- Research
- Sensory & Motor
- Sensory Integration Therapy (Ayres SI)
Does sensory integration therapy help autistic children?
It depends what you mean — a structured, standardized form shows some promise; generic “sensory” add-ons don't.
Podcast·Two-voice deep dive
Listen to the discussion
Chapters & transcript
This is an AI-generated discussion of this article, transcribed automatically. The written summary below is the reviewed source — if the two differ, go by the summary.
0:00What is this discussion about?
Host A Okay, so let's unpack this. You know, usually when we talk about a medical diagnosis, there is this underlying expectation of precision. Like, it's almost like engineering.
Host B Right, yeah. It's truly binary.
Host A Yeah, exactly. Like, you fall off your bike, you break your arm, and the x-ray shows that jagged white line on the screen.
Host B And the doctor just points at it and says, you know, there it is.
Host A Right. Here's the cast. See you in six weeks. It's clean. And frankly, it is incredibly comforting to have that level of certainty.
Host B Oh, absolutely. We like things to be visible and easily categorized.
Host A Right. But then you step into the world of neurodevelopment. You step into the world of autism therapies. And suddenly that x-ray machine is just completely broken.
Host B Yeah. It's a completely different landscape.
Host A It's incredibly murky. Parents get an autism diagnosis for their child and almost immediately they are flooded with this avalanche of recommendations from doctors, from schools, you know, other parents on the Internet.
Host B Oh, yeah. It's overwhelming.
Host A Yeah. And one of the most common things they hear is that their child desperately needs something called sensory integration therapy.
Host B Ubiquitous. If you are in the autism space, you have heard this term.
Host A Right. And on the surface, it sounds fantastic. It sounds gentle. It sounds scientific, like exactly what you would want for your kid. But today we are doing a deep dive into the actual evidence behind that umbrella term, what works, what doesn't, and how we even measure the difference.
Host B Yeah, and to do that, we have a really fascinating set of sources to guide us today.
Host A We really do.
Host B We are looking at something very specific, very cutting edge. We have our hands on a demonstration run of Spectrum Connect's AI engine.
Host A Right, which is just super cool to see in action.
Host B It is. So for a little context, this is a sophisticated system designed to ingest and synthesize massive amounts of complex medical evidence. We're looking at both their internal audit document and a parent facing findings report. And the sole focus of this specific AI run is evaluating the evidence for sensory integration therapy for autistic children. Specifically, we're looking at kids mostly under the age of 10.
Host A Right. Now, before we get into what the AI actually found. We need to put a fairly large asterisk on this right up front, don't we?
Host B Absolutely. This is a crucial caveat for our listeners. What we are examining today is a demo run. It was executed to demonstrate how the AI engine reasons its way through medical evidence. It is not a finalized, published medical directive. In the system's own internal terminology, this specific run did not pass Gate D.
Host A Okay, wait, Gate D sounds a bit like something out of a sci-fi movie.
Host B I know, I know.
Host A Let's translate that. What does failing to pass Gate D actually mean in terms of how we should view this data?
Host B In simple terms, it means it lacked a human co-pilot.
Host A Ah.
Host B So for a full conformant run to be published as verified medical guidance by the system, it requires dual independent human review. Real human experts have to double check the AI's provisional judgments.
Host A Got it. So it needs that human oversight.
Host B Exactly. It also has to go through rigorous methodological checks like prospective registration and reproducible database counts. Right. But even as a demo, this document is a masterclass in how we evaluate medical evidence. It pulls back the curtain on a therapy that is widely used, heavily debated, and honestly, fundamentally misunderstood by a lot of well-meaning people.
3:27Is "sensory therapy" one thing or several?
Host A Misunderstood really seems to be the operative word here, because when families are offered sensory therapies at a local clinic, the science, at least as synthesized by this AI audit, shows that this single umbrella term is actually masking two wildly different interventions.
