Researchers at the U.C. Davis MIND Institute has discovered regions in the state of California that have notably higher autism incidence. But the story is more complicated, and more sad, than one might think at first. Instead of indications of an “autism epidemic”, these clusters point to the fact that minority and poor children are much less likely to receive autism diagnoses.
I don’t have the paper yet (I’m still trying to find the abstract), but articles in the Woodland Daily Democrat and the San Diego Union-Tribune are reporting the story.
The clusters do not appear to point to environmental causes. Instead…well, read for yourself:
Researchers said that in this investigation the clusters probably are not correlated with specific environmental pollutants or other “exposures.” Rather, they correlate to areas where residents are more educated.
Children with autism diagnoses in these clusters are more likely to be White and have parents with high education levels. Again, a quote:
“In the U.S., the children of older, white and highly educated parents are more likely to receive a diagnosis of autism or autism spectrum disorder. For this reason, the clusters we found are probably not a result of a common environmental exposure. Instead, the differences in education, age and ethnicity of parents comparing births in the cluster versus those outside the cluster were striking enough to explain the clusters of autism cases,” said senior author Irva Hertz-Picciotto.
Kids in the “clusters” are about twice a likely to be diagnosed autistic and kids in nearby areas.
Twice as high.
To the many of us armchair epidemiologists who who have looked closely at the California Department of Developmental Services (CDDS) data, this comes as no surprise.
For me, the most memorable discussion of the autism clusters came from Autism Diva, in her post from July 1997, Malibu and Compton: Compare and Contrast.
Here is a graph from that post:
The South Central Regional Center, in a predominantly non-White, poor area of the Los Angeles basin, had an administrative prevalence of 33 per 10,000. Compare that to Westside Regional Center with a prevelance of 84. Westside is a much more affluent are with a higher proportion of White families.
From the San Diego Union-Tribune:
“There is mounting evidence that at least some of this clustering results from the greater access and utilization of services by those with more years of schooling,” the UC Davis researchers wrote.
Yes, there is a certain “I told you so” moment here. This blog, Autism Diva, Autism Natural Variation, Autism Street and others have been pointing out the apparent autism clusters in the raw CDDS data for years. Long before I started blogging. But the real story isn’t the effect such clusters have on the idea of the “autism epidemic”. Rather, this is a clear indication that we are underserving the disabled in our minority and poor communities. This is just plain wrong.
It is long past time for real autism advocacy organizations to work on increasing awareness and access to services in underserved areas. The autism “clusters” are probably not real. From where I sit, what is real are the “anti–clusters” of undiagnosed autistics, minorities, the poor, and, yes, adults.
Left Brain Right Brain 
From the American Journal of Epidemiology (2003) Article entitled “Infant Sleep Position and the Risk of Sudden Infant Death Syndrome in California, 1997-2000” by Li, Pettitti, Willinger, et al.:
Sleep position of Control infants
White Infants
62.2% put to sleep on their backs
Hispanic Infants
51.6% put to sleep on their backs
African American Infants
42.4% put to sleep on their backs
If we willingly accept that sleep position can have such a profoundly positive impact on SIDS rates can we at least consider that it may also have some profound negative effects also?
Infants who sleep supine compared to infants who sleep in the prone position are impacted in the following ways:
– Social skills delays at 6 months (Dewey, Fleming, et al, 1998)
– Motor skills delays at 6 months (Dewey, Fleming, et al, 1998)
– Increased rates of gastroesophageal reflux (GER) (Corvaglia, 2007)
– Below norm AIMS scores (Majnemer, Barr, 2005)
– Milestone delays (Davis, Moon, et al., 199
– Increased duration of sleep apnea episodes during REM sleep at both 2.5 months and 5 months (Skadberg, Markestad, 1997)
– 6% decrease in sleep duration (Kahn, Grosswasser, et al.,1993)
– 1 in 300 infants had plagiocephaly in 1974 (Graham, Gomez, et al., 2005)
– 1 in 60 infants had plagiocephaly in 1996 (Graham, Gomez, et al., 2005)
“Infants with deformational plagiocephaly were found to have significantly different psychomotor development indexes and mental developmental indexes when compared with the standardized population.”
Kordestani, et al. in their study “Neurodevelopmental Delays in Children with Deformational Plagiocephaly”
“There are indications of a rapidly growing population of infants who show developmental abnormalities as a result of prolonged exposure to the supine position.”
