Wednesday, January 23, 2008

Is Autism more common in girls than previously thought

I have just received an email that on tonight's nightline, there is going to be a program claiming that autism has been underdiagnosed in girls. This has also been posted on the Age of Autism web page. The literature has consistently reported a 4:1 ratio of autism in boys as opposed to girls. The topic of tonight's nightline show is not a spanking brand new idea. This has been a topic that has been proposed many times, often by females with autism. They claim that the 4:1 ratio is a myth. That girls are more sociable than boys. They do a better job of hiding their autism. Or diagnostic substitution exists and they are diagnosed as something else. This was also proposed by Ivar Lovaas and Tristram Smith when Ralph Boyd criticized Lovaas' famous 1987 study claiming that it was not a representative sample of autistics and that the bias of having so many girls in Lovaas' control group as opposed to the experimental group could account for the results of his study. This is because autism in girls is often more severe than in boys. It seems strange to see this post on the Age of Autism's web page and in the email list that the organization unlocking autism has. The reason it is strange is because this view is usually espoused by some of their biggest detractors, the anti-curebie neurodiversity people. The neurodiversity group is often very critical of the mercury causes autism crowd because they want a cure for their disabled children with chelation therapy and other treatments of extremely dubious value. I believe these parents are well-intentioned though misguided.

I suppose I am getting a little off-topic here, so back to the original topic at hand. One of the reasons this alleged argument is so interesting is that the neurodiversity proponents at times often seem to imply that they are representative of all autistics. Sometimes they speak for all autistics. This phenomena has been noted by Harold Doherty's The Royal We post at I notice that it would seem that an extremely disproportionate number, probably more than 50% of these people are of the female gender. One can look at the postings of neurodiversity people on the internet and see how many females there are. If the allegations that autism is more common in girls than previously thought is true then perhaps neurodiversity's proponents might be more representative of those on the autistic spectrum than I thought. If these people are mistaken, then the predominence of females in the neurodiversity movement would seem to me to be problematic if they claimed that they were any sort of "real voice of autism" as implies. It would mean they were less representative of autistics, so that their experiences would not mirror a typical autistic person, who would in reality have an 80% probability of being a male.

Though the internet seems to have a pervasive population of autistic females I know that in AGUA, the autism support group that I helped to start with Jerry Newport and some other people back in 1993 has at least a 10:1 ratio if not higher of males to females. This is greater than what is often presented in the literature. Though, some of the literature indicates that the ratio may be as high as 10:1 in the higher functioning groups as autistic females are often more severely afflicted than males as I mentioned previouslyThis is one thing that compels me not only to believe that neurodiversity is not a very representative sample of autistic people but that the arguments of Brenda Myles Smith and others who will appear on tonight's show are incorrect.

Michael Wigler, a geneticist who was written about in this month's issue of scientific american has a theory that females have a protective mechanism that would often make them the carrier for genes that could account for autism, but be less likely to give them the condition themself. Therefore, they could pass the gene on to their offspring but not end up being autistic.

Let's look at the diagnostic substitution argument that autistic females are often diagnosed as something else. I am curious as to what they are diagnosed as? Are they diagnosed having ADHD? Are they diagnosed as having dyslexia? What about stutterers? I suppose it is possible they will be more specific and answer the question on tonight's show, so I guess I will have to watch it to find out what the answer to that question is. After I watch the show, I may have to post an addendum to this post. However, this issue is of such great interest to me I could not curb my enthusiasm and had to post something about it today-before the show airs tonight. The problem with the diagnostic substitution argument is that not only has there been reported a 4:1 ratio of autistic boys to girls. The 4:1 ratio has also been reported in the above-named conditions as well. Here is one reference to the 4:1 ratio in stutterers

This would tend to negate the diagnostic substitution of argument of myles smith and others as this would mean there would be either parity in these other conditions between boys and girls or at least all of these undiagnosed girls with autism would end up being diagnosed as dyslexics, stutterers, etc. and the ratio would at least be substantially lower than 4:1 in these conditions. If these autistic girls were being diagnosed with something else there would have to be some sort of zero sum effect. The consistencies of ratios across conditions compels me to believe no such zero sum effect exists.

The next question is are dyslexia, ADHD, stuttering etc. underdiagnosed in girls also? How do the social arguments, ability to pass, less aggressive than boys arguments hold up for these conditions? What substitute diagnosis are these people given? This would also seem to negate that argument. How would female stutterers be able to hide their condition and pass for nonstutterers? Does this mean that reading problems in girls is not noticed. I am very curious to the answers to these questions?


Anonymous said...

Hi Jonathan, your new blog looks good and is easy to comment on. :)

I think Wigler's hypothesis makes sense. In general, there are sex-related differences in the genes that control social behavior, and it seems quite reasonable to suggest that male and female social behavior genes interact in different ways with autism genes.

If autism genes are dominant in males as Wigler indicates, while behaving like recessive genes in females, that would suggest a ratio approaching 3:1 as with Mendelian inheritance generally.

So it's possible that both sides of the underdiagnosis argument could be right; perhaps autism is significantly more common in males than females, but because there is some amount of underdiagnosis in females, the actual ratio is something less than the 4:1 that is often reported.

Cultural factors may account for the high percentage of autistic female bloggers. There are some folks who believe that keeping a diary or blog is a feminine activity. Perhaps other kinds of social interaction tend to appeal more to autistic men, such as online gaming or, as you mentioned, support groups.

jonathan said...

hi ventura, thanks for commenting and you do raise some good points. However, none of this accounts for similar ratios in the other developmental disabilities including stuttering. The fact that these ratios exist in these other conditions argues against cultural differences in autism and sociologic expectations from girls

Anonymous said...

I followed your link to the article about stuttering, and although it does refer to a 4:1 ratio for stuttering overall, it also indicates that the ratio is much lower for inherited stuttering and that males often become stutterers for nongenetic reasons. This is very different from autism, which is estimated to be over 90% heritable.

Although some biomedical enthusiasts have argued that males often become autistic for nongenetic reasons, they have no scientific evidence to back up that claim, and I don't believe there is any validity to it. Whatever is responsible for the male-female ratio in stuttering seems to be a completely different process than in autism.

I did not watch the Nightline program, but it's my understanding that the diagnostic substitution argument asserts that some autistic girls have been diagnosed with depression and/or anxiety because these are conditions more often found in females. Diagnostic substitution would be much less likely as to the predominantly male conditions of dyslexia, ADHD, and stuttering.

jonathan said...

hi ventura i re-read it and it seemed to say that there were two different types of stuttering, those that often spontaneously resolved because females were more easily able to overcome it and the genetic case in which it persisted into adulthood and there were similar ratios to autism, ADHD, dyslexia, etc. so that this form of stuttering had similar ratios. Also, it said that it involved a single gene so that the genetics could be more easily understood, the other forms of stuttering could also have a genetic origin involving multiple genes, but apparently there are types of stuttering that have that 4:1 ratio or close to it, so this would be a mechanism similar to autism.

jonathan said...

also there are other disorders such as dyslexia and ADHD which have similar ratios. At some point i may try to find links and post those.

Anonymous said...

Here's a link to a new study on ADHD that reports a male-female ratio of 5.7 to 1.