Archive | May, 2008

Thimerosal on trial- the incredible shrinking epidemic

13 May

The audio recordings of the first day of the thimerosal-only portion of the Autism Omnibus Proceedings hearings are now available here: ftp://autism.uscfc.uscourts.gov/autism/thimerosal.html. They are mp3 files.

Here’s some of what I heard yesterday via telephone and comments on what I think the parents’ lawyers seem to be implying now, maybe you will listen to the same discussion and take away different key points:

A lawyer for the petitioners (Mr. Williams, I think) said, as if a fact: there has been an autism epidemic, and he added that there is no such thing as a “genetic epidemic”.

They know this because no one could “miss” regressive autism in the past. I guess they might have missed other non-regressive autism and other ASDs.

The only kind of regressive autism they are interested in is the “clearly regressive” subtype, which they seem to be saying is about 2% or less of all ASD children born during the 1990s.

Apparently, they are only interested in the children of the “epidemic” era when kids got more thimerosal exposure.

There are so few of their target group that when these kids started to be “added” to the “epidemic” no one could see it happening, and likewise when the exposure to thimerosal dropped of precipitously, even though the numbers of these target group kids must have dropped off precipitously, no one could see that change in the larger epidemiological data.

So the epidemic might continue but it has nothing to do with thimerosal exposure now.

The numbers of “clearly regressive” autistics, however should be obviously diminishing. Because it’s a small group and not all of them “clearly regressed” following a vaccine containing thimerosal. These supposedly thimerosal-damaged clearly regressive kids must be disappearing by now, but maybe they’ve been replaced by kids who “clearly regress” due to another actionable agent. If they regress because of an non-actionable agent, like, say, oxygen or exposure prenatally to mom’s immune system, no one cares. Then logically, if all of the “clear regressing” autistics were caused to regress only by thimerosal, then there should be very few, or none, younger “clearly regressed” autistics in areas where thimerosal is not used for toddler age vaccines now and hasn’t been used in the past few years.

Apparently, they are claiming that thimeosal in vaccines only causes a subset of regressive autism, not including early-onset autism. So apparently there’s no way for a baby who got the birth dose of Hep B to be made autistic, since it can’t “clearly regress” shortly after birth. And if the baby only got the Hep B dose (if preserved by thimerosal), that alone couldn’t cause a regression months later. I think they are only interested in vaccines given right before a toddler regresses, at say age 12 months to 36 months.

Also, it seems that the PSC believes Eric Fombonne’s research is reliable when they want to make a point with it. They used his research to support the numbers of autistics who regress if I recall.

The transcripts will be available eventually (maybe soon), but we don’t know when. I think it would be interesting to compare get them to explain how many of this tiny group of ASD kids also have mitochondrial diseases or disorders. I wonder if they are trying to imply that the rest of the “epidemic” is caused by tuna mercury, chicken mercury or MMR, aluminum, assortative mating or what?

Autism Omnibus and shrinking hypotheses

13 May

The number of people who have made confident assertions about thiomersal causing autism over the last four years or so is astounding.

It’s now 2005…..[W]e should see fewer cases entering the system this year than we did last year.

– David Kirby

if the total number of 3-5 year olds in the California DDS system has not declined by 2007, that would deal a severe blow to the autism-thimerosal hypothesis…..total cases among 3-5 year olds, not changes in the rate of increase is the right measure.

– David Kirby

“Late 2006 should be the first time that rates go down” said Handley. “If they don’t, our. hypothesis will need to be reexamined.”

– JB Handley

…I would like to make a virtual wager that within the next 18-24 months scientific evidence will make the thimerosal-autism link a near certainty.

– Richard Deth, March 22, 2006.

All these statements have one thing in common, they promote the idea that mercury (thiomersal in particular) causes autism in either all, or the vast majority of cases.

However, listening to the Autism Omnibus yesterday provided a very interesting change from this perspective:

In some kids, there’s enough of it that it sets off this chronic neuroinflammatory pattern that can lead to regressive autism,” said attorney Mike Williams.

ABC.

Note the new language: ‘some kids’….’regressive autism’…..’can lead’.

It seems the days of ‘all autism is mercury poisoning’ are long gone.

Petitioners presented a very interesting expert witness yesterday – a Dr Sander Greenland from UCLA who is a Professor of Epidemiology.

Dr Greenland argued some strange facts for the PSC but completely in line with this new tack that I can’t even remember being argued in the Cedillo hearing (thiomersal may cause regressive autism in some kids).

