, p. 1594796. doi: 10.3389/fmed.2025.1594796
Daylor, V., Griggs, M., Weintraub, A., Byrd, R., Petrucci, T., Huff, M., Byerly, K., Fenner, R., Severance, S., Griggs, C. and Norris, R.A. (2025) ‘Defining the chronic complexities of hEDS and HSD: a global survey of diagnostic challenges, life-long comorbidities, and unmet needs’, Journal of Clinical Medicine, 14(16), p. 5636. doi: 10.3390/jcm14165636
Yao, L., Subramaniam, K., Raja, K.M., Arunachalam, A., Tran, A., Pandey, T., Ravishankar, S., Suggala, S., et al. (2025) ‘Association of postural orthostatic tachycardia syndrome, hypermobility spectrum disorders, and mast cell activation syndrome in young patients; prevalence, overlap and response to therapy depends on the definition’, Frontiers in Neurology, 16, p. 1513199. doi: 10.3389/fneur.2025.1513199