Living on the Spectrum — Episode Show Notes
Overview
Aaron and Jamie dig into a striking CDC figure: the share of 8-year-old autistic children in the United States classified as having "profound autism" has fallen from 27 percent in 2023 to nearly 17 percent. They explain why the measuring stick changed rather than the children, walk through the competing definitions behind the number, and examine what the underlying data can and cannot tell us. The second half of the episode turns to non-medication approaches for ADHD — exercise, sleep, complementary treatments, and movement-based programs — with an honest look at where the evidence is strong, where it is merely suggestive, and where it is still missing.
Prevalence of profound autism declines with new definition
Aaron opens with the new figure, reported by Brianna Abbott in The Transmitter (22 October), based on 2022 surveillance data from 13 sites in the CDC's Autism and Developmental Disabilities Monitoring Network and published in the Journal of the American Academy of Child and Adolescent Psychiatry.
Jamie explains that the children didn't change — the definition did. The term "profound autism" was coined by the Lancet Commission in 2021 to describe autistic adults who need 24-hour care, who in most cases have an IQ below 50 or very limited verbal ability. That was the definition behind the 2023 estimate. Earlier this year, a Delphi consensus group of more than 70 researchers, caregivers, autistic people, and clinicians settled on a revised definition: adaptive functioning far below age level, plus either an IQ below 50 or minimal to no verbal abilities. The adaptive-functioning requirement is the key addition. Adaptive behavior refers to what a person actually does with their skills in real settings, rather than what they can demonstrate on a test.
Matthew Siegel (Boston Children's Hospital, who served on the Delphi group but was not involved in this study) attributes the drop to the more precise definition, and also notes that the overall pool of autistic children has grown — so the group meeting the definition is a smaller, more precise slice of a larger whole. He still called 17 percent a sobering figure, since the group of children and families needing intensive support and services remains large. Jamie highlights the tension: a declining percentage can still represent more children in absolute terms if the measured population has grown. The dataset covered 7,084 children with autism.
Data limitations. Aaron asks how solid the underlying data is, given that the figure will appear in grant applications and school funding discussions. The researchers are candid: only 61 percent of children had recorded IQ information, about 74 percent had data on verbal status, 56 percent had adaptive scores, and just 54 percent had everything needed. The team used statistical imputation to fill gaps. Catherine Lord (UCLA, who participated in the Lancet and Delphi groups but not this study) said plainly that "the missing data is a huge problem," and added that the missing data probably aren't random — IQ scores are hardest to capture in exactly the children most likely to fall into this category. Jamie notes that imputation therefore does real interpretive work. Lord also frames a familiar trade-off: a diagnosis can come from a quick visit with a psychiatrist or pediatrician, while a full picture of what a child is actually capable of takes longer to build.
Site variation and definitional sensitivity. Prevalence ranged from 10 percent in Utah to 35 percent in Laredo, Texas. Joshua Anbar (Arizona State University, a study investigator) says evaluation and diagnostic practices vary across communities, and that what counts as profound autism for one person might not for another — which is why an agreed definition matters for describing this population. The team tested definitional sensitivity directly on the same children: compared with the Delphi definition, the 2021 Lancet definition produced prevalence 34 percent higher; raising the IQ cutoff to 70 produced prevalence 71 percent higher; dropping IQ entirely in favour of verbal status and adaptive scores produced prevalence 33 percent lower (an approach that mirrors a proposal this year from the federal Interagency Autism Coordinating Committee). Jamie stresses that for a school district planning intensive supports, these definitional choices shape the conversation.
Siegel flags one remaining gap: none of the competing definitions actually measure the thing the term is meant to describe — whether someone needs round-the-clock access to an adult caregiver. Every study relies on proxy measures, which he calls a gap in the field waiting to be filled.
Exercise & Sleep Are Non-Negotiables
Joel Nigg (Oregon Health and Science University) argues that exercise and sleep are underused tools for children with ADHD.
Exercise. Nigg cites a 2014 monograph from the Society for Research in Child Development concluding that exercise improved academic performance in children more than an equivalent amount of extra class or study time. Jamie notes that Nigg's own article caveats that this literature is still short on high-quality randomized trials, which he calls the acid test. Mechanism work is suggestive: animal studies suggest exercise triggers epigenetic changes — biological processes that alter how genes behave without changing the DNA sequence — in brain regions central to self-regulation, such as the hippocampus and frontal cortex. Reviews from roughly 2013 to 2016 found fitness is associated with better working memory, response inhibition, and learning in children. Jamie is careful to distinguish association from proven cause in humans, with stronger causal evidence in animals. Practically, the research cited suggests about 30 minutes of walking four times a week is enough, and skill-based activities such as martial arts or ballet appear especially effective.