Greenland essentially argued that all the current epidemiology regarding autism and thiomersal was not good enough to detect thiomersal causation in regressive autism – this is from his submitted report:

A simple example may clarify this point. If a vaccine is not associated with any type of disorder in the category, we should expect to see the same risk when comparing vaccinated to the unvaccinated. Suppose, however, that in reality the vaccine is associated with a two-fold increase in the risk of a type of disorder in the category, but not associated with any other type. Suppose also that, without the vaccine, the associated type represents only one-tenth (10%) of the disease category, and that the total number of cases in the category would be 100. Then, without the vaccine, the number of cases with the associated type would be 100/10 = 10. With the vaccine, however, the number of cases with the associated type would double, to 20, an excess of 10 cases over the original 10 with the associated type. This excess produced by the vaccine would result in a total of 100+10 = 110 cases over the full category, which is only a 10% increase in the risk of any type in the category. Thus, the risk ratio for getting any type in the category would be only 110/100 = 1.1. Such a small risk ratio cannot be reliably distinguished from 1 by ordinary epidemiologic studies.

In other words, the amount of autism caused by vaccines is in fact too small to be detected by epidemiology. If, of course, it is associated with it at all – a point made later by Dr Greenland:

The brief overview given above supports the idea that the association of MCV (mercury containing vaccine) with autism is small or nonexistent.

But really his point is that if thiomersal does cause autism (and whilst he professes to have ‘no opinion’ on the matter it may be telling that he refers to the idea as a hypothesis throughout his report, not a theory) it causes it in very, very small numbers indeed.

Dr Greenland passed no opinion the validity of the hypothesis itself. Rather, he was there to study epidemiology. We have to respect his opinion even if we disagree with it.

The more telling aspect for me was this sudden conversion from ‘vaccines cause autism’ to this suddenly tiny percentage – so small to be undetectable by epidemiology up to this point. That’s quite a step back. What will become of the Omnibus cases that are not considered’ regressive’? Or the ones (like Michelle Cedillo) who were claimed to be regressive but were, upon viewing the video evidence, clearly not. Are the PSC really throwing cases away?

The Autism Epidemic Meme is Behind Almost All Autism Woo: A Call for Additional Research

13 May

After learning former US president Bill Clinton had indicated he believes that “the number of children who are born with autism [is] tripling every 20 years” (hat tip Orac), an understandable reaction might be to point out his ignorance. Understandable, yes, but I think we are looking at a bigger problem than lack of scientific literacy or political pandering in this case; a problem that is going to have to be addressed in a manner that is clear and generally convincing.

A lot of the discussion in the autism community centers around the anti-vaccination movement. It is true that anti-vaccination could potentially become a major problem for the world as a whole, and it is also true that it is a source of stigma for autistics. Some of us have taken it to be our fight, even though it should probably be the CDC’s or the WHO’s fight, if they were not, as it seems, asleep at the wheel. Nevertheless, I think the persistent autism epidemic meme is a much bigger issue as far as the autism community is concerned. Not only is the notion of an epidemic stigmatizing, but it results in ideas that are more than just theoretically harmful to autistics, such as the idea that autistic adults don’t exist. These ideas will be around regardless of the existence of an anti-vaccination movement.

In my regular blog I have discussed the evidence against the notion of an autism epidemic at length. If I may say so myself, I might have even managed to half persuade a few people from the other side of the debate.
What I want to do here, however, is to essentially critique the evidence I’ve discussed thus far. Let me explain why.

Those of us who are immersed in scientific discussions involving autism are well aware, for example, of diagnostic substitution, of an apparently high prevalence of autism in adults, of the changing characteristics of autistics over time, of regional prevalence differences that resemble time-dependent differences, of the stability of cognitive disability as a whole, of the stability (even the decline) of institutionalization rates, of what went on in the past, and so forth. Taken as a whole, this evidence is overwhelming and convincing to someone such as myself who has studied and perseverated on it for years. Fundamentally, though, it is evidence that has a number of problems: It is too numerous, complex, disjoint and most importantly, lacking in precision; none of it is decisive on its own. We are talking about many bits and pieces of evidence that need to be put together and thought through in order to arrive at the conclusion that there is no such thing as an autism epidemic. I don’t expect someone such as Mr. Clinton to be aware of this evidence, understand it, or think through it, much less be able to analyze some of the publicly available data that is not yet available through the scientific literature.

You see, there’s no such thing as an IOM report on the autism “epidemic.” While I’m personally not that fond of basing my beliefs on what authority tells me I should believe, I think a pronouncement by major authorities on the matter would help inform the general public of the state of the debate and the evidence. For this, however, I believe additional research that specifically addresses the matter in a clear way is needed. Allow me to propose some avenues of future research that could potentially answer the remaining questions once and for all. I encourage readers to propose their own ideas.

1) Replicate Lotter (1967). We know that the prevalence of autism as currently defined is relatively high. We also know that the prevalence of autism as defined in the 1960s was relatively low (4.5 in 10,000). What we don’t know is whether the prevalence of autism ascertained using Lotter’s operationalized criteria and methods is still relatively low in 2008. I think it should be feasible to replicate Lotter’s methodology and criteria today and find out the prevalence, not of DSM-IV autism, but of autism as it was thought of in the past. Without meaning to be disrespectful, this should preferably be done while Lorna Wing is still with us. She claims to know which kinds of children Vic Lotter considered autistic and which he didn’t.

