Sunday, 2 December 2012

The continuous nature of autism symptoms

The proposed revisions to the DSM re-definition of autism becoming an actuality are fast approaching over the clinical horizon as we are told that the die has now been cast. Although there is still considerable debate about what the changes will do to the numbers of people eligible to receive a formal diagnosis of the label autism as per previous revisions, one of the more well-received changes to the criteria is the formal acceptance that autism is a spectral condition, and that the severity of presentation of the soon-to-be dyad of core symptoms, reflects as much.

Prisms @ Wikipedia  
The spectrum concept, whilst not wholly supported by everyone, provides an intellectually satisfying description of the heterogeneity present in autism but also implies a few important principles must be accepted.

One of them is the assumption that for autistic traits to become clinically relevant, there must be a start point in the same way that the visible light spectrum has a very fuzzy beginning. Following this argument, and accepting that diagnosing clinicians are still as much artists as they are physicians when it comes to autism and other developmental disorders solely reliant on observation and developmental history, the question is: where do the symptoms of autism begin?

A recent paper by Kamio and colleagues* (open-access) looked at this question and came to some very interesting conclusions.

  • The name of the Kamio game was to look at the quantitative distribution of autistic traits in a rather large normative population sample of children (N=22,529) in Japan aged 6-15 years based on parental ratings of the Social Responsiveness Scale (SRS). There were also other data included in the results based on child psychiatric patients with an autism spectrum disorder (ASD) (n=257) and without (n=157) but receiving other diagnoses such as ADHD, schizophrenia and learning disability. And also a typically developing group (n=61) with no neuropsychiatric history. These latter groups were included primarily to validate and calibrate the Japanese version of the SRS. Normative data from a US sample used to validate the US version of the SRS were also included as some points.
  • Results: there were quite a few of them. The main message is, as per the title of this post, that results "add substantial evidence in support of the continuous nature of autistic traits in the general population" and specifically "there was no evidence of a natural cutoff that differentiated children categorically affected from those unaffected by ASD".
  • The male : female differences were also confirmatory that boys exhibit quantitatively higher autistic trait scores than girls as per other findings.
  • An interesting thought is entertained based on the capability of the SRS to distinguish autism from other conditions, "autistic traits, when present, exacerbate other types of psychopathology when they cooccur with autistic traits as comorbid conditions". In effect autism might magnify the symptoms of other comorbid conditions such as ADHD.

Some of this data takes me back to my Ph.D write-up days, and what I was once told by someone involved in my studies. It went something like this: autism is a distinct condition diagnosable on the basis of that triad (soon to be dyad) of symptoms, but most if not all of the traits present in autism are to some extent also present during typical development at specific periods. I must admit that at the time I had some trouble believing this, after all autism can be associated with some very extreme characteristics and symptoms.

As the years have gone on however, I've understood that those lining up of toys, spinning of wheels, flicking of fingers, even pronoun reversal and other communicative features, are present in typically developing children too at specific times of development; autism merely describes the unusual persistence of these behaviours or quantitative differences in their presentation. Of course this takes no account of symptoms which might not necessarily be totally psychological in nature, as per what can happen when gastrointestinal symptoms are present and potentially exert an effect on behaviour for example.

With all that in mind, it's not so surprising that there is no distinguishable natural cut-off point when it comes to autism or not-autism in those border regions outside of the lines in the sand that we draw, or rather our diagnosing clinicians and clinical symptom makers impose. I assume that's why we also have the concept of the broader phenotype further illustrating these fuzzy boundaries.

That being said Kamio and colleagues do suggest that "segments of the autistic continuum may be comprised of small clusters of discrete disorders". This point really intrigues me, in that they are in effect talking about more than one type of autism; a spectral type which basically runs from typical development through to autism, and a second type which includes a number of known conditions which manifest autism, such as Fragile X syndrome and indeed, that recent in-born error of branched chain amino acid condition. The rise of the autisms indeed.

I must finally pass some comment on that comorbidity suggestion which basically theorises that autism might magnify the effects of certain comorbidities. It really is an interesting finding which is deserving of a whole lot more study, particularly in light of how much increased risk of other comorbidity such as ADHD a diagnosis of autism can convey. Readers might know that I have a bit of a thing for comorbidity and autism on this blog, and indeed how tackling comorbidity - however this manifests - can sometimes impact on core presentation. Think dietary intervention as one example.


* Kamio Y. et al. Quantitative autistic traits ascertained in a national survey of 22 529 Japanese schoolchildren. Acta Psychiatr Scand. November 2012.

---------- Kamio, Y., Inada, N., Moriwaki, A., Kuroda, M., Koyama, T., Tsujii, H., Kawakubo, Y., Kuwabara, H., Tsuchiya, K., Uno, Y., & Constantino, J. (2012). Quantitative autistic traits ascertained in a national survey of 22 529 Japanese schoolchildren Acta Psychiatrica Scandinavica DOI: 10.1111/acps.12034