With the wild-west of IVF and sperm donation bypassing the mechanisms of natural selection, I anticipate all sorts of bad results from sub optimal gamete connections.
It's fairly easy to test if there are substantially more severe autism cases in IVF kids vs naturally conceived kids. However, even in the wealthiest societies, IVF kids are only 2% of the total or so.
IVF will increase until the societies in which it predominates collapse or are killed off by societies which are not depending on it.
IVF necessarily involves humans trying to bypass the natural selection with their own priorities by making vast excess of humans, killing off the non preferred ones, and sending disproportionate numbers of preferred ones to a lifetime of medical dependency.
Humans are not qualified to run the selection process.
IVF is an expensive, accelerated process for accumulating genetic and epigenetic anomalies. It is not sustainable.
Most of the IVF is due to breeding too late. Some due to STDs like mycoplasma destroying the normal passage through the fallopian tubes. Or due to endometriosis. None of that is strictly genetic, most is bad nurture. So if the next generation no longer goes for bad nurture, these are one-off events.
Obviously, cystic fibrosis leads to sterility, but I haven't heard of many cases trying IVF for that reason.
Breeding too late is a really big deal that is swamping the remaining available health care supply, even when it occurs without IVF.
Look up the causes of endometriosis and ponder how these could affect the environment of a developing IVF baby.
There are hormonal anomalies and numerous other environmental factors which can effect genetic performance in multigenerational ways, even if the actual DNA sequence is unaltered.
Please also review what's going on with the actual supply of health care. Many of us are getting out in the effort to do no harm, and for numerous other reasons. It is no longer attractive to people with problem solving and diagnostic minds. Those that remain are getting very burned out.
Increased health care demand is being engineered into an environment of reduced supply and quality.
I think people meeting in college and getting married is a mistake for a myriad of reasons with increasing the chances of autism in their offspring one of them. IQ is like height in that it is a good gene and it will get stronger over time. The problem with IQ and height is that at some point you get diminishing returns. My solid 125 IQ is just as good as the 150 IQs I was around growing up in a community filled with high achievers (if I told you what institution was nearby you would understand). So if a college graduate marries a non college graduate we will get more 125s and fewer 150s…but we probably have too many 150s and we don’t need more law school professors with 150s anyway. So 150s often want to be semi famous and not work too hard while being wealthy and so they are attracted to professions like lawyer that are pretty dumb professions.
Yes, the assortative mating theory is something I've seen play out in my own personal experience and I think it's actually a big driver. I also suspect that previous social systems did a better job of socializing mildly autistic people, helping them get over their problems and fit in normally. Sadly neither of these hypotheses are being pursued by RFK
Change requires a cultural and social change. There was greater social pressure to conform, a belief that children had to be molded by their parents and a subordination of the self to the community and its social norms. All of these have been undermined by liberalism and the focus on the individual over the community.
Being socially awkward and autistic is now an identity group with people celebrating neurodivergent people. It’s now who they are and they will fight tooth and nail against any effort to change “who they are.”
Yes, this theory could explain some of the effects. Do you know anything about the relationship between Asperger's and "true" autism?( do two high functioning Asperger's parents have a higher chance of having a kid with "true" autism than two non-Asperger's parents?)
At the rate with which Psychiatry is medicalising eccentric behaviours the day will come soon when having a "normal" child will be a statistical rarity.
While I absolutely agree with the argument that raw data is hugely confounded leading to a dramatic increase of autism diagnoses independent of actual rates, it may be helpful to also note that maternal age at birth is a contributor to diagnosed autism, and maternal age at birth has been increasing since at least 1990 or so.
"The study found that the incremental risk of having a child with autism increased by 18 percent for every five-year increase in the mother’s age. A 40-year-old woman’s risk of having a child later diagnosed with autism was 50 percent greater than that of a woman from 25 to 29 years old."
Yeah, I was wondering about this correlation as well, one reputed to be particularly significant when both parents have higher education featuring advanced mathematics and adjacent STEM disciplines.
I wish there were a version of this written for the layman. Or an video version with illustrations maybe. There are several people I'd like to send it to but I can't imagine any of them making it more than a few paragraphs in due to the technical language. I'm not suggesting you should take on that responsibility, I'm sure this was weeks of work in itself, I'm just saying I wish a version existed that was specifically intended for people with an IQ of ~90. Thanks for writing it! If nothing else, I can reference it myself to mine for data when the topic arises.