2) Determine the prevalence of autism in adults. This one is non-trivial, as there are some ethical issues to consider, but it seems they will attempt it in the UK. I hope it’s not another case of trying to find how many adults are diagnosed with autism or receiving services under an autism category. This wouldn’t teach us anything new and would just be fodder for David Kirby’s blog. I also hope they don’t assume all autistics must be psychiatric patients, for example. They should find a lot of autistics in the general population, and there is evidence they should find many who might not be diagnosable with autism despite meeting criteria, for various technical reasons. Of course, they also need to look in institutions and group homes, since a ready rebuttal will be that “low functioning” autism must therefore be what’s rare in adults.

3) Determine if regional differences in prevalence are real. When you study administrative databases in some detail, one thing that immediately jumps out is that there are huge disparities in the administrative prevalence of autism between certain regions, be it states, regional centers or counties. I have reasons to believe these differences are not real. If these differences are not real, I’d suggest it would be reasonable to hypothesize that time-based differences in administrative autism prevalence are of the same nature. I have suggested, for example, screening children with mental retardation from different regional centers in California to determine, at the very least, if there are real discrepancies in the prevalence of autism within the population with mental retardation. Another question that needs to be answered is why population density correlates so well with administrative prevalence (independently of things like environmental pollution, as I’ve recently found).

4) Explain the changes with a mathematical model. The plausible mechanisms that explain the rise in diagnoses of autism have been discussed at some length. They might include increased awareness, changes in official criteria, an increased availability of specialists, an increased availability of certain services, changes in cultural beliefs, and so on. I have even discussed the internet as a potential driving force behind increased awareness, particularly in the 1990s. But let’s face it, these are all essentially unproven mechanisms. No one has done a multivariate analysis that gives us a coefficient for each variable. Granted, some things are hard to quantify. It would be a lot like trying to quantify word of mouth. But some of this should be doable.

Tags: , , ,

Presto Chango

12 May

Now, there’s nothing wrong with making a mistake. Nothing at all. People make mistakes all the time – as I saw on the back of a window cleaners van the other day – ‘guano happens’.

For a trivial mistake (spelling etc) its easy to change things on a blog. I can simply edit and re-save the post. I don’t need to tell anyone my Bluto sized fingers have typed ‘teh’ instead of ‘the’ again. I can just change it and republish.

However, sometimes, you make a mistake that is rather more important. A mistake that changes the factual interpretation _and_ the tone of a post. These should be altered _and_ a little note be made close to the alteration to point out the error and the fact its fixed. Trying to get away with making such large scale errors and hoping no one notices is bad form.

If one is a journalist – a professional writer – you would expect the notification as a matter of course. Don’t journalists pride themselves on their accuracy and attention to detail?

So it was something of a surprise to see that the article Sullivan discussed written by David Kirby had undergone a mysterious and totally unremarked upon alteration.

This (click for larger image and then use your browsers ‘back’ button to return here after viewing) is the original post David made on the Huffington Post. The page was recovered from Google Cache. As you can see, this contains the erroneous ‘34,000’ figure and all that flows from it. The maths error that Sullivan noted.

However, visiting the Huffington Post post today reveals the following:

.

(Again, click for bigger).

As we can see, the post has undergone very significant change of a key part of important factual information. With no (that I can see) notification that the data has been altered. I took a screen shot of the entire page as of 09:43 on Mon 12th May 2008 and couldn’t see such a notification. Maybe someone else can see one?

Tut tut.

However, even more curiously, the same article David posted at the Age of Autism blog still contains the error.

(Again, click for bigger).

Why? Are AoA readers less interested in facts? Is David too busy packing for his trip over here in June (which I am _very_ much looking forward to by the way)?

Enforced Vaccination

11 May

I don’t like this, I really don’t.

I know I advocate for the undoubted and scientifically established benefits of vaccination and will continue to do so, but the news that the influential Fabian Society have recommended a policy of enforced vaccination is not good.

In an article for the Fabian Society, leading public health expert Sir Sandy Macara called for child benefit to be linked with vaccination uptake.

And Labour MP Mary Creagh said children should have to prove they are vaccinated before they start school to improve uptake of MMR.

Call for vaccine opt-out penalty

I’m all up for improving the uptake of MMR, I think that is a worthy and vital goal. But is this – educationally and financially punishing children – the right way to do it? Because make no mistake, the parents won’t particularly care that their kids are home schooled. And the type of parent who doesn’t vaccinate (wealthy, white middle class) won’t miss the child benefit. But the child at the heart of these penalties may well miss scholastic education. As a home schoolers ourselves (less through choice than lack of any other option) one of the things we are keenly aware our child misses is the company of her peers in an educational setting.