Dr. Ralph Pelligra regarding the impact of the Back to Sleep Campaign
http://cgi.thescientificworld.co.uk/cgi-bin/processHtml.pl?Id=2005.03.71.html&format=Dreamweaver
“Since the implementation of the “Back to Sleep” campaign, therapists are seeing increasing numbers of kindergarten-aged children who are unable to hold a pencil.”
Susan Syron, Pediatric Physical Therapist
“In its fundamental purpose it has been largely successful. The incidence of SIDS has been reduced dramatically. However, as many orthotists can attest, this important gain has not been without its lesser comorbidities. The one we tend to think of has been the rapid increase in the incidence of positional plagiocephaly and positional brachycephaly. However, there have been whispers and rumors of other effects.”
Phil Stevens, MEd, CPO regarding side effects of the Back to Sleep Campaign.
http://www.oandp.com/edge/issues/articles/2006-12_02.asp
The social inequality is the key factor.
Convergent data , see also this :
Texas Study Confirms Lower Autism Rate in Hispanics
“Autism could be under diagnosed among Hispanics, Palmer and his team note, given that these children are less likely to have health insurance and more likely to have trouble accessing medical care.”
http://abcnews.go.com/Health/wireStory?id=9441843
The debate ‘inequality link’ vs ‘environmental link’ seems to be nearly closed.
What is interesting in information genealogy is how statistics are used.
And information genealogy is the best way to avoid time vaste in false links.
LB/RB “underserving the disabled in our minority and poor communities. This is just plain wrong.”
“@ageofautism: Why do rich white kids have more autism? Does CAVIAR cause autism???”
“@ageofautism: Autism Centers spring up where there is autism and in areas where folks can afford the treatment. ARGH!”
Liz,
I gave myself a gift this Christmas–a few days not reading AoA!
My favorite historical cluster, for geographic reasons, is Brick NJ in 1998. The ingredients? A less than 5 year old IDEA legislation; add one heaping helping of an active special needs parent group called ‘Parents of Special Services and Education’, mix in complete cooperation of the schools, health care to records, and bake with the comparatively unlimited resources of the federal government to gather and analyze data. The result: a ‘cake’ that mirrored autism rates 10 years later.
That’s why I have some trouble with the ADDM data on an individual state basis. Too many ‘cooks’ making their cakes in different ways. Their data pieces may be the same, but the politics, demographics, psycho-social, and educational policies may be vastly different.
The question is how many of these minority and “under-served” kids are diagnosed with other conditions, other than autism. For example, we discovered in our area, white kids were diagnosed with autism, and minority children were diagnosed- with the same presentations- as mentally retarded. In special education, we had disproportionality with autism (percentage of white children high), mental retardation (black children high) and emotional disturbance (black and hispanic children high). The good news was that minority children were being given support services- the bad news was they were being mislabeled, and those labels can impact a child’s future throughout adulthood, changing what opportunities are made available and what services can be accessed.
Hi Sullivan –
I saw this, thanks for posting it.
Considering that there are many more poor people than rich people, what happens to our prevalance numbers once all of these people get evaluated and diagnosed? Even by only counting the rich people, we seem to have reached the number of one percent that is fundamental to the argument of a static incidence of autism.
– pD
That’s a good question. It will go over 1%, I believe, first of all.
Now, it’s true that administrative prevalence of ASD can be much lower in counties with low population density. I’ve analyzed this using California IDEA data, and the correlation is roughly logarithmic and quite good.
What would happen if all those counties started counting all autistic children? Probably not as much as might seem intuitive at first glance, depending on the state. Note that in California about 90% of the population appears to be an urban population.
Additionally, IDEA prevalence and CalDDS prevalence are nowhere near 1%. (This San Diego study, like all MIND studies, are probably based on samples of children from CalDDS.) IDEA reports available list autism when it’s a “primary disability” only. The phone survey uses a completely different counting method. The ADDM data is based on specific sites.
So basically, special education and service databases seem to undercount a lot more than other methods, and they have a lot more room for growth. I do think IDEA prevalence could in theory reach Minnesota 7 year-old levels across the US (1.4% or so), which would be a huge jump, but the increase won’t be as pronounced when you use other counting methods.
This isn’t surprising at all. Socio-economic class dictates that middle-class white Americans are going to get services and diagnoses while poor minorities are going to lack that.