Sleep. The argument is that sleep deprivation can mimic ADHD symptoms outright, and sleep is when the brain consolidates learning. Children with ADHD often have more sleep problems, including higher rates of sleep apnea and restless legs. A 2014 randomized trial found a professionally guided sleep-training program improved mood and adjustment in children with ADHD, with gains after parents received just two sessions of guidance. Nigg cites National Sleep Foundation figures of around 10 hours a night for school-age children and 9 to 10 for teens — not eight. Jamie adds an unresolved nuance: some stimulants can worsen sleep, while some experts believe stimulants can improve sleep in ADHD by quieting an overactive mind. The article presents both views without resolving them, so a sleep struggle is a conversation with a prescribing clinician about timing or formulation rather than a reason to change anything unilaterally.
Non-medication ADHD treatment options
Aaron and Jamie survey a medically reviewed Additude piece by Sharon Saline covering the wider landscape of complementary approaches — omega-3s, CBT, vitamins, exercise, brain training, neurofeedback, and mindfulness.
The core message is deliberately unglamorous: between 80 and 85 percent of people with ADHD respond well to stimulant medication, and the landmark 1999 MTA trial concluded that medication combined with behavior therapy is optimal for school-age children. The article frames complementary approaches as additions rather than replacements, and notes that behavior therapy should be first-line for children under five.
David Rabiner (Duke) makes the point Jamie feels deserves the most airtime: there is no way to predict in advance whether any individual treatment will help a person, even medication. The practical recommendation is to track symptoms with a simple daily log and adjust alongside a professional. Aaron reads this as liberating — you don't have to get it right on the first try.
Stephanie Sarkis and Ari Tuckman add a caution: placebo effects can be potent, so anything expensive or time-consuming should be held to a higher evidence bar. Neurofeedback runs two to five thousand dollars for a course, and per the piece its efficacy remains inconclusive, with weak or opposing evidence.
Evidence quality varies across approaches:
- CBT: a 2024 meta-analysis in the Journal of Attention Disorders found CBT plus medication outperformed medication alone for adult ADHD symptoms, with the advantage persisting at least three months.
- Omega-3s: modest — some studies show small-to-medium improvements in some symptoms.
- Iron: one small study found low ferritin in 84 percent of children with ADHD versus 18 percent of controls, but Jamie stresses that is one small study, and iron or zinc supplements should never be started without a doctor's supervision.
The Science of Movement
Julia Basso (Virginia Tech) wrote about dance and movement for neurodivergent people. Her team ran an eight-week musical theater program with autistic children and adults and observed what she calls significant inter-brain synchrony — coordinated brain activity between people during shared movement. Parents reported lower social phobia and separation anxiety afterward, and participants described brighter moods and a stronger sense of belonging.
Jamie flags the important caveat: those social and mood outcomes are parent- or participant-reported rather than independently measured, making this promising, exploratory work rather than a clinical conclusion. Even so, Aaron returns to the practical floor of the discussion — movement you actually enjoy, five minutes, in your living room.
Links to all original pieces are on the episode page.
Podcast Transcript
Aaron: Welcome to Living on the Spectrum. I'm Aaron.
Jamie: And I'm Jamie.
Aaron: Jamie, I want to start with a number that stopped me mid-coffee this week. The CDC's estimate of how many 8-year-old autistic children in the United States qualify as having 'profound autism' has dropped to nearly 17 percent, down from 27 percent back in 2023. That's from reporting in The Transmitter by Brianna Abbott, published on the 22nd of October. And my first reaction was honestly a bit of whiplash — did something happen to all those kids overnight?
Jamie: That's exactly the right first question to ask, and the honest answer is no. The children didn't change. The measuring stick changed. This new figure comes from a study published in the Journal of the American Academy of Child and Adolescent Psychiatry that applied a newer definition — something called the Delphi consensus definition — to 2022 surveillance data from 13 sites in the CDC's Autism and Developmental Disabilities Monitoring Network.
Aaron: So walk me through the definitions, because I think this is where most listeners will glaze over, but it's genuinely the whole story.