For those who don’t know, the Fabian Society is a ‘middle-left’ think tank that recommends policy to Labour Party members, particularly influential whilst we have a Labour government (as we do now). They reached this recommendation apparently after:

A poll by YouGov for the Fabian Society suggested that the public would back government action on MMR to address large rises in mumps and measles’ cases. It found that 63% of the public felt that immunisation only worked if everyone was covered, and only 31%felt if was purely up to families to make the choice.

MMR press release

YouGov are a well thought-of (in terms of results accuracy) market research agency. I’ve little doubt the figures they collected are correct. I still don’t like it though. I think that something needs to be done, but this? The penalties seem targeted to ‘hit’ the kids. It also seems tantamount to admitting that attempts to utilise the excellent, freely available science that has killed the MMR hypothesis is pointless.

I’m also frankly disturbed by this quote from Fabian review author Sir Sandy Macara:

One ought to recognise that mothers have a responsibility for ensuring their children are protected.

Mothers? Not parents?

This seems ill thought out, knee-jerk-ish and guaranteed to play into the hands of the conspiracy theorists. We need to do better – much better – than this.

Autism and Mental Illness

10 May

So, the family have been away for four days on holiday – our first ever holiday! A very, very good time was had by all 🙂

But in the meantime it seems like the Autism News Juggernaut hasn’t even slightly slowed. I came back to a deluge of emails on subjects touching on autism but one really caught my eye.

This is the story about autism being linked to mental illness:

Parents of autistic children are twice as likely to have had psychiatric illness, researchers have discovered…A child’s risk of autism was 70% greater if one parent was diagnosed with a mental illness, and twice as high as average if both parents had psychiatric disorders, according to a report in the Pediatrics journal. The finding suggests autism and psychiatric problems may sometimes have a common cause and genetic link.

I’m trying to get ahold of this paper to read for myself but its totally unsurprising to me that this should be the case. As some of you know I have manic depression (bipolar as its known in the US) for which I have been receiving treatment for approaching 30 years. I have long suspected that there is an overreaching link between many flavours of mental difference – a hypothesis, born out in the scientific work of David Porteous who has been involved in pioneering science regarding mental illness and DISC 1 mutations. Long term readers of this blog may know that DISC 1 has a high association with autism too.

Indeed, last year, David Porteous gave a fascinating talk at last years MDF Conference in which he talked about the DISC1 connection to manic depression and included ASD amongst the constellation of ‘mental disorders’ that have some kind of interrelationship.

So, this news was no surprise to me at all. Yet to some others it seemed as if it was a slap in the face. A comment from a reader who saw this item reported at CBC said:

So what is being implied here? That mental illness in parents is an indicator /cause of autism in off-spring, or autism in children causes mental illness for their parents? On behalf of parents of autistic children I feel offended by this type of garbage research…

Which is a frankly bizarre way to look at this study. The study itself seems to be saying only what is presented in its abstract.:

This large population study supports the potential for familial aggregation of psychiatric conditions that may provide leads for future investigations of heritable forms of autism.

Its step one. Nothing about _cause_ has been discussed as far as I can tell from reading the abstract. Does that make it ‘garbage research’? Hardly.

Adults, Autism and Scotland

10 May

I have been thinking recently how nice it is that the online autism community has moved on from the quarterly analyses of the CDDS data. For those who are blissfully unaware–the California Department of Developmental Services (CDDS) publishes statistics on the people it serves. They do this every three months.

These data are a favorite of people who would like to promote the idea of an autism ‘epidemic’. Mr. David Kirby has a book and enough power point slides for three debates which are filled with (mis)interpretations of these data.

For the past year or so, every three months the CDDS publishes the data followed by people stating, “The CDDS autism count has gone up, this proves there is an epidemic” and “the CDDS autism count has gone down, this proves the epidemic”. Both seemingly contradictory statements being made on the same dataset. These were quickly followed by multiple bloggers pointing out that the interpretations made were incorrect.

Three things happened that made for a break. (1) Mr. Kirby declared that he was moving on from autism, (2) the CDDS is on a break while they rework the way they compile the data and (3) A case was conceded in the Autism Omnibus which shifted the debate (and ended item (1)).

I was very happy to see the CDDS phase of the autism discussion end.

Then, much to my dismay, the same arguments started up again. This time it is data from Scotland, not the CDDS being misused. Otherwise, it is the same old arguments and the same bad analyses. Well…almost. Some new bad analyses have been added.

Mr. Kirby has a discussion of the Scottish data on another blog. Let’s avoid the conceptual mistakes (such as assuming that somehow everyone is properly identified and receiving services). Before we get to the real implications of this, let’s take a break and look at the math errors, shall we?

Mr Kirby takes this graph of data from the Scottish report:

And states:

Let’s look at the numbers. There are approximately 34,000 young people with autism in Scotland, born during the 16 years from 1987-2002. That is an average of 2,125 cases per birth cohort. But among older people, born during the 31 years between 1955 and 1986, there are only about 600 reported cases, or just over 19 cases a year.