Hi Cremieuex, Thanks for the analysis. It sparked me to comment that increased awareness and media coverage of autism has also increased and this might have led to a significant shift in societal "attitudes" (for bad or worse) towards the condition. In Finland, for example, autism has become a more openly discussed topic, with a growing number of reality series and documentaries featuring autistic individuals (our "NPR" airs the show on TV on primetime). Same goes to ADHD, dyslexia, eating disorders, etc. (the latter might have even statistically increased, I'm no scientist so cannot comment). Anyway, my question: is this autism discourse part of an "awareness trend" or a cultural change towards "embracing value" in "human diversity"?
"it is still to be lauded, because their lives still have value."
It says something rather dark about modern society that such a statement would be expected to result in condemnation and cancelled subscriptions rather than being widely regarded as praiseworthy (or at least unremarkable for stating the obvious).
Thank you. Autism runs in my family and I grew up in a medically fragile foster care home. I have several adopted siblings who are severely disabled by a variety of conditions. I love them dearly. The value of a human life should never be measured by IQ or functionality, we are all so much more than merely an equation of net economic contribution vs economic cost to society.
Huh, well thanks for this. I think I'm the target audience since I was concerned about the rise in autism and supportive of RFK. So you've made me more skeptical. Certainly the change in diagnosis is a big concern. I would like to see some type of cross age study here in the USA. I've got nothing against the Swedish study. But I do worry that we are eating more 'weird.. toxic' stuff here in the US and so it could be worse here. If there is some environmental effect. I've listened to RFK some and my impression is that if someone said to him, "You know we don't really know if autism is increasing." He would be for more data on the real trends. Thanks again for writing this.
Thanks for the most comprehensive analysis of the supposed autism rate increase. This has become a charged issue, and most of the time, reason is a casualty.
"But is this a good priority? To determine why autism rates have increased, we first have to know if autism rates have increased."
Excellent point, and it is certain by any means that they have increased.
"In fact, all intertemporal comparisons of autism rates are biased for a simple but important reason: the definition of autism has changed!3"
And that is a significant fact to consider.
I sense there is no increase in autism over the years, and you laid out the logic for the conclusion. I am sure you will be inundated with indignation.
Your section on profound autism didn't answer my main question here: now or decades ago, people would notice someone was non-verbal and banging their head against the wall no matter what the diagnostic criteria. Do we have any data on the frequency of such people in, say, the 80s vs now? 25% of the exploding number of diagnoses are "profound" - even allowing that some of that is just "normal" low IQ, it seems like there is a real increase here. How many people were they hiding in those asylums?
Back in the early '80s I hitched a ride with a woman who was going to visit her autistic son, in an asylum, who had been diagnosed and committed as an infant.
Spending time with my nonverbal autistic grandson, I run into other grandparents with their grandchildren on the spectrum, and the question is always the same, “Do you remember having classmates with these kinds of traits when you were a kid?” The answer is always, “No.“ I don’t think we remember having friends who had siblings in institutions either. My mother died last year at 99 years old and she also says cases of cognitive decline were very very rare in her experience. I think this kind of anecdotal data is excellent and I think the uptick in autism (along with all the other neurological diseases, such as MS, Parkinson’s, Alzheimer’s, ) is very, very real.
You might also have to consider that medical advances are keeping children alive who didn’t survive long enough to reach school age, and that families used to keep neurodivergent children that did survive hidden away. Finally there was far less reluctance to institutionalize before the emergence of what you might call the “inclusion” movement.
As a mother navigating the pre-assessment process for a four-year-old, I deeply appreciate your article. My son’s symptoms are so mild that, as parents, we sometimes wonder whether all this labeling is truly necessary. A private speech therapist even declined to continue seeing him, saying she didn’t want me to waste my money. Yet the school insisted on pursuing an assessment to secure extra funding.
It makes one question whether labeling a child with extremely mild symptoms truly helps them—or if it risks shaping their behavior and self-esteem in ways that might not be positive, while also straining the financial health of the system.
Making decisions and thinking clearly as a parent in such a complex and emotionally charged situation is incredibly difficult. I’m grateful to find such well-detailed information.
Absolutely unneeded. We just need Galtonian eugenics. Pay the sickly to breed less, pay the healthy to breed more. And it's mostly about the latter since the TFR is sub 2.
One of the great mistakes we made as a nation was the canonization into law (ADA) of the idea that accomodations are due to individuals based on their membership in some named category. Not only has it wrecked havoc on schooling and medical diagnosis it bears substantial blame for our oppression olympics culture by leaving anyone whose needs don't go into a labeled box out in the cold.