Host B Yeah, that is the defining problem of this entire landscape. The audit officially categorizes this as a definitional scope split. Basically, when people say sensory therapy, they're usually talking about one of two completely different things, but they are treating them as if they are identical.
Host A So break that split down for us. What are the two distinct things hiding under this umbrella?
Host B So on one side, you have Ayres Sensory Integration, which is usually abbreviated as ASI.
Host A ASI. Okay.
Host B Right. This is an incredibly structured, highly specific approach. It is delivered one on one by a specially trained occupational therapist, and it demands what the literature calls high fidelity.
Host A Meaning it has to adhere to a very strict, uncompromising standard of delivery. Like it has a manual, it has rules.
Host B Exactly. But then on the other side of the split, you have what the literature refers to as sensory based interventions or SBI. These are much looser. They're essentially generic add ons. We are talking about things like weighted vests, brushing techniques, which is sometimes called the Wilbarger protocol. Or generalized sensory diets where a child is just given various sensory toys throughout the day.
Host A Okay, so if ASI is this strict, highly trained clinical regimen, the other side, the SBIs, it sounds a bit like the Wild West of sensory toys.
Host B It really is.
Host A It sounds like someone telling you they're doing fitness.
Host B Yes, perfect analogy.
Host A Right. Because that phrase could mean they're training with an Olympic coach following a highly calibrated biomechanically perfect regimen. Or it could mean they're casually wearing a pair of running shoes while walking through the grocery store.
Host B Right. Yeah.
Host A We obviously can't evaluate those two things as if they were the same intervention.
Host B Yeah.
Host A I mean, if I'm wearing running shoes to buy milk, I'm not going to see the same cardiovascular results as the Olympic athlete.
Host B That is an excellent analogy. Now, imagine a scientific researcher trying to figure out if fitness works. And to do that, they lump your grocery store trips in with the Olympic training data.
Host A Oh, wow. Yeah.
Host B The overall average result would be completely skewed. It misestimates both activities. Right. It makes the grocery store trip look more effective than it is. And it waters down the results of the Olympic training. And that is exactly the historical problem that the Spectrum Connect engine flagged in the medical literature regarding autism therapies.
Host A Because they were averaging the running shoes with the Olympic coach.
Host B Precisely. When past reviews pooled these two distinct things together under the banner of sensory therapy, it created massive confusion.
Host A I can imagine.
Host B But when you actively separate them in the data, like the landmark Case-Smith 2015 systematic review did, a very clear, stark picture emerges.
6:21Do weighted vests and brushing help?
Host A Okay, let's get into that picture because this is where the rubber meets the road for parents. What happens when we evaluate them separately?
Host B Well, the Case-Smith review found that Ayres Sensory Integration, the structured high fidelity one, actually shows emerging evidence of benefit.
Host A Okay.
Host B But the looser sensory based add ons, the SBIs like weighted vests and brushing protocols, are simply categorized as not supported.
Host A Wait, really?
Host B Yeah. The evidence for those passive interventions is either wildly inconsistent or downright insufficient. They show no measurable benefit.
Host A I really want to pause and emphasize that because that is a massive paradigm shift. I mean, a weighted vest is something you see recommended constantly.
Host B All the time.
Host A You see it on parent forums. You see it used in special education classrooms, catalogs, everywhere. And what this AI synthesis is telling us is that the evidence basically says it does nothing. But that brings up a really important question about the mechanism of action, right? Like, why does one work while the other fails? What is it about ASI that actually creates a benefit, whereas strapping a heavy vest to a kid does not?
Host B What's fascinating here is it all comes down to the neurology of active versus passive engagement. The underlying theory of ASI is built on neuroplasticity, the brain's ability to reorganize itself. For the brain to actually rewire how it processes sensory information, it can't just be a passive recipient. If I just brush a child's arm with a specialized brush or put a heavy vest on their shoulders, the child's brain doesn't have to do anything with that information. It's entirely passive.
Host A Right. They are just sitting there receiving a sensation.