Jamie: It is. The term 'profound autism' was coined by the Lancet Commission back in 2021 to describe autistic people who, as adults, need 24-hour care and who in most cases have an IQ below 50 or very limited verbal ability. The CDC's 2023 estimate used that definition. But not everyone agreed with it, and criteria have varied across studies. Earlier this year, a Delphi consensus group — more than 70 researchers, caregivers, autistic people, and clinicians — landed on a revised definition: adaptive functioning far below age level, plus either an IQ below 50 or minimal to no verbal abilities. The key addition is that adaptive-functioning piece.
Aaron: Which is basically how a person manages day-to-day life — dressing, eating, communicating needs, that sort of thing.
Jamie: Roughly, yes. Adaptive behavior is what someone actually does with their skills in real settings, not just what they can demonstrate on a test. And Matthew Siegel, who's chief of clinical enterprise in the psychiatry and behavioral sciences department at Boston Children's Hospital and was on the Delphi group though not involved in this new study, says that more precise definition contributed to the drop. He also points out something else: the overall pool of autistic children has grown. So the group meeting the definition is a smaller, more precise slice of a bigger overall pool.
Aaron: He put it nicely — the group meeting the definition is getting smaller and more precise at the same time. And yet he still called the 17 percent figure sobering, which I appreciated. Because it's easy to hear 'it went down' and feel relief, and he's saying, no, this is still a very large group of children and families who need intensive support and services.
Jamie: And that tension is worth sitting with. A percentage that declines can still represent more children in absolute terms if the overall population being measured has grown. The study covered 7,084 children with autism in the dataset.
Aaron: Okay, here's my next worry, and this is where I want your careful read. How solid is the underlying data? Because 17 percent is going to end up in grant applications and school funding conversations.
Jamie: It's a fair worry, and the researchers are candid about it. Only 61 percent of the children had recorded IQ information. About 74 percent had data to determine verbal status, and just 56 percent had adaptive score data. Only 54 percent had everything needed. So the team used statistical imputation — predicting missing values to fill the gaps. Catherine Lord, a distinguished professor at UCLA who participated in the Lancet and Delphi groups but wasn't involved in this study, said plainly, 'The missing data is a huge problem.' And she added something important: the missing data probably aren't random. IQ scores are harder to capture in exactly the children most likely to fall into this category.
Aaron: Which means the imputation isn't just tidying up — it's doing real interpretive work.
Jamie: It is a workaround, and an imperfect one. Lord also framed a trade-off I think every parent will recognize: getting a diagnosis from a psychiatrist or pediatrician is one thing; getting a full picture of what a child is actually capable of is another, and you're less likely to get that from a physician under time pressure.
Aaron: So the label can come from a quick visit with a psychiatrist or a pediatrician, while the fuller picture of what a child can actually do takes longer to build. I also noticed the site-to-site variation was wild. Ten percent in Utah, 35 percent in Laredo, Texas.
Jamie: Joshua Anbar, an investigator on the study and a clinical assistant professor at Arizona State University, says evaluation and diagnostic practices vary across communities, and that may partly explain the range. His comment gets at the deeper issue: 'What is considered profound autism for one person might not be considered profound autism for another person.' Coming to an agreed definition matters so we can describe this population with significant needs.
Aaron: And they tested that sensitivity directly, right? Different definitions, same kids, very different numbers.
Jamie: They did. Compared with the Delphi definition, the 2021 Lancet definition produced a prevalence 34 percent higher. Raising the IQ cutoff to 70 produced a prevalence 71 percent higher. And dropping IQ entirely — using only verbal status and adaptive scores — produced a prevalence 33 percent lower. By the way, that last approach also leaves IQ out, which is the same feature as a proposal this year from the federal Interagency Autism Coordinating Committee.
Aaron: So the same dataset, the same children, and depending on the definition you're roughly a third lower or about 70 percent higher than that 17 percent figure. That's not a small methodological quibble. For a school district trying to plan how many kids need intensive support, those definitions shape the conversation.
Jamie: Exactly. And there's one more gap Siegel flags as key. None of the competing definitions actually measure the thing the term is supposed to describe — whether someone needs round-the-clock access to an adult caregiver. Every study is using proxy measures. He says that's a gap in the field begging to be filled.
Aaron: Which I think is a good bridge to what I keep coming back to on this show. Definitions matter for research and budgets, but families are living in the proxy-measure gap every day. They need practical supports that work. And that's the thread connecting to our second set of articles this week — a cluster of Additude pieces on non-medication approaches for ADHD, particularly exercise and sleep.