And this was never a sensible way to handle the situation. For instance, take accomodations in schooling. As someone who couldn't focus or remain awake in any lecture class I'm a firm believer in making sure no one is denied an education or a grade because of 'incidental' factors. But without an official diagnosis I had a few professors make life very difficult for me (if I'd gone to a less elite school I may not have graduated at all). And this just doesn't make sense, there is no need for an arbitrary cut off below which you are punished for a trait and above which you get special accomodation and it makes people resent disability accomodations because they feel they let people gain unfair advantage. For instance, in my case it is unacceptable for a professor to deny a student who shows themselves to have mastered the material via performance on exams, homework and any projects, a good grade because they didn't imbibe their wisdom in class. At some point students need to learn self-motivation and until then you can make it merely inconvenient not to attend.
We could instead identify what aspects of evaluation are truly central to the course -- those should never be relaxed for a disability. If it is the department's judgement that being able to complete problems in a limited time frame is an essential skill (bullshit imo) then you can no more offer extra time accomodations than you could accomodate muscular weakness in weight lifting). If it isn't -- which it never is profs are just too lazy to write tests appropriately (often me too) -- then anyone who wants that accomodation should be able to request it. Sometimes it may need to be discouraged and annoying to get (some ppl really do need laptops to take notes but making it a hassle can help discourage distractions).
This holds more generally. What businesses do or don't need to have by way of accessibility features is ultimately a function of cost, incidence in the population and alternatives not threshold effects. A better run government in a country with higher social trust would delegate such regulation to an independent expert panel to weigh costs and benefits -- but it's not feasible if it has to be in the executive branch. And yes that might include issuing little licenses for certain service animals (again no need to check the level of impairment only that the animal is professionally trained to the appropriate level and owner took a class to deter the people who don't really need the emotional support animal).
Instead we get worst of all worlds. Half the country thinks accomodations are a joke ruining the country while those with partial issues find it difficult to get partial accomodation. And no one actually does the cost benefit calcs to see what kind of improvements a buisness should need to make.
Your footnote 20 on ADDM data seems strange. You elaborate how profound cases may shift to non-profound as the child ages, but why would that undermine the reported growth, if it is measured at the same year (e.g. 8 year old) across time? An apples to apples comparison in https://pmc.ncbi.nlm.nih.gov/articles/PMC10576490/ ends up reporting
"During 2000-2016, the prevalence of non–profound autism increased from 3.94 to 14.26 cases per 1000 children aged 8 years; the prevalence of profound autism increased from 2.68 to 4.59 cases per 1000 children aged 8 years."
You do not cite such evidence in the post and delegate talking about the rise of profound cases according to ADDM data to a footnote, where you find ways to downplay it. You talk about different states having different diagnostic criteria, etc., and end with "This fact should be a lesson to those who cite the ADDM’s data as an indication of a “real” autism epidemic". What kind of lesson? That the data is safe to ignore? Isn't the onus on you to explain how "variation in reporting sources, practices, and policies" between states was likely to have led to a marked and sustained increase over time?
These seem like isolated demands for rigor. "Variation in reporting sources" is everywhere to be found. Generally the situation with data is complicated everywhere, including in stats that favor your story, but you do not pay such minute rhetorical attention to these issues then. E.g. you cite the Swedish data comparing symptoms with diagnoses without explaining that the symptoms data comes from phone surveys of parents of twins, while the diagnoses come from the national register of all children. Lots of reasons why the temporal behavior of the symptoms of autistic behavior in twins (reported by parents in a phone questionnaire, too!) may not reflect the entire population. You do not discuss the reasons or even the fact of the different populations, but find plenty to criticize in the ADDM data.
Great post but very noticeably absent: the assortative mating theory, of which there is at least some evidence: https://pubmed.ncbi.nlm.nih.gov/31200929/
Great point. I did miss that and it might make a contribution. I may go back and add it later.
With the wild-west of IVF and sperm donation bypassing the mechanisms of natural selection, I anticipate all sorts of bad results from sub optimal gamete connections.
It's fairly easy to test if there are substantially more severe autism cases in IVF kids vs naturally conceived kids. However, even in the wealthiest societies, IVF kids are only 2% of the total or so.
IVF will increase until the societies in which it predominates collapse or are killed off by societies which are not depending on it.
IVF necessarily involves humans trying to bypass the natural selection with their own priorities by making vast excess of humans, killing off the non preferred ones, and sending disproportionate numbers of preferred ones to a lifetime of medical dependency.