Host B Exactly. But true ASI requires an adaptive response. In an ASI clinic, a child might be navigating a suspended obstacle course where they have to time their jump onto a swinging bolster to grab a puzzle piece.
Host A Oh, wow. So it's very involved.
Host B Very involved. Their brain is actively doing complex math, integrating visual input, balanced proprioception, and tactile feedback, all to execute a specific goal.
Host A Yeah.
Host B That active participation is what is theorized to create neurological change.
Host A Okay, so it's the difference between watching a workout video on your couch versus actually lifting the weights.
Host B Exactly.
Host A The brain has to sweat a little bit to change. Yeah. Okay, so if the evidence is so bifurcated, if active ASI shows promise and passive vest do not, a parent's immediate hurdle is knowing which version they're actually getting when they walk into a local clinic.
Host B Right, which is tough.
Host A Because the sign on the door probably just says occupational therapy.
8:51What does it mean to deliver Ayres Sensory Integration to fidelity?
Host B And that brings us to the concept of fidelity, which is the absolute linchpin of the ASI evidence base. For Ayres Sensory Integration to show any promise in the data, it cannot just be a loose collection of fun sensory activities. It must be performed under strict fidelity.
Host A Let's define that. Because outside of a clinical trial, I feel like fidelity usually means a marriage vow or the audio quality of a record player.
Host B Fair enough, yeah.
Host A What does it actually mean for a therapist?
Host B So in clinical research, fidelity means the intervention was delivered exactly as the creators designed it to be delivered. There is an actual published ASI fidelity measure.
Host A Oh, there is?
Host B Yes. It is a checklist. It means the therapist has specific, rigorous postgraduate training in ASI. It means the physical clinic environment meets very specific structural requirements, you know, with suspended equipment.
Host A Okay.
Host B And most importantly, it means the therapy is actively tailored to the child in real time, not just a static passive routine.
9:52How can I tell whether a clinic offers true Ayres Sensory Integration?
Host A So from a practical standpoint, how does a family spot true ASI in the wild versus the generic practices? If a clinic puts a weighted vest on a child or tells the parent to brush the child's arms every two hours, is that an immediate red flag that they are not receiving high fidelity ASI?
Host B Yes, it is. The findings report generated by the AI explicitly tells families how to navigate this exact scenario.
Host A Oh, that's helpful.
Host B It suggests that if you are considering occupational therapy, you cannot just assume you are getting the evidence-backed version. You need to ask the clinic director directly like, Is this Ayres Sensory Integration delivered to fidelity by a trained OT?
Host A Right.
Host B Because if the clinic's approach relies heavily on generic passive sensory add-ons, you are stepping out of the realm of promising evidence and into territory that the science does not support.
Host A That is incredibly empowering for a parent to know what question to ask. But Okay, let's play this out. Let's say a family navigates that hurdle. They ask the hard questions. They find the right clinic. And they verify, yes, we are getting high-fidelity Ayres Sensory Integration from an authentically trained OT.
Host B Okay, best case scenario.
Host A Right.
11:01Does sensory integration therapy treat autism itself?
Host A The next massive question we have to tackle is, what is this intense therapy actually supposed to be fixing? Because this is where the AI audit uncovered a huge discrepancy in how the scientific community measures success.
Host B And this is where parents and clinicians alike have to radically recalibrate their expectations. If you read the older medical literature, or if you just look at the broad marketing claims sometimes made on the internet, you might easily assume that sensory therapies are designed to treat autism itself, to reduce the core symptoms. But the data shows something completely different.
Host A Wait, it doesn't treat the autism?
Host B High fidelity ASI does not treat the core traits of autism. It does not magically change a child's social communication differences. It doesn't reduce repetitive behaviors in a vacuum. It doesn't make a child less autistic.
Host A Then what are the researchers actually measuring in these landmark trials to claim that the therapy works?
11:55How do the studies measure progress?
Host A If they aren't curing autism, what is the goalpost?
Host B The goalpost is everyday functional participation.