Duh. It’s basically stating the obvious. Only people that can afford it, get diagnosed. So what are we going to do about it? Should we start planning out systems to get free diagnostic care for the poverty class? As well as systems in place for poor working classes to get services for their autistic child? Because here is another idea. If poor folks can’t afford diagnosis, what makes you think that they can afford ABA, OT or Speech therapy?
We what should we tell the government?
The UK doesn’t seem to be as good at collating this sort of data (taking Joseph’s points into account) but it should be noted that if I recall, the UK study that found a prevalence of 1% was concentrated in (I believe) a very affluent area of London. It would be interesting to see a study in an area with a much less well educated and well resourced area of London as well.
To further elaborate on pD’s question, I looked at some county-level IDEA data from a few years back (that I believe I got from Sullivan.)
Admin prevalence for the state as a whole was about 50 in 10,000. The most dense counties (LA area) have a prevalence of about 60 in 10,000, which is not that different than the state as a whole. This is even though sparsely populated counties have a prevalence as low as 1.5 in 10,000 (e.g. Inyo county.)
CalDDS is probably going to work in a similar manner. The most densely populated counties will dominate state prevalence, despite ridiculously low prevalences in sparsely populated ones.
To put it another way, if the researchers did this San Diego study in the Los Angeles area, their findings will probably not be as strong.
If all counties caught up with the LA area, prevalence would only increase 1.2-fold.
I need to correct that. It’s about 8 in 10,000 for Inyo, which is still quite low. I was looking at the wrong column. Alpine county had an IDEA prevalence of zero, but its population is only 1,261.
Does it ever occur to anyone that since much autism/Asperger’s is genetically based, that it is quite possible that some races may naturally have a higher incidence? We wouldn’t dare say the white community is underserved because they don’t have enough sickle cell anemia clinics, would we?
It is possible. For example, if I recall correctly, Dr. Fombonne found a lack of autism in the Innuit. The possible “Amish Anomaly” could be genetic, especially considering the dramatic rise in vaccination uptake in the past 20 years.
However, there are some studies which have discussed prevalence amongst racial groups. Those studies showed a lack of variation by the races they studied, as I recall.
But, if there are races that have higher incidence or lower incidence, why do they all seem to be increasing in administrative data? If
It is well worth studying if there is a variation by race. First, we need accurate numbers (which educational and services data like CDDS are not). Until then, the responsible thing to do is assume that the prevalence values are the same as the highest found. Seek out those who may be getting the wrong or even no support.
I think it should be higher priority to study if there is true variation by region. For example, do counties like Inyo really have fewer autistic children? This is the sort of thing the IACC should be doing as a prelude to environmental factor research. In other words, test the assumptions first.
First of all I do not agree that this year is 2010 (I don’t even agree it is a year necesseraly) But putting such pernicketiness aside I shall tackle Sullivan.
Fombonne would not recognise autism if he saw it face to face, the reason, it is a cultural construction.
It is no accident that the demographics of Kanners originals was very waspish leading to rather odd conclusions at the time.
Heck it is all cultural and it is no surprise that Kuru does not exist in the UK but new variant CJD does.
It used to be said that the best cure for Schizophrenia was a trip accross the atlantic (can’t recall which direction but I am not perfect)
If one believed the statistics one might well assume that in the UK there was something particular about Afro Caribean genetics that led to a higher rate (being cagey about incidence and prevelance here that ain’t relevant cos tis disputed) of Sz in that population.
Well it does not take an Einstein to say something is rotten in the state of Denmark there (sorry I forgot that Greenland is still part of Queen Margarethe’s domain.)
I suggest that Eric Fombonne is a latter day Lord Franklin to an anthropologist Amundsen, who could tell him that he wouldn’t know Inuit Autism if it shat on his shoes.
BTW the no autism amongst the Amish myth has been busted I believe.
Dr. Fombonne never published a paper about the Inuit, did he? I can’t find one, anyway.
If you recall, the preliminary finding was nothing more than a database search. In other words, the Inuit don’t get psychiatric diagnoses for autism that get coded in a database – that’s all.
I wonder what happened to that research. My impression was that there would be a follow-up effort with proper case-finding. Did it not get done? Were the results not as interesting as the researchers hoped? Is it still ongoing?
“Kids in the “clusters” are about twice a likely to be diagnosed autistic as kids in nearby areas”
This looks like it should be “than kids”.