Based on this, he has determined that if the true incidence of autism is constant, about 1 in 110 of the adults are missing from the count.

OK, go back and click on that image for me. I know you skipped over it, but, go take a look at the bigger version.

Did you see it? Yep, the number is not 34,000, but 3,400 adults with autism in the Scottish survey. A factor of 10. Don’t worry if you missed it. Mr Kirby (who spent some time ‘analyzing’ the data) and at least 20 people who responded to his post missed it too.

At this point, I can hear the screams of “So what, that’s just a small mistake. You are trying to distract us all from the big picture.” Because, in the end, even though Mr. Kirby is off by a factor of 10 and there aren’t 100 times more kids than adults receiving services with an autism label in Scotland, there is a roughly factor of 10 difference in the administrative prevalence of autism.

A factor of 10 is still big. I’d argue it’s huge. In fact, I’d scream right back at the people who are trying to use this for political gain.We can argue back and forth whether it’s real. But, consider some of the possibilities for the Scottish survey:

  1. the numbers are correct, all autistics are correctly counted.
  2. People are getting appropriate services, but some are under the wrong label (e.g. intellectual disability).
  3. Some people are getting the wrong services because of an incorrect label (e.g. schizophrenia).
  4. People who really should be getting services and supports are not getting any.

Let’s face it, if there’s a chance that people are getting the wrong services, we should be looking. And, yes, it is a very real possibility. Remember the big stink some people made when it was implied that some adults with the label of Schizophrenia might actually be autistic? Well, David Mandell is scheduled to talk about this at IMFAR this year in his paper: Evidence of autism in a psychiatrically hospitalized sample.

A £500,000 project to look for adults with Autism in the UK has been recently announced. To quote one of the researchers on this project:

“Adults with autism and Asperger’s syndrome are too often abandoned by services with their families left to struggle alone. Equally, people are frequently missaprorpriately referred to either mental health or learning disability services

“This study will inform the development of a national strategy designed to ensure that adults with autism and Asperger’s syndrome are supported to have full lives.”

“We still don’t know enough about autism, but we do know that left unsupported, it can have a devastating impact on those who have the condition and their families. One of the key gaps in our knowledge is simple – we don’t know how many people have the condition in any given area. That is why I am ordering a study to address this. “

It sounds like a really tough project. I don’t know if this study can really be accomplished. But, I have hopes that it could help improve the lives of adult autistics.

Now, as long as we brought up the Scottish survey, why not look at some of the details that were missed by others?

One question was how many of the individuals had “other behavioural or biomedical conditions?”. For the adults, this was 30% of the total. For the children, this was 3% of the total. Is this an indication that the kids being identified today actually have less severe symptoms than the adults? Even without that, only 1% have “other…biomedical conditions”?!? Where are all the kids with all the conditions like bowel problems that some groups claim define autism?

Another interesting fact from the survey is that over 50% of the children are in mainstream schools.

Yet another factoid: about 32% of the children in the survey have Asperger syndrome. Of those, half are listed as having ‘no learning disabilities’. For adults, only 14% are listed as having Asperger syndrome.

Surveys of those getting services are prone to a lot of errors–as has been discussed in the past for the CDDS data many times. So, these data should not be taken as hard epidemiological counts of the actual number of people in Scotland with autism. However, these data do not support the idea that the younger generation of autistics have greater challenges than the adults had to overcome.

I am actually glad that the subject came up of autism in Scotland. Why? Because in looking for some of these data, I found the website for the National Autistic Society Scotland. I particularly liked this page: I Exist: the message from adults with autism in Scotland.

For me, I have just finished a box of McVitie’s HobNobs while writing this. I don’t know if they are Scottish, but I love them. Perhaps I’ll open the other box to celebrate a study of adults with autism.

Additional

Sullivan’s catch of David’s maths error is good but I thought to myself as soon as I heard about this Scottish report that I’d heard about it before. I had. I blogged on this audit three years ago. One of the most fascinating aspects of the paper was when local authorities were asked for their opinions on the following question:

Research tells us that prevalence rates of autistic spectrum disorder represent an underestimate. To what extent do you consider the numbers above to be an accurate reflection of all those who live in your area?

The answers were very interesting. About 45% of the areas questioned said that the prevalence for adults was grossly underestimated, badly reported and that a lot of these adults exist without diagnosis. For example:

Argyll & Bute Council
It is believed that the figures represent a significant under-representation of those with ASD in Argyll and Bute. This was thought to be due to a historical under-diagnosis and the absence of clearly defined referral pathways and multi-agency assessment processes for adults.

East Renfrewshire Council, NHS A&C and Greater Glasgow NHS

…as a result of changing patterns of diagnosis over recent years there are likely to be substantial numbers of adults with ASD who are not known to services and are not diagnosed as having ASDs.

AYRSHIRE AND ARRAN
It is apparent that information collection and collation for adults is almost non existent.