Humans are not qualified to run the selection process.
IVF is an expensive, accelerated process for accumulating genetic and epigenetic anomalies. It is not sustainable.
Most of the IVF is due to breeding too late. Some due to STDs like mycoplasma destroying the normal passage through the fallopian tubes. Or due to endometriosis. None of that is strictly genetic, most is bad nurture. So if the next generation no longer goes for bad nurture, these are one-off events.
Obviously, cystic fibrosis leads to sterility, but I haven't heard of many cases trying IVF for that reason.
Breeding too late is a really big deal that is swamping the remaining available health care supply, even when it occurs without IVF.
Look up the causes of endometriosis and ponder how these could affect the environment of a developing IVF baby.
There are hormonal anomalies and numerous other environmental factors which can effect genetic performance in multigenerational ways, even if the actual DNA sequence is unaltered.
Please also review what's going on with the actual supply of health care. Many of us are getting out in the effort to do no harm, and for numerous other reasons. It is no longer attractive to people with problem solving and diagnostic minds. Those that remain are getting very burned out.
Increased health care demand is being engineered into an environment of reduced supply and quality.
This is not sustainable.
I think people meeting in college and getting married is a mistake for a myriad of reasons with increasing the chances of autism in their offspring one of them. IQ is like height in that it is a good gene and it will get stronger over time. The problem with IQ and height is that at some point you get diminishing returns. My solid 125 IQ is just as good as the 150 IQs I was around growing up in a community filled with high achievers (if I told you what institution was nearby you would understand). So if a college graduate marries a non college graduate we will get more 125s and fewer 150s…but we probably have too many 150s and we don’t need more law school professors with 150s anyway. So 150s often want to be semi famous and not work too hard while being wealthy and so they are attracted to professions like lawyer that are pretty dumb professions.
With generative AI coming, there will be a premium on 130+ males.
Yes, the assortative mating theory is something I've seen play out in my own personal experience and I think it's actually a big driver. I also suspect that previous social systems did a better job of socializing mildly autistic people, helping them get over their problems and fit in normally. Sadly neither of these hypotheses are being pursued by RFK
Change requires a cultural and social change. There was greater social pressure to conform, a belief that children had to be molded by their parents and a subordination of the self to the community and its social norms. All of these have been undermined by liberalism and the focus on the individual over the community.
Being socially awkward and autistic is now an identity group with people celebrating neurodivergent people. It’s now who they are and they will fight tooth and nail against any effort to change “who they are.”
Yes, this theory could explain some of the effects. Do you know anything about the relationship between Asperger's and "true" autism?( do two high functioning Asperger's parents have a higher chance of having a kid with "true" autism than two non-Asperger's parents?)
At the rate with which Psychiatry is medicalising eccentric behaviours the day will come soon when having a "normal" child will be a statistical rarity.
Excellent analysis. It reminds me of the over diagnosis of Covid using the PCR test.
While I absolutely agree with the argument that raw data is hugely confounded leading to a dramatic increase of autism diagnoses independent of actual rates, it may be helpful to also note that maternal age at birth is a contributor to diagnosed autism, and maternal age at birth has been increasing since at least 1990 or so.
https://pubmed.ncbi.nlm.nih.gov/22525954/
https://ourworldindata.org/grapher/period-average-age-of-mothers?time=1922
"The study found that the incremental risk of having a child with autism increased by 18 percent for every five-year increase in the mother’s age. A 40-year-old woman’s risk of having a child later diagnosed with autism was 50 percent greater than that of a woman from 25 to 29 years old."
https://www.ucdavis.edu/news/uc-davis-study-confirms-link-between-advanced-maternal-age-and-autism
Noted here: https://www.cremieux.xyz/i/161498315/actual-causes
Also confounded by assortative mating, given that later childbirth (and less children) robustly correlates with higher education.
Yeah, I was wondering about this correlation as well, one reputed to be particularly significant when both parents have higher education featuring advanced mathematics and adjacent STEM disciplines.
I hadn't thought of that! Interesting.
Shoot. Sorry!
I wish there were a version of this written for the layman. Or an video version with illustrations maybe. There are several people I'd like to send it to but I can't imagine any of them making it more than a few paragraphs in due to the technical language. I'm not suggesting you should take on that responsibility, I'm sure this was weeks of work in itself, I'm just saying I wish a version existed that was specifically intended for people with an IQ of ~90. Thanks for writing it! If nothing else, I can reference it myself to mine for data when the topic arises.