I suggest an experiment to test a suspicion I have voiced before: Check whether poor areas with lower reported autism rates have correspondingly higher “retardation” rates.
I can tell you this is not the case in the San Diego RC. It is the case in other RCs. San Diego has a relatively low prevalence compared to LA, but the prevalence of autism in the population with MR is not low. It was 9.26% as of Q4 2005, which was comparable to LA RCs.
This means that the missing autistics are “high functioning.”
In CalDDS RCs, I’ve observed that recognition of autism in the populations with MR and without MR do not necessarily correlate. I have an analysis related to this. I see in it that the Central Valley RC is an example of an RC with very low recognition of autism in the population with MR (2.7% as of Q4 2005.)
Let’s not forget that clustering can occur because some districts provide better services (or, simply services) than others, so people move to those areas. In other words, they’re not sources of autism causes but of autism effects!
It appears that the study looked at place of birth for the clusters, not place of current residence. That should remove the possible bias you mentioned.
Hello friends –
This occurred to me in regards to the very strange Somali cluster in Minnesota. A lot of the African cultures are still highly tribally based, as was the basis of the genocide in Rhwanda. Anyways, it occurred to me that one possible reason we might see more Somalis in one area with autism, but not in others was that individuals from a certain tribe, with consequent genetic likenesses, relocated to Minnesota and encountered an environmental input.
– pD
At its most fundemental the problem is that people are using the wrong measurements.
The CDC is the wrong instrument for any objective estimates of prevelance in that it is not an academically designed instrument with checks and balances, it is a political statistic, and a heavily socially nuanced one as well, as has been clear for some time.
It is like trying to use a tachometer to measure the speed of a car when the speedo has failed, unless you understand the underlying mechanism and ratio of the gears rpm tells you nothing.
One thing is always the case, a small sample is always a skewed sample because it is not large enough to contain every eventuality that occurs in nature.
Given that autism as concieved statistically is a medical diagnosis in the context of a very haphazard system of social practice.
Current systems can’t even get a realistic prevalence of swine flu right because there isn’t independant confirmation that every case reported is not seasonal flu.
So much more difficult it is with autism, a socially defined condition without rigid boundaries, whose diagnosis (and the very word means opinion in itself) is driven by clinical opinion, based upon individual contexts and conditions that vary according to the families status and knowlege as much as the clinicians status and experience.
Autism itself is partly the fault, it is a syndrome, a collection of symptoms not a distinct entity with a discrete biomarker.
Put it another way, if I were to emigrate to California, and never seek a service, how would my UK diagnosis impact on the statistics if at all?
Another issue, which I’ve never seen discussed, is the effect of advocacy at local level. If the nationwide stats say 1 in 110 children have an autistic spectrum disorder, but in Lincoln county only 1 in 60 have been diagnosed, then one can infer that pediatricians may come under pressure of some sort (possibly even from their own desire to do the best) to lower the threshhold, or in some other way find children, so as not to leave unaddressed needs. This would potentially be a particular issue as the importance of early identification is increasingly recognized.
This, in turn, will push up the national stats, next time around.
Am I wrong to wonder about this?
I find it difficult to believe there could be any children with Kanner’s Autism unknown to the authorities under some guise or another.
All I can see is a mish-mash of different definitions, methodologies and screening instruments, with little standardisation across different studies and groups. It appears futile to try and extract anything useful that can be stated with conviction.
Passionless Drone, did it ever occur to you that the genetic basis of autism is but a hypothesis wanting any decent confirmation.
There is no absolute genetic code for autism, there are no consistent findings, only a range of studies indicating totally different morphologys.
This indicates if anything that there is no genotype of autism and that what is labled together as a “syndrome” has a variety of etiologies.
Does it never occur that looking for genetic causes for Cancer, there is no ‘cancer’, only cancers.
One is essentially looking at the genetics of Unicorns.
The differences we see in diagnostic rate are social constructs.
Indeed I believe that over time when the various genetic studies have either been replicated or not, it will become clear how much of a construct autism is, and we will be looking at it more in the way that one looks at the concept of race itself.
Autism is a complex phenomenon as complex as personality and a long way from being simplified or understood in medical terms.
You are all on the road to defeat. The tragedy is that people will be eliminated on the basis of the current unsound genetics.