Host A Okay.
Host B They are trying to help the child exist more comfortably and capably in their daily environment. And the way researchers measure this is through a mechanism called goal attainment scaling, or GAS.
Host A Goal attainment scaling. That sounds like a metric a middle manager would use at a corporate retreat.
Host B I know. I know.
Host A How does that actually work with a child?
Host B It is actually a highly individualized, incredibly personal metric. In a lot of medical trials, you use a standardized checklist for every single patient. Did the blood pressure go down? Yes or no.
Host A Right.
Host B But GAS works entirely differently. Before the therapy even starts, the therapists, the parents, and sometimes the child sit down together and set very specific functional goals tailored only to that child's life.
Host A Give me a real-world example of what one of those goals looks like.
Host B So for one child, the goal might be Johnny can get dressed independently in the morning, tolerating the feeling of his socks without having a 30-minute meltdown.
Host A Okay.
Host B For another child, the goal might be Sarah can sit at the noisy family dinner table for 10 consecutive minutes without needing to leave the room.
Host A Got it.
Host B They establish that as the baseline. Then they grade the outcome on a scale. Zero means they met the accepted goal. Plus two means they vastly exceeded it. Minus two means they fell far short.
Host A Oh, I see. So instead of judging the therapy on a rigid standardized test that asks, you know, did we cure the underlying neurological condition of autism? We are grading it on a highly personalized rubric that asks, can Johnny put his coat on and get to school today?
Host B Yes.
Host A That fundamentally changes the entire definition of what clinical success looks like.
Host B It changes it entirely. And this is exactly why the AI engine flagged this whole topic under the category of discordance.
Host A Discordance.
Host B Yeah. Discordance basically means scientists are yelling across the room at each other because they are reading two completely different thermometers.
Host A Wow.
Host B Older reviews were evaluating ASI based on whether it changed standardized core autism symptoms. They were looking for a reduction in autism traits. They didn't find it. And so they declared the therapy a failure.
Host A But the newer reviews are looking at the GAS scores.
Host B Exactly. The newer reviews, like the 2025 AOTA systematic review, which is heavily featured in our audit, rely on those individualized GAS scores. They found that when ASI was conducted under high fidelity, autistic children significantly improved in their individualized goals related to occupational performance and daily function.
Host A That's great. Yeah.
Host B But we must be absolutely clear, the success is Johnny putting on his coat. It is not a change in his core autism neurology.
Host A Which brings up a very real conversation about the investment of time and resources versus the functional life improvements, right?
Host B Absolutely.
Host A Because when families are looking at the massive investment of time, energy, and clinical costs for these therapies, they need to know if the anticipated payoff is a fundamental change in their child's neurotype or if it's just gaining the ability to put on a coat.
Host B Right.
Host A And what the data is saying is that it's the coat. which is hugely important for daily life. Don't get me wrong, but it's a very specific lane.
Host B It is a critical distinction for managing a family's resources and emotional bandwidth.
Host A But wait, I have to pull a thread here. Here's where it gets really interesting.
Host B Okay, pull away.
Host A If the success of this therapy is being measured by highly personalized goals set by the therapist and the parent, rather than standardized blind tests, I mean, we have to look very closely at the people grading the test.
Host B Yes, we do.
Host A Because if the therapist sets the goal, delivers the therapy, and then asks the parent, hey, do you think he's doing better at putting on his coat? I mean, that feels incredibly subjective.
Host B It is, and that lack of blinding is a widely acknowledged limitation in almost all behavioral therapies. You physically cannot blind a therapist to the therapy they are actively delivering.
Host A Of course not, yeah.
Host B And parents usually know if their kid is receiving an intense intervention, so they are naturally primed to look for improvements.
Host A So if the grading is subjective, the integrity of the people running the trials becomes paramount.
16:01Who did most of the research on Ayres Sensory Integration?
Host A And going through this AI audit document, the engine actually picked up on a glaring red flag regarding the researchers conducting these landmark trials, didn't it?
