Monday, 23 January 2012

Can you grow out of autism?

An intriguing question to open this post commenting on this study by Heather Close and colleagues* on comorbid conditions and the stability of autism as a diagnosis. In essence the paper suggests two things: (1) autism is rarely a stand-alone condition; a recurring theme on this blog, and (2) autism is not a static condition (see my previous post on diagnostic instability).

To summarise:

  • Based on the 2007 National Survey of Children's Health, over 90,000 parents of children aged up to 17 years old were contacted. 
  • Of the total population, 1,366 children were identified via parental report as having either a past or current diagnosis of an autism spectrum condition. 453 parents said their child had a diagnosis of autism but didn't anymore.
  • Based on age divisions (young children, 3-5 years old), children (6-11 years old) and adolescents (12-17 years old), the authors analysed various responses of background information and the presence of comorbid conditions to ascertain whether there were any differences between those who had a current diagnosis of an autism spectrum condition (ASC) compared with those who had a past but not current (PBNC) diagnosis of an ASC. There were in effect 3 experiments running comparing those with a current diagnosis with those who 'lost' their diagnosis, across the 3 age groupings.
  • The findings: youngest children with a current diagnosis of an ASC were more likely (a lot more likely) to have a concurrent learning disability or delayed development compared with the PBNC group. In the 6-11 age bracket, current diagnosis children were more likely to have past speech and hearing problems and concurrent anxiety issues than the PBNC group. In adolescents, past hearing problems and current speech and epilepsy issues were more likely in the current diagnosis group. Currently diagnosed children across the age groups were also more likely to have more comorbidities than the PBNC group. 
  • The authors suggest that the presentation of these comorbid features, past or current presentation, likely influences whether or not a diagnosis is retained or 'outgrown'.

There are a few media reports and opinions about this work already. has the headline 'Why some children may 'grow out' of autism'.

Whilst very interested in these results, I find myself in a bit of a quandary. On the one hand is the pretty persistent line presented over the years that autism is a 'lifelong condition' which whilst exemplified by the changing/fluctuating presentation of symptoms as a consequence of things like maturation, in essence is immutable in terms of whether someone has autism or not. On the other hand, we have parents reporting in this study that some of their children having had an autism diagnosis, were no longer considered to have a current diagnosis. I suppose one could argue that the authors were not able to independently test these assumptions out either when diagnosis was originally given or after diagnosis was 'done away with' so a degree of subjectivity should be expected. One perhaps cannot also rule out the subclinical presentation of symptoms such like the ideas on the broader autism phenotype.

Having said that, quite a few people over the years have reported similar things in terms of their child no longer appearing to present on the autism spectrum. Few of these cases have been spontaneous; more often that not associated with some kind of intervention or following mis-diagnosis. I make no value judgements on either of these factors.

One of the authors of the current study, Andrew Zimmerman has quite an extensive autism research career which it has to be said, is perhaps slanted more towards the possibility that other factors/conditions/comorbidities seem to be associated with autism. So for example, familial autoimmunity and medical risk of autism, issues related to inflammation and neuroinflammation, and even a role for fever in abating some of the symptoms of autism. The current findings perhaps extend his interest.

What this latest study does reiterate is that a diagnosis of autism should not be the end-point to determining why a person behaves the way they do. Given recent posts on SPAD, CFD and all manner of other conditions potentially more prevalent in cases of autism (glutathione issues?), a diagnosis of autism should really be the starting point to try and ascertain whether these and/or other conditions contribute to the presentation of autism or at least whether they can adversely affect quality of life. Whilst token mention has been made about the impact of intervention on some cases of autism in this current paper, it does not seem beyond the realms of possibility that intervention might have had a role to play (with the caveat that much more targeted research is needed).

So the question remains: can you grow out of autism?

* Close H. et al. Co-occurring conditions and change in diagnosis in autism spectrum disorders. Pediatrics. January 2012.