“If the nationwide stats say 1 in 110 children have an autistic spectrum disorder, but in Lincoln county only 1 in 60 have been diagnosed, then one can infer that pediatricians may come under pressure of some sort (possibly even from their own desire to do the best) to lower the threshhold, or in some other way find children, so as not to leave unaddressed needs.”
Bit confused over the word ‘only’, since that infers that 1:60 is fewer than 1:110. Which it isn’t. 1:60 is approximately 0.01667 (4 s. f.), whereas 1:110 is approximately 0.009091 (again, 4 s. f.) … and the ratio of the two is approximately 1.834 (yet again, 4 s. f.) – which is a significant difference.
However, the inference regarding the difference could be correct, since almost twice as many children are getting the diagnosis than the overall national statistics figures would suggest. Where I used to live, the situation was – for a hell of a long time – quite the reverse of this: the diagnosis rate was practically zero. This wasn’t because there were actually no autistic children in that borough… it was because they had been avoiding diagnosing us!
I agree that socioeconomic factors may play a role in the failure to diagnose clusters of disadvantaged children. However, it may go beyond simple diagnoses. Recent research is beginning to identify prenatal folate supplementation as a risk factor for autism spectrum disorders.
Simply, folate supplementation allow embryos with deficiencies in folate metabolism to survive (they would have normally miscarried). These embryos develop, and are born. Once born, folate supplementation stops, and brain development is hindered. It just so happens that many individuals with autism spectrum disorders have an increased rate of mutations with genes involved in folate synthesis. Is it possible that affluent parents also have more access to prenatal vitamins? Lack of diagnosis of disadvantaged children plus an “environmental” factor such as enhanced folate supplementation of affluent parents could account for the clusters observed in the study.
The UCDAVIS publication :
http://referentiel-autisme.fr/#e20100105T070930
or the direct url but longer
http://www.ucdmc.ucdavis.edu/newsroom/newsdetail.html?key=3479&svr=http://www.ucdmc.ucdavis.edu&table=published
Hi MJC –
I had precisely this thought today. My wife was getting an ultrasound; her technician was a hispanic. We were talking and she mentioned that ‘hispanics just don’t get prenatal care as frequently’. It likely also means they aren’t getting lots of folate during pregnancy. It occurred to my wife a while ago that this was a big experiment; no doubt it has stopped spina bifida; but once the spinal column is formed, which happens relatively early in pregnancy, what are the effects of all this extra folate? We don’t know.
– pD
Passionless Drone, (and others)
You are at it again, confusing incidence of diagnosis for prevalence.
The rate of autism has not increased, nor is it I suspect different between Hispanics and Whites.
Autism is around the same prevelance as it always has been, and what is different is when, and whether it is diagnosed.
Hasn’t anybody got that yet.
We simply do not have this argument in the UK, it’s a non issue.
Laurentius Rex,
I don’t know where you get your certainty from. Autism researchers as a group certainly don’t have it. While I believe an increase in diagnoses has bumped the total number of autism cases over the past two decades, a quick read through the scientific literature shows that many qualified and respected researchers are pursuing environmental hypotheses regarding the rise in autism rates (folate supplementation being one of them). As is true in most controversial fields, one answer is hardly the only answer. There is always room for additional research into possible environmental factors. Do we just throw our hands up in the air and stop asking questions? That is not how science works.
There is a simple way to solve the dispute. Perform. the. experiment. Were the disadvantaged mothers in California taking an equal amount of prenatal folate as their affluent neighbors? If so, case closed. If not, there is still a question mark…
Hi Laurentis Rex –
While I can understand that you seem to relish your position of someone who thinks they know better than everyone and anyone else who researches autism, this seems a bit at odds with simultaneously being amazed that your statements are not being taken seriously.
It is possible to acknowledge that autism is a condition with extremely diverse manifestations and indeed causes, with the understanding that our current tools for measurement have serious imprefections; they are still the best tools we have available. I, for one, am pleased that everyone has not thrown their hands up and proclaimed that this social construct is far too nebulously defined to be understood effectively. You are free to continue lobbying for this; funny enough, it serves my purposes quite well, to my mind. Keep it up!
The fact that a discussion is had or not had in the UK is without utility in determining if it is an important discussion to be had or not.
– pD
Well yes I am certain, somebody has to be, and if I am wrong, then I am, but I am prepared to wait it out and see, and I don’t think I will be the one who is revising his opinions.
What I do predict is that there will be a new group of putative causes and the ones being debated here will have been put to bed as surely as Bettelheim has been.