Host B It did. This raises an important question because the AI engine utilizes a source checker protocol, and it surfaced what it formally calls a developer or proponent group COI, a conflict of interest.
Host A Now, I don't know the specific names in this field, but I'm guessing the AI found a pattern in the authorship of the studies that show positive results.
Host B That's exactly it. The positive evidence for ASI, the major systematic reviews, the landmark randomized controlled trials, they are almost entirely authored by the exact same small group of researchers.
Host A Really?
Host B In the data, you see the same four or five names popping up over and over again. Schaaf, Mailloux, Parham, May-Benson. These are the specific experts who developed and championed the ASI fidelity measure in the first place.
Host A But wait, hold on. The people who invented the ruler are the ones measuring their own success.
Host B That is exactly what is happening.
Host A If the developers of the therapy are the ones conducting the landmark trials and they are using personalized, unblinded goals to declare victory, doesn't that inherently compromise the trustworthiness of the entire evidence base?
17:16Why is the certainty low?
Host B It introduces a profoundly high risk of bias, and the AI engine maps this out mathematically using established academic appraisal tools. It ran the systematic reviews through a tool called AMSTAR 2, and it ran the clinical trials through a tool called RoB 2.
Host A Okay, let's drop the acronyms for a second. What are those tools actually doing? ELI5, that for us.
Host B Sure, sure. Think of AMSTAR 2 like an audit of an audit. It's a checklist that asks, you know, did these researchers look at all the available literature fairly or did they cherry pick the studies that make them look good?
Host A Right.
Host B And RoB 2, which stands for risk of bias, is basically a lie detector for the methodology of a trial. It looks for leaks in the study's design. It asks, are there blind spots here where the researchers' own desires could have artificially inflated the positive results?
Host A Got it. And when the AI ran the ASI evidence through these lie detectors, what was the verdict?
Host B For the randomized trials, the judgment landed on some concerns moving heavily toward a high risk of bias. And the AI explains exactly why. The sample sizes of kids are small. The trials are unblinded by their very nature. The outcome is measured by that subjective goal attainment scaling. And the developers of the therapy are deeply involved in running the trials.
Host A I mean, that is the non-drug equivalent of pharmaceutical industry funding.
Host B Yes, exactly.
Host A If a drug company runs all the trials showing that their new pill is a miracle, we naturally demand independent verification before we trust it.
Host B Right.
Host A The AI is essentially arguing that we need to apply that exact same skepticism here. But how does the AI handle that? Does it just throw the evidence in the trash?
Host B No, no, it doesn't discard the evidence because the trials were conducted, but it drastically downgrades the certainty of the evidence. The engine uses the GRADE framework, which is a standardized system, to rate how confident the medical community can be in a set of data. Because these are randomized controlled trials, they start at a high certainty rating. But then the AI drops the rating down a level because of the unblinded subjective GAS measures.
Host A Makes sense.
Host B And then it explicitly flags that a human reviewer needs to weigh the severity of that developer conflict of interest, which could easily push the final certainty rating all the way down to low or even very low.
Host A That is a staggering reality check. We started this deep dive looking at a therapy that is casually recommended to almost every single family navigating an autism diagnosis. And what the engine has distilled is that ASI is ultimately categorized as promising but contested. It shows potential for functional life skills, but the lack of independent third-party confirmation means families have to tread incredibly carefully.
Host B It really does. The AI notes that larger, truly independent, meaning non-developer randomized trials are exactly what this field desperately needs to move forward. It does mention that some independent comparative trials are currently in progress for 2025, which is great news. But right now, today, the evidence rests on a very fragile, heavily biased foundation.
Host A So what does this all mean?
20:17What are the key takeaways?
Host A Let's bring all of this home for the listener. If you are a parent sitting in an IEP meeting or a clinician trying to write a referral or just someone trying to understand the neurodevelopment space, here are the core synthesized takeaways from the Spectrum Connect demo run.
Host B Go for it.