DUMFRIES AND GALLOWAY
There is little doubt that this number is far short of the actual number of adults in Dumfries & Galloway with ASD.

GRAMPIAN
There is low diagnosis for longstanding clients, whom workers are aware have autism as well as a learning disability.

HIGHLAND
It is believed that these figures comprise a significant underestimate due to the lack of a diagnostic process particularly for adults. It is believed that the figures for younger children are accurate due to the development of diagnostic tools for children are accurate due to the development of diagnostic tools for children and the establishment of multi-disciplinary partnerships which include education.

LANARKSHIRE
The estimated numbers provided for the pre-school and primary school ages are thought to be a reasonably accurate reflection of the true picture. However the estimated number of secondary school children is less accurate and the estimated number of adults with ASD is likely to be a considerable underestimate of the true prevalence.

ORKNEY
Figures for children are an accurate representation of needs. One or two children may yet be diagnosed. Figures for adults are under estimated as diagnosis has not been made and access to specialists is variable.

Perth & Kinross Council
Figures for adults reflect the national findings that the numbers known to services/diagnosed represent a significant underestimate of those individuals likely to be affected. For example day centre managers locally consider a number of people to be on the spectrum who have had no formal diagnosis.

Pretty interesting stuff I think you’ll agree. This means that about 45% of the areas questioned said that the prevalence for adults was grossly underestimated, badly reported and that a lot of these adults exist without diagnosis. The two really stand out quotes for me were:

There is low diagnosis for longstanding clients, whom workers are aware have autism as well as a learning disability……day centre managers locally consider a number of people to be on the spectrum who have had no formal diagnosis.

So as well as the excellent points Sullivan raised, I’d also like to ask how it is possible to place any kind of interpretation of the data when the fact that adult prevalence is grossly under-reported is so well established?

Age of Autism Excels Itself

4 May

It’s my opinion that the blog Age of Autism has not ever once published a post that has contributed anything to the sum of human knowledge in a general sense, nor has it ever published a post that is designed to actually help autistic people live their lives.

However, every once in awhile, it publishes a post that is so monumentally stupid that I literally think the worse of myself for wasting time reading it. And here I am actually blogging about one. Sigh.

Such a post appeared today. It is entitled ‘CDC triggers measles outbreak’. The author of this post, ex-UPI journo Dan Olmsted says:

I’m starting to think we should rename the CDC the Centers for Disease Contagion. You’ve all seen the news that there are suddenly more measles cases in the United States and the CDC is blaming it in part on the increasing reluctance of parents to vaccinate their kids.

But it’s the CDC’s fault, and no other. Getting the “measles shot” means getting the MMR, and the MMR is “the autism shot” in the minds of many, many parents.

So, let me get this straight. It is the CDC’s fault that measles is making a return across the US? I see.

Its not, for example, the fault of the non-vaccinating upper-middle class soccer-mommies and daddies, for example:

Of the 64 people infected by the measles virus, only 1 had documentation of prior vaccination. Among the other 63 case-patients were 14 infants who were too young to be vaccinated. Many of the cases among US children occurred in children whose parents claimed exemption from vaccination due to religious or personal beliefs, or in children too young to be vaccinated.

Hell, no. _That_ couldn’t be the issue, right? Its obviously the CDC’s fault. Damn them for providing the vaccines and a schedule that has led to serious measles epidemics being held at bay in the US and the UK prior to the last 10 years of utter complacency and idiocy.

And why is Dan Olmsted happy to blame the CDC?

Let me tell you one reason why I’m not shy or circumspect about squarely blaming the CDC for this — because Jon Poling, Hannah’s dad, predicted something like this, or much worse, just a few week ago

And as we all know:

Dr, Poling is the real deal, educated at Johns Hopkins, devoted both to his daughter and his patients, tempered by reality. He’s mild-mannered. He’s mainstream. He’s credible.

Riiiiight. This is the same Jon Poling who was recently described by his co-authors as ‘muddying the waters’. The same Jon Poling who’s wife has been a subscriber to the vaccine hypothesis since at least 2001. The same Jon Poling who knowingly uses incorrect epidemiology.

I’m afraid that Jon Poling is right now in the process of extricating himself from the mainstream. And also from any concept of credibility. His refusal to approve access to information that would provide more accuracy to public statements members of his clique have made about the situation is testament to a man who is not governed by any reality other than a desire to push a pre-conceived agenda.

But really, the attempt to point the finger elsewhere by Dan Olmsted is nothing more than a childish ‘It wasn’t me! Its not my fault!’ when both logic and morality show quite clearly that if people decide to eschew something that might not only save their kids lives but the lives and/or well-being of the society in which they live then the finger of responsibility can only point in one direction.

Association Between Autism and Environmental Mercury Exposure Disappears Once Population Density is Controlled for

2 May

california-pollution-autism-analysis

[Correction 5/4/2008: Please see this comment. The trends and conclusions don’t change. The scatter of the graphs is not affected in a way that is noticeable, but the Y ranges do change. The adjustment formula also changes. See the corrected spreadsheet for details.]