You see what the scientists do is follow the money and there is a lot of money riding on causation, it takes dedication to pursue something you actually believe in, rather than continue an existing line of research simply because that is the only thing to do post doctoral. It is like everything social artefact.
Someday somebody is going to discover something new and interesting, and probably quite unexpected too but not biomed.
Its the unknown unknowns, all that is going on now is that there are so many rival road maps being printed, that is to say there is a vast amount of literature, but it will take time to see just how much of it is “noise”
The problem still is with the definition of autism and the failure to understand it’s wider connections, the science needs to step back and broaden out somewhat, rather than effectively defining it’s own outcomes in circularity.
Or to put it another way, it is like when all those crop circles appeared.
On the one hand you have all the new agers and ufologists (the equivalent of the anti vax brigade) blethering on about extra terrestial causation, the new age of aquarias or whatever, whilst the scientists, deeply perturbed by this desperately try to come up with explanations that fit within the current science, freak vortices, electrically charged particles, I dunno.
All the while a group of nocturnal ramblers are laughing in the pub with a plank and a bit of string beneath the table.
The scientists are trying to explain something that probably doesn’t exist, they have made the fundemental mistake in following the notion that there is an increase in prevalence before it has been proven to exist. They have jumped the gun.
“Well yes I am certain, somebody has to be, and if I am wrong, then I am, but I am prepared to wait it out and see, and I don’t think I will be the one who is revising his opinions”
–I get the feeling that it isn’t the following hypothesis: the increase in autism rates is caused by more effective diagnoses, that you are trying to prove – rather your superhuman ability to foresee the future…
“What I do predict is that there will be a new group of putative causes and the ones being debated here will have been put to bed as surely as Bettelheim has been”
–Yeah, that’s how science works… You create a hypothesis, try to disprove it, and, if necessary, move on to a new hypothesis. Very insightful “prediction”.
“You see what the scientists do is follow the money and there is a lot of money riding on causation, it takes dedication to pursue something you actually believe in, rather than continue an existing line of research simply because that is the only thing to do post doctoral. It is like everything social artefact”
–Scientists have a lot more creative control of the experiments they perform than you think. Do they have to write grants to get the money? Yes. Once they receive the money, however, they can perform risky experiments that the scientific community wouldn’t necessarily agree with. Do you really believe that all of the scientific advances over the last 100 years were detached from government funding? The structure of DNA? The development of the smallpox vaccine? The poliovirus vaccine? Artificial limbs and organs? Surgical and chemotherapeutic techniques to cure some forms of cancer? The entire human genetic code? You can thank the American taxpayer and the NIH…
“Someday somebody is going to discover something new and interesting, and probably quite unexpected too but not biomed”
— Are you kidding me? New and interesting things are discovered each day. Have you ever read a science journal in your life? Not biomed? What the heck do you even mean? Three-dimensional structures of viruses, human proteins, bacterial components, etc. are crystallized and solved each day. This data is new, interesting, and is changing the world as we know it. You can thank “biomed” for that! Why would the autism field be any different?
–I have to admit, I’m a molecular virologist who works from cash given through NIH grants. While I don’t specifically work on autism, I am insulted that anyone would write off the professional scientists in that field. For what gain? No scientist ever gained anything by sticking to a specific hypothesis– refusing to engage in an open and objective debate. Maybe that is the issue at hand. Perhaps Laurentius (and others) aren’t trained scientists. If true, why would Laurentius spend so much time on an autism blog? Rather than respectfully being a part of the dialog, it seems that Laurentius has an agenda to push. If not true, I doubt Laurentius’ scientific career has progressed much…
Of course I have an agenda to push, it is a disability rights agenda, an agenda to be treated ethically and respectfully by medical science, something that does not happen a great deal.
In this country the usual way into a academic research career is to compete for funding from the appropriate board, usually on research topics already set by the University, unless one can bring ones own funds into it.
I have already stated somewhere I am sure that my field is educational research, now methodology is just as important, the rules are the same, but the experiments are not. On the way one learns of the impact of sociology and social psychology on the workings of the researcher, nobody is free from pressures or influences of some sort.
Whether or not I can predict the future is neither here nor there, what I do, like anybody else is watch trends.
The point to be taken is that for a long time in the UK and elsewhere in Europe, it has pretty much been accepted that the autism rate is much higher than has been orthodox thinking in the US. Indeed a start (albeit an uncertain one) has been made on adult epidemiology.