Host A Number one. Passive sensory add-ons, the weighted vests, the brushing protocols, the generic sensory diets, the evidence synthesis is clear. They have no supported benefit for core autism traits or functional goals.
Host B Correct. The data simply does not back up passive interventions. The brain requires active engagement.
Host A Number two, high-fidelity Ayres Sensory Integration. When it is delivered perfectly, actively, by a specially trained OT, it shows real promise for helping kids achieve individualized life skills, like getting dressed, sitting at the dinner table. But it does not treat the core neurological traits of autism.
Host B Right. It targets functional daily participation, not the autism itself.
Host A And number three, the trustworthiness of that promise is currently severely hampered because the evidence is heavily intertwined with the therapy's own creators, and it relies on highly personalized, unblinded metrics to prove its success.
Host B Which perfectly encapsulates why the AI's final status for ASI lands on promising but contested.
Host A Exactly. And it is so vital that this kind of rigorous, impartial analysis makes its way out of dense academic journals and into the hands of the people actually making decisions on the ground. which is exactly the mission here. Currently, Spectrum Connect is bridging the science of autism therapies to make it more approachable for parents. If you are a credentialed expert, an MD, PhD, PsyD, or DO, and would like to help us in validating our findings, you can email us at research@spectrumconnect.us.
Host B Yeah, the more independent experts we have reviewing and challenging this data, the better equipped families will be to make informed choices.
Host A Absolutely. Now, before we wrap up, we always like to leave our listeners with something provocative to chew on, something to mull over on your own after the deep dive ends. Yeah. And there's this one little detail from the audit surveillance and gaps section that just keeps nagging at me.
Host B Oh, the phantom treatment idea?
Host A Yes, exactly. Think about this. The audit explicitly notes that the benefits of ASI only appear under incredibly strict fidelity.
Host B Right. The suspended equipment, the highly trained OT, the constant active adjustment.
Host A But typical community sensory occupational therapy often does not adhere to this strict fidelity.
Host B Sadly, no. It's very hard to find outside of a major academic center.
Host A Right. So here is the puzzle to ponder. If the therapy only actually works in a perfectly controlled, high-fidelity academic setting, are families in the real world essentially chasing a phantom treatment that doesn't actually exist in their local community clinics? Are they paying for the Olympic trainer but only getting the grocery store run?
Host B It's a tough question to face.
Host A It really is. Something to think about next time you see sensory therapy on a clinic brochure. Until next time, keep diving deep.
Key Takeaways
“Sensory therapy” means two different things — a specific, structured approach called Ayres Sensory Integration (ASI), delivered one-on-one by a trained OT to a strict standard, is different from looser “sensory” add-ons like weighted vests or brushing. They don't have the same evidence behind them.
ASI shows some promise, done to standard — when delivered exactly as intended, several small trials suggest children make progress on their own individual goals, like getting dressed or joining a routine. But it doesn't treat autism itself.
Weighted vests and brushing aren't well supported — reviews haven't found good evidence that these common “sensory” add-ons improve autism-related outcomes, even though they're often offered under the same “sensory therapy” label as ASI.
Most of the positive evidence comes from the therapy's own developers — nearly all the supportive ASI research is authored by the people who created it. That doesn't make it wrong, but independent confirmation still matters and is limited.
If you're considering occupational therapy for sensory differences, it's worth asking whether it's Ayres Sensory Integration delivered to fidelity by a trained OT — or generic sensory add-ons, which have weaker evidence.
What this means for you
“Sensory integration therapy” is an umbrella term covering two quite different things. Ayres Sensory Integration (ASI) is a specific, structured occupational-therapy approach, delivered one-on-one by a trained OT following a strict standard (“fidelity”), using playful sensorimotor activities tailored to a child's own goals. Sensory-based interventions are looser add-ons — weighted vests, brushing protocols, or “sensory diets.” They are not the same, and the evidence for each is different.