This is a critique of Palmer et al. (2008), a recent study claiming to associate the administrative prevalence of autism in Texas school districts and proximity to coal-fired power plants, as well as mercury emissions. Normally I would just point out the likely problems of the paper, but this time I will go further and test a key hypothesis of my critique using California data in a way that is straightforward enough for readers to verify.

Background

Palmer et al. (2008) is not the first study of its kind. Palmer et al. (2006) claimed to document that “for each 1000 lb of environmentally released mercury, there was a 43% increase in the rate of special education services and a 61% increase in the rate of autism.” The more recent paper by Palmer et al. does not result in such remarkable estimates, considering its finding that “for every 1,000 pounds of release in 1998, there is a corresponding 2.6 percent increase in 2002 autism rates.”

Windham et al. (2006) is a case-control study done in the San Francisco Bay Area which claims to associate autism with emissions of Hazardous Air Pollutants (HAPs).

Then we also have Waldman et al. (2007), which I consider a study of the same type, except it associates autism with precipitation (as a proxy of television exposure) instead of environmental pollution.

My primary criticism of these types of studies is that they are attempting to find a cause for an epidemiological phenomenon that could very well not require an environmental explanation. That is, administrative data (special education data in particular) is not equipped to tell us if there are real differences in the prevalence of autism from one region to the next. No screening has ever demonstrated that substantial differences in administrative prevalence between regions are not simply diagnostic differences.

That said, the studies have been done, and they have found statistical associations. This usually means they either found a real effect or they have failed to properly control for some confound.

As I have noted repeatedly over the last couple of years, the glaring confound that most likely mediates these types of associations is urbanicity. The association between urbanicity and autism was documented even before these studies were carried out. It is plausibly explained by a greater availability of autism specialists in urban areas and by greater awareness in the part of parents who live in cities.

Palmer et al. (2008) does control for urbanicity, which might be one of several reasons why its findings are underwhelming compared to those of Palmer et al. (2006).

Is the control for urbanicity in Palmer et al. (2008) adequate?

There are two main problems with the control for urbanicity, described in the paper as follows.

Urbanicity. Eight separate demographically defined school district regions were used in the analysis as defined by the TEA: (1) Major urban districts and other central cities (2) Major suburban districts and other central city suburbs (5) Non-metropolitan and rural school districts In the current analysis, dummy variables were included in the analysis coding Urban (dummy variable 1, and Suburban (dummy variable2), contrasted with non-metro and rural districts which were the referent group. Details and specific definitions of urbanicity categories can be obtained at the TEA website http://www.tea.state.tx.us/data.html

.

1. It is too discrete. Within the set of urban districts, some districts will be more urban than others. The same is true of rural districts. Palmer et al. (2008) is effectively using a stratification method to control for urbanicity, but this method is limited, especially considering the paper looks at 1,040 school districts. A better methodology would be to use population density as a variable.

2. Modeling for distance. The paper models autism rates based on distance to coal-fired power plants. It follows that a control variable should model distance to urban areas rather than urbanicity of each district. Granted, this would not be easy because, as noted, urbanicity is not a discrete measure. But it needs to be noted as a significant limitation of the analysis. Consider school districts in areas designated as “rural” that are close to areas designated as “urban.” Such proximity would presumably provide access to a greater availability of autism specialists than would otherwise be the case.

California Analysis

This time around I thought it would be a good idea to run some actual numbers in order to test this population density confound hypothesis that up to this point has been simply theoretical. I will use county-level data from the state of California, which was fairly easy to obtain on short notice. The data used is the following.

  • Special education autism caseload data at the county level for 2005 was obtained from a California resident who had requested it from the California Department of Education.
  • County population and density data for 2006 was obtained from counties.org.
  • Atmospheric mercury concentration data was obtained from the EPA’s 1996 National Air Toxics Assessment Exposure and Risk Data for 2006.
  • All of the raw data, intermediate data, formulas, and resulting charts can be found in this spreadsheet which I am making available for readers to verify and tweak as needed.

Population Density vs. Autism

Autism prevalence was calculated by dividing the special education autism caseload of each county by its population (Column G). This is not a precise determination, of course, but it should not affect the analysis. In any given California county, the population under 18 is roughly a fifth of the total population of the county.

A first attempt at modeling population density vs. autism prevalence (Chart A) suggested the relationship was logarithmic. So I modeled log(population density) vs. autism prevalence, which resulted in the clear correlation you see in Figure 1 (Chart B).

Pop. Density vs. Autism Prevalence

Figure 1: Pop. Density vs. Autism Prevalence

–

This is as expected. You will note, however, there is one significant outlier in the lower-right quadrant. That is San Francisco county. Presumably, because of its peculiar geographic characteristics, its population density is the highest in the state. Nevertheless, San Francisco is an important data point since it is a significant urban area which happens to have a relatively low special education prevalence of autism. Let’s leave it in and see how it affects things.