If it is established, as I am expecting it will be going by past trends, the work of Wing and Gillberg for instance, that the rate of autism amongst adults is as great, or very close as makes no difference within a reasonable margin of error, I am afraid all this search for new causational factors will be seen to be a little ridiculous.
If there is an environmental factor, something I don’t rule out, I think it will be found to be something that has been around a very long time.
I have a hunch, that is where we all start, that hunch becomes a hypothesis, and if one cannot disprove it, it remains more likely to be true than not.
This is a testable hypothesis but the funding to test it is not as forthcoming as it ought to be, that is a political and a social factor.
I do happen to know, from political circles that the funding for adult epidemiology was botched when the UK government made it’s committment, and what has appeared so far is but a pilot.
“If it is established, as I am expecting it will be going by past trends, the work of Wing and Gillberg for instance, that the rate of autism amongst adults is as great, or very close as makes no difference within a reasonable margin of error, I am afraid all this search for new causational factors will be seen to be a little ridiculous.
If there is an environmental factor, something I don’t rule out, I think it will be found to be something that has been around a very long time.
I have a hunch, that is where we all start, that hunch becomes a hypothesis, and if one cannot disprove it, it remains more likely to be true than not.
This is a testable hypothesis but the funding to test it is not as forthcoming as it ought to be, that is a political and a social factor.
I do happen to know, from political circles that the funding for adult epidemiology was botched when the UK government made it’s committment, and what has appeared so far is but a pilot.”
Thank you so much for your erudite views Rex.
There is a simple way to solve the dispute. Perform. the. experiment. Were the disadvantaged mothers in California taking an equal amount of prenatal folate as their affluent neighbors? If so, case closed. If not, there is still a question mark
I am not surprised that you picked something which raises a question mark, by your own definition.
In a recent study in Texas “Folic acid awareness and supplementation among Texas women of childbearing age”, they found that during pregnancy about 40% of non-Hispanic White mothers took daily folate. By comparison, about 24% of Hispanics did so.
Does this really tell us something?
Here’s a better question–do people of any ethnicity use prenatal supplements more than those outside of a cluster?
There are a million or more such questions that can be posed. I hope that the epidemiologists are spending the time to ask smart questions rather than taking a simple Edisonian approach.
People who actively advocate for environmental causation research need to take this study seriously. If there are environmental causes to be found it isn’t going to happen without knowing and controlling for factors like education and access to services.
If they are sincere in their advocacy, they should welcome studies like this and Bearman and King’s recent study.
“There are a million or more such questions that can be posed. I hope that the epidemiologists are spending the time to ask smart questions rather than taking a simple Edisonian approach”
Edisonian approach? There are plenty of well-respected studies which support folate as a contributing factor to autism. Autistic individuals are more likely to have defects in genes related to folate metabolism for instance… Folate has also been shown to epigenetically modify the embryo. I don’t know if you can call a hypothesis with such support from the current literature Edisonian. With such support, what makes the “folate hypothesis” less smart than another?
Hi MJC & Sullivan –
It occurs to me that maybe the CHARGE or MARBLES studies currently underway could give us some insight into this. If I understand correctly, they are gathering markers from mothers and children from families that already have a child with autism. Perhaps maternal folate levels are included in that list of values being analyzed for.
– pD
It’s the same as the rise of diagnoses, isn’t it? It’s like an instinctive reaction. People think “if there are clusters made up of wealthy people, what is it that the wealthy people are doing to create autism?”
Thinking like that probably leads nowhere but dead ends, unfortunately. Hypotheses based on that sort of idea can’t explain certain key observations, e.g. why is the prevalence of ASD in a semi-urban location of Sri Lanka over 1%? (When it was probably zero before any autism screening was done there.)
To demonstrate that autism is actually associated with affluence, you can’t just select groups of diagnosed autistics. I’m getting a little tired of pointing this out, actually. You have to find all autistics in a population, diagnosed and undiagnosed, and then check for levels of affluence or anything else.
Hi friends –
I got an email from the MARBLES team this afternoon; maternal folate levels are being taken into consideration but it will be a few years before enough data, and presumably age appropriate children will be available for analysis.