When ASI is delivered exactly as intended, several small trials suggest children make progress on their own individual goals — such as getting dressed, joining a routine, or taking part in daily activities. Two caveats: it's about individual functional goals, not autism's core traits, and much of the research comes from the group that developed ASI, so independent confirmation matters. For the looser sensory add-ons, the evidence is weak — reviews haven't found good support for weighted vests or brushing protocols improving autism-related outcomes.
Who was studied. Most studies included young children under about 10. The ASI trials measured progress on each child's own goals (Goal Attainment Scaling) rather than a standardized measure, and the therapy has to be delivered to a strict standard for the evidence to apply — which is often not how community “sensory” work is delivered in practice.
Where the studies landed
Promising but contested — Ayres Sensory IntegrationThese 6 sources are grouped by what they found for Ayres Sensory Integration (ASI) specifically — the better-evidenced form. Generic sensory add-ons are covered separately below, in “What the strongest evidence says.”
5 Points toward ASI helping
1 Different form — sensory add-ons, not supported
Tap any tile to read it. Most of the reviews pool the same few trials, so they aren't independent evidence.
Points toward ASI helping5
Different form — sensory add-ons, not supported1
Spectrum Connect reviews published research on interventions parents are exploring for their autistic children — so you can see where the evidence actually stands. No agenda, no selling, no cherry-picking. Just the studies, our method, and what it means for you.
See the research behind this
For anyone who wants to check our work · 6 sources
Where we looked
This run was a scoping search only — done via general web search, not the reproducible Boolean search of record and not the PubMed/Epistemonikos API layer we use on a fully conformant run. That means we can't publish reproducible per-database counts or a formal PRISMA flow for this run. Below is the search string a full conformant pass would run against PubMed, the Cochrane Library, CINAHL, PsycINFO, Embase, and OTseeker — we haven't executed it against the database APIs yet.
("autism spectrum disorder"[MeSH] OR autistic[tiab] OR ASD[tiab]) AND ("sensory integration"[tiab] OR Ayres[tiab] OR ASI[tiab] OR "sensory-based"[tiab] OR "sensory processing"[tiab] OR "weighted vest"[tiab] OR "sensory diet"[tiab]) AND ("occupational therapy"[tiab] OR intervention[tiab] OR efficacy[tiab] OR systematic[sb] OR "meta-analysis"[pt] OR "randomized controlled trial"[pt])
Run on PubMed →
What we did with what we found
What the strongest evidence says
When delivered to fidelity by a trained OT, several small trials suggest children make progress on their own individual goals — getting dressed, joining a routine, daily activities. Much of this evidence comes from the therapy's own developers.
Neither approach has been shown to change core social communication. Improving everyday functioning and participation is the realistic aim — not autism's core traits.
Reviews haven't found good support for weighted vests or brushing protocols improving autism-related outcomes. These are often offered as “sensory therapy” but aren't the same as ASI, and don't have the same evidence behind them.
Keep reading
What would you like to do?
Want to talk to a parent who’s been there?
We match you with families whose kid is like yours. Free, takes two minutes.
Heard of something we haven’t covered?Suggest an article
If another parent, a clinic or a video mentioned a therapy you can’t find here, tell us. We’ll read the research and review it next.
Suggest an articleThink we got something wrong?Report a factual error
We publish the whole record so it can be checked, and that only counts if we act on what you find. If a number looks wrong, a study is missing or has been retracted, or we’ve read a finding in a way the evidence doesn’t support, tell us.
You don’t need a research background to file one. “This doesn’t match what our doctor told us” is a useful report. Every one reaches a person: we reply within seven days, and within thirty we have either corrected the page or told you when we will. Substantive reports send the affected steps back through the protocol and need fresh sign-off before anything here changes.
Report a factual errorAre you a clinician or researcher?Review this article
If you have read this and the evidence holds up, you can sign off on it. We send you a personal review link where you confirm your credentials, and our team checks every review before anything is published. Once approved, your name and a link to your profile appear on this article.
Ask to review this