I will use a simple standardization method of adjustment for population density. Basically, I will standardize autism prevalence in each county, such that population density is no longer a factor. Think of it this way. If the population density of each county grew such that its log were now about 3.5, how would we expect autism prevalence to be affected? The following formula is what I came up with.

Adjusted(Y) = Y + 7 – 1.93 * X

The fact that the adjusted prevalence (Column H) is not dependent on population density can be verified graphically (Chart C). Readers can click back and forth between Chart B and Chart C to better understand the effect of the adjustment. I will come back to this adjusted prevalence.

Mercury Exposure Concentration vs. Autism

I obtained atmospheric mercury exposure concentrations for each county from 1996 EPA data (Column I). More recent data would’ve been better since our population density data is from 2006, but it is not clear if newer data is available. I learned of the 1996 data because that is what Windham et al. (2006) uses. I’m working under the assumption that changes in population density in the last decade have been roughly uniform across the state.

Let’s first look at Figure 2 (Chart E), a graph of log(mercury exposure) vs. autism prevalence, without adjustment for population density.

Pop. Density vs. Autism Prevalence

Figure 2: Mercury Exposure vs. Autism Prevalence

–

There is a graphically noticeable trend in Figure 2, which is not surprising. The question is, does the trend remain after adjustment for population density?

Pop. Density vs. Autism Prevalence

Figure 3: Mercury Exposure vs. Autism Prevalence Adjusted For Pop. Density

–

Figure 3 (Chart D) is a graph of log(mercury exposure) vs. standardized autism prevalence; that is, autism prevalence adjusted for population density as previously calculated. In this figure we see there’s no longer a graphically discernable correlation between environmental mercury and autism. In fact, Excel produces a linear fit that indicates there’s somewhat of an inverse correlation between environmental emissions and autism prevalence.

Granted, if we were to remove San Francisco as an outlier, the trend would be pushed upwards. But then in this graph there appear to be two additional outliers in the middle upper part of the graph, Orange county and Los Angeles county. Keep in mind we have not adjusted for wealth. Regardless of how we might adjust the analysis, I fail to see that the graph would support a statistically meaningful association between mercury exposure and autism.

Further Confirmation

So far I have provided evidence that, in California, an association between environmental mercury exposure and autism disappears once we control for population density. This is clear to my satisfaction, but I thought it would be a good idea to attempt an inverse exercise as an illustration of the adjustment method. That is, let us try adjusting prevalence for mercury exposure, and see if the correlation with population density remains.

This is similar to what I did previously. A linear model is discerned from the correlation between log(mercury exposure) and autism (Chart E). This is used to derive an adjustment formula (Column K) whose validity can be verified graphically (Chart F). The new adjusted prevalence (Column K) is used in a new graph of log(population density) vs. autism: Figure 4 (Chart G).

Pop. Density vs. Autism Prevalence

Figure 4: Pop. Density vs. Autism Prevalence Adjusted For Mercury Exposure

–

What Figure 4 (Chart G) tells us is that even after we control for mercury exposure, there is still a clear correlation between autism and population density. In other words, population density wins bigtime – I believe that is the epidemiological term.

Conclusion

An analysis of California data suggests that correlations between the administrative prevalence of autism and environmental mercury emissions are fully mediated by population density. Palmer et al. (2008) suggests there is a real effect in Texas, but its results are not convincing primarily because its control for urbanicity is limited and inconsistent with the hypothesis the paper tests.

Tags: , ,

What Makes You Smile?

2 May

This post is dedicated to a wonderful woman I know on another site who is terminally ill. She wants to be remembered for being happy and to go out with lots of shoes and a bang. So this post is just about being happy. Specifically, some of the things that makes Tom happy.

Tom loves the colour blue

http://i117.photobucket.com/albums/o65/bullet046/Picture002.jpg

 

He loves icecream

http://i117.photobucket.com/albums/o65/bullet046/ghost028-1.jpg

 

And walks

http://i117.photobucket.com/albums/o65/bullet046/ghost030.jpg

 

He loves his little brother

http://i117.photobucket.com/albums/o65/bullet046/ghost035.jpg

 

And he likes water

http://i117.photobucket.com/albums/o65/bullet046/radio010-1.jpg

 

He adores lifts

http://s117.photobucket.com/albums/o65/bullet046/?action=view&current=cooking015.flv

 

And he loves singing and action rhymes. He also loves looking at himself in the mirror, which is what he’s doing during these three clips.

http://s117.photobucket.com/albums/o65/bullet046/?action=view&current=082.flv

 

http://s117.photobucket.com/albums/o65/bullet046/?action=view&current=083.flv

 

http://s117.photobucket.com/albums/o65/bullet046/?action=view&current=084.flv

 

 

← Older Entries
Newer Entries →