This was a day's write, and I thought it was written for laymen.
I could try to write up a short summary that's even simpler.
LLMs do such summation well, though I expect you know this.
Hi Cremieuex, Thanks for the analysis. It sparked me to comment that increased awareness and media coverage of autism has also increased and this might have led to a significant shift in societal "attitudes" (for bad or worse) towards the condition. In Finland, for example, autism has become a more openly discussed topic, with a growing number of reality series and documentaries featuring autistic individuals (our "NPR" airs the show on TV on primetime). Same goes to ADHD, dyslexia, eating disorders, etc. (the latter might have even statistically increased, I'm no scientist so cannot comment). Anyway, my question: is this autism discourse part of an "awareness trend" or a cultural change towards "embracing value" in "human diversity"?
I think that's a large part of it!
"it is still to be lauded, because their lives still have value."
It says something rather dark about modern society that such a statement would be expected to result in condemnation and cancelled subscriptions rather than being widely regarded as praiseworthy (or at least unremarkable for stating the obvious).
Thank you. Autism runs in my family and I grew up in a medically fragile foster care home. I have several adopted siblings who are severely disabled by a variety of conditions. I love them dearly. The value of a human life should never be measured by IQ or functionality, we are all so much more than merely an equation of net economic contribution vs economic cost to society.
Huh, well thanks for this. I think I'm the target audience since I was concerned about the rise in autism and supportive of RFK. So you've made me more skeptical. Certainly the change in diagnosis is a big concern. I would like to see some type of cross age study here in the USA. I've got nothing against the Swedish study. But I do worry that we are eating more 'weird.. toxic' stuff here in the US and so it could be worse here. If there is some environmental effect. I've listened to RFK some and my impression is that if someone said to him, "You know we don't really know if autism is increasing." He would be for more data on the real trends. Thanks again for writing this.
Have long thought it was a combination of liberalized diagnostic criteria and older parents. But boy do people get upset when you mention the latter!
Thanks for the most comprehensive analysis of the supposed autism rate increase. This has become a charged issue, and most of the time, reason is a casualty.
"But is this a good priority? To determine why autism rates have increased, we first have to know if autism rates have increased."
Excellent point, and it is certain by any means that they have increased.
"In fact, all intertemporal comparisons of autism rates are biased for a simple but important reason: the definition of autism has changed!3"
And that is a significant fact to consider.
I sense there is no increase in autism over the years, and you laid out the logic for the conclusion. I am sure you will be inundated with indignation.
Your section on profound autism didn't answer my main question here: now or decades ago, people would notice someone was non-verbal and banging their head against the wall no matter what the diagnostic criteria. Do we have any data on the frequency of such people in, say, the 80s vs now? 25% of the exploding number of diagnoses are "profound" - even allowing that some of that is just "normal" low IQ, it seems like there is a real increase here. How many people were they hiding in those asylums?
Hiding in those asylums 20 years ago*
Back in the early '80s I hitched a ride with a woman who was going to visit her autistic son, in an asylum, who had been diagnosed and committed as an infant.
Spending time with my nonverbal autistic grandson, I run into other grandparents with their grandchildren on the spectrum, and the question is always the same, “Do you remember having classmates with these kinds of traits when you were a kid?” The answer is always, “No.“ I don’t think we remember having friends who had siblings in institutions either. My mother died last year at 99 years old and she also says cases of cognitive decline were very very rare in her experience. I think this kind of anecdotal data is excellent and I think the uptick in autism (along with all the other neurological diseases, such as MS, Parkinson’s, Alzheimer’s, ) is very, very real.
You might also have to consider that medical advances are keeping children alive who didn’t survive long enough to reach school age, and that families used to keep neurodivergent children that did survive hidden away. Finally there was far less reluctance to institutionalize before the emergence of what you might call the “inclusion” movement.
thanks. this is the experience of every 30+ year veteran school teacher. Something is amiss.
As a mother navigating the pre-assessment process for a four-year-old, I deeply appreciate your article. My son’s symptoms are so mild that, as parents, we sometimes wonder whether all this labeling is truly necessary. A private speech therapist even declined to continue seeing him, saying she didn’t want me to waste my money. Yet the school insisted on pursuing an assessment to secure extra funding.
It makes one question whether labeling a child with extremely mild symptoms truly helps them—or if it risks shaping their behavior and self-esteem in ways that might not be positive, while also straining the financial health of the system.