Jamie: And these are useful to discuss together because the evidence quality varies a lot, and part of what we do here is help people tell those levels apart.
Aaron: Let's start with the strongest stuff. Joel Nigg, a professor at Oregon Health and Science University, has an article arguing that exercise and sleep are underused tools for kids with ADHD. He cites a 2014 monograph from the Society for Research in Child Development concluding that exercise improved academic performance in children more than an equivalent amount of extra class or study time. That floored me.
Jamie: It's a striking comparison, and worth noting that Nigg's article itself caveats that the exercise-and-academics literature is still short on high-quality randomized trials, which it calls the acid test. But the mechanism work is interesting. Animal studies suggest exercise triggers epigenetic changes — biological processes that alter how genes behave without changing the DNA sequence itself — in brain regions central to self-regulation, like the hippocampus and frontal cortex. And reviews from around 2013 to 2016 found fitness is associated with better working memory, response inhibition, and learning in children.
Aaron: Association, not proven cause, in humans, with stronger causal evidence in animals. Got it. And practically, it doesn't need to be fancy — research cited in these Additude pieces suggests about 30 minutes of walking four times a week is enough, and skill-based activities like martial arts or ballet are especially effective.
Jamie: Sleep is the other half, and the argument there is that sleep deprivation can mimic ADHD symptoms outright. Sleep is when the brain consolidates learning, and children with ADHD often have more sleep problems — including higher rates of sleep apnea and restless legs. A 2014 randomized trial found a professionally guided sleep-training program improved mood and adjustment in children with ADHD, with gains after parents received just two sessions of guidance.
Aaron: And the National Sleep Foundation numbers Nigg cites are a useful reality check for a lot of us — school-age kids around 10 hours a night, teens 9 to 10. Not eight.
Jamie: One nuance worth adding: some stimulants can make sleep worse, while some experts believe stimulants can improve sleep in ADHD by quieting an overactive mind. The article presents both without resolving it. So if sleep is a battle, that's a conversation with a prescribing clinician about timing or formulation, not a reason to unilaterally change anything.
Aaron: Which brings me to the broader 'natural remedies' question, because there's an older Additude piece, medically reviewed by Sharon Saline, that surveys the whole landscape — omega-3s, CBT, vitamins, exercise, brain training, neurofeedback, mindfulness. And what I liked was how unglamorous its core message is. Between 80 and 85 percent of people with ADHD respond well to stimulant medication, and the landmark 1999 MTA trial concluded that medication combined with behavior therapy is optimal for school-age kids. The article frames the complementary approaches as additions, not replacements — and it also notes that behavior therapy should be first-line for children under five.
Jamie: And David Rabiner at Duke makes the point that I think deserves the most airtime: there is no way to predict in advance whether any individual treatment will help a person, even medication. So the practical recommendation is to track symptoms — a simple daily log — and adjust with a professional.
Aaron: Which is liberating, in a way. You don't have to get it right on the first try. Stephanie Sarkis and Ari Tuckman add a useful caution too: placebo effects can be potent, so anything expensive or time-consuming — neurofeedback runs two to five thousand dollars for a course — should be held to a higher evidence bar. And neurofeedback's efficacy, per the piece, remains not conclusive, with weak or opposing evidence.
Jamie: Some approaches have clearer evidence than others. A 2024 meta-analysis in the Journal of Attention Disorders found CBT plus medication outperformed medication alone for adult ADHD symptoms, with the advantage persisting at least three months. The omega-3 evidence is modest — some studies show small-to-medium improvements in some symptoms. And on iron: one small study found low ferritin in 84 percent of children with ADHD versus 18 percent of controls, but that's one small study, and iron or zinc supplements should never be started without a doctor's supervision.
Aaron: Last quick one, because I thought it was lovely. Julia Basso at Virginia Tech wrote about dance and movement for neurodivergent people. Her team ran an eight-week musical theater program with autistic children and adults and observed what she calls significant inter-brain synchrony — coordinated brain activity between people during shared movement. Parents reported lower social phobia and separation anxiety afterward, and participants described brighter moods and a stronger sense of belonging.
Jamie: With the important caveat that those social and mood outcomes are parent- or participant-reported, not independently measured. It's promising, exploratory work, not a clinical conclusion.
Aaron: But even as a starting point — movement you actually enjoy, five minutes, in your living room — that feels doable. Jamie, thank you. That's our episode.
Jamie: Links to all the original pieces are on our episode page. Thanks for listening, and take care of each other.