– pD
Hi Joseph –
I’d be curious on your take of the Faroe Island study, wherein every child in school was evaluated and a prevalance of .56% was found.
http://www.springerlink.com/content/b758767u364g5255/
This study got thrown around a lot when the mercury hypothesis was in vouge, but I don’t recall anyone pushing Wing as evidence that their methods were flawed because they must be missing one half of the people on the spectrum. I haven’t read the study, just the abstract; how did their sample size compare to the four diasgnosed children in the Sri Lanka study?
But thinking about the tools in Sri Lanka gives me pause as towards if we should consider this a key observation or not. They only identified four kids with autism, and all were aged two or less. If one of them lost their diagnosis over time, something that gets bandied about when expedient, our prevalance in Sri Lanka plumets to ~ .75%. Aren’t we subject to large jumps in conclusions based on very small absolute changes with samples this size? On the other hand, however, if we want things to match up with Baron Cohens findings in adults, even a single loss of diagnosis hurts quite a bit.
– pD
This social class association is very un-new. See my discussion of the stark social class bimodal distributions already found 40 years ago, in my 1993-published paper linked at top of my blog. That was before the autism increase, and this report now is after it. So even after this supposed explosion of awareness, the level of unawareness remains in place? That seemingly does not make sense. Conclusion: –as I had concluded decades ago– autism is partly mainly genetic and partly mainly environmental.
Startlingly not news, Robin. Never mind autism there is an association between socio-economic class and just about any medical condition you care to mention, going back to the Black Death in the 14th century and I dare say even further to whatever it was used to bother the ancient egyptians, pyramid builders back maybe.
All the same you only have it two thirds right, part genetic, part environmental and a very big part socially constructed.
Put it all together and it is all inevitable anyway including the responses to it.
We not only, only see what we look for we only see what we want to look for and I bet Donald Rumsfeld wished he had said that 🙂
“Any medical condition you care to mention….”. Indeed, but isn’t it usually biased towards the lower classes? The only other upper-class illnesses I know of are manic-depressive, gout (in historical connection to meat-eating), and caviar allergy. Oh, and allegedly dyslexia.
Laurentius Rex wrote:
‘One is essentially looking at the genetics of Unicorns”.
Great quote, can I use it? In line with your observation, Moss and Howlin recently revisitd the genetic syndromes and ASD and found that the genetic syndromes have distinct and meaningful differences beyond ‘Autism’ and provided good evidence that the association between the genetic syndromes and ‘autism’ is superficial and related to the degree of cognitive impairment. Cognitively impaired people share one or another symptom with ‘true’ autism and researchers have never been able to control for cognitive impairment in the genetic syndromes.
Take way the single gene disorders (Tuberous Sclerosis, Fragile X, Angelman’s Syndrome, 16P mutation, Rhetts Syndrome, Downs Syndrome etc. ) and all that the researchers are left with is ‘One is essentially looking at the genetics of Unicorns”.
http://www.ncbi.nlm.nih.gov/pubmed/19708861?
Not to minimize specific examples of helping specific people but this is another good example of the broader problems created by supporters of and the current leadership in the Congress, Senate and Presidency. The more things are done to “help the poor” the more “poor” is perpetuated. More help is available when a broader base of success is prevalent in the country and more people will benefit. When people are more free they are more successful and more people benefit broadly.
Besides doing specific things please consider the broader concept as well so more of us can help or benefit.
This is beginning to get a bit political.
I think what you find is that the more the rich help themselves the more “poor” is perpetuated.
Wealth does not trickle downwards, it pools in the rich districts and creates ever increasing barriers for people in poor neighbourhoods.
Look at EEG’s of infants who sleep on their backs compared to infants that sleep on their stomachs. Something as simple as a change in sleep position causes huge physiological changes. But, just because it’s not a drug or a pollutant or “genetics” it isn’t analyzed as a possible cause of autism.
If a known vaccine or pollutant caused an increase in sleep apnea, lowered subcortical arousal thresholds, obliterated Stage 3 NREM sleep, obliterated stage 4 NREM sleep, increased the number of infant arousals, and was also shown to cause increases in plagiocephaly, torticollis, hip subluxation, and strabismus there would be people screaming in the streets. But, since the back sleep recommendations are mandated by the medical community it’s ok. Just like thalidomide the long-term implications of the back to sleep campaign have never been analzyed. Never. People in Asia have put their infants to sleep on their backs for generations but that is a “cultural practice” which is far different than a medical recommendation. Just food for thought.