Making decisions and thinking clearly as a parent in such a complex and emotionally charged situation is incredibly difficult. I’m grateful to find such well-detailed information.
I've said it before, and I'll say it again; if we want a healthier population, we must dramatically increase infant and childhood mortality rates.
Morbid.
But you know I'm right.
Absolutely unneeded. We just need Galtonian eugenics. Pay the sickly to breed less, pay the healthy to breed more. And it's mostly about the latter since the TFR is sub 2.
the power of positive thinking
Well, if you get rid of all the sick people you will have a perfectly healthy population.
One of the great mistakes we made as a nation was the canonization into law (ADA) of the idea that accomodations are due to individuals based on their membership in some named category. Not only has it wrecked havoc on schooling and medical diagnosis it bears substantial blame for our oppression olympics culture by leaving anyone whose needs don't go into a labeled box out in the cold.
And this was never a sensible way to handle the situation. For instance, take accomodations in schooling. As someone who couldn't focus or remain awake in any lecture class I'm a firm believer in making sure no one is denied an education or a grade because of 'incidental' factors. But without an official diagnosis I had a few professors make life very difficult for me (if I'd gone to a less elite school I may not have graduated at all). And this just doesn't make sense, there is no need for an arbitrary cut off below which you are punished for a trait and above which you get special accomodation and it makes people resent disability accomodations because they feel they let people gain unfair advantage. For instance, in my case it is unacceptable for a professor to deny a student who shows themselves to have mastered the material via performance on exams, homework and any projects, a good grade because they didn't imbibe their wisdom in class. At some point students need to learn self-motivation and until then you can make it merely inconvenient not to attend.
We could instead identify what aspects of evaluation are truly central to the course -- those should never be relaxed for a disability. If it is the department's judgement that being able to complete problems in a limited time frame is an essential skill (bullshit imo) then you can no more offer extra time accomodations than you could accomodate muscular weakness in weight lifting). If it isn't -- which it never is profs are just too lazy to write tests appropriately (often me too) -- then anyone who wants that accomodation should be able to request it. Sometimes it may need to be discouraged and annoying to get (some ppl really do need laptops to take notes but making it a hassle can help discourage distractions).
This holds more generally. What businesses do or don't need to have by way of accessibility features is ultimately a function of cost, incidence in the population and alternatives not threshold effects. A better run government in a country with higher social trust would delegate such regulation to an independent expert panel to weigh costs and benefits -- but it's not feasible if it has to be in the executive branch. And yes that might include issuing little licenses for certain service animals (again no need to check the level of impairment only that the animal is professionally trained to the appropriate level and owner took a class to deter the people who don't really need the emotional support animal).
Instead we get worst of all worlds. Half the country thinks accomodations are a joke ruining the country while those with partial issues find it difficult to get partial accomodation. And no one actually does the cost benefit calcs to see what kind of improvements a buisness should need to make.
Your footnote 20 on ADDM data seems strange. You elaborate how profound cases may shift to non-profound as the child ages, but why would that undermine the reported growth, if it is measured at the same year (e.g. 8 year old) across time? An apples to apples comparison in https://pmc.ncbi.nlm.nih.gov/articles/PMC10576490/ ends up reporting
"During 2000-2016, the prevalence of non–profound autism increased from 3.94 to 14.26 cases per 1000 children aged 8 years; the prevalence of profound autism increased from 2.68 to 4.59 cases per 1000 children aged 8 years."
You do not cite such evidence in the post and delegate talking about the rise of profound cases according to ADDM data to a footnote, where you find ways to downplay it. You talk about different states having different diagnostic criteria, etc., and end with "This fact should be a lesson to those who cite the ADDM’s data as an indication of a “real” autism epidemic". What kind of lesson? That the data is safe to ignore? Isn't the onus on you to explain how "variation in reporting sources, practices, and policies" between states was likely to have led to a marked and sustained increase over time?
These seem like isolated demands for rigor. "Variation in reporting sources" is everywhere to be found. Generally the situation with data is complicated everywhere, including in stats that favor your story, but you do not pay such minute rhetorical attention to these issues then. E.g. you cite the Swedish data comparing symptoms with diagnoses without explaining that the symptoms data comes from phone surveys of parents of twins, while the diagnoses come from the national register of all children. Lots of reasons why the temporal behavior of the symptoms of autistic behavior in twins (reported by parents in a phone questionnaire, too!) may not reflect the entire population. You do not discuss the reasons or even the fact of the different populations, but find plenty to criticize in the ADDM data.