Thursday, 23 January 2014

The potential impact of DSM-V on autism numbers

I don't really want to get too bogged down with the various discussions which have passed to-and-fro when it came to the expected impact of the introduction of DSM-V on autism prevalence estimates. I couldn't however let the publication of the paper by Matthew Maenner and colleagues* pass by without making some comment on it (or at least directing readers the various comments being made about it: see here and see here).

To summarise the new study: retrospectively apply the new DSM-V criteria for receiving a diagnosis of autism (sorry, autism spectrum disorder) to data pertaining to a cohort of children previously diagnosed with DSM-IV autism and see how much overlap there is. I think back to similar study when it came to comparing even earlier versions of DSM on autism prevalence estimates (see here) and the interesting effects that seemed to have.

The good news from the Maenner study: 81% of those DSM-IV classified as having autism were also DSM-V classified as having autism. Interestingly, also that those with comorbid intellectual disability or with a history of developmental regression were slightly more likely to meet the DSM-V criteria for autism. The not-so-good news: er, what happens to those children who didn't quite reach the DSM-V diagnostic thresholds..? Social communication disorder (SCD)? Well, we don't know yet because SCD wasn't looked at in the current study, and as far as I can ascertain, SCD is still wanting in terms of research into the diagnosis and indeed, outcomes. Just as an aside, I wonder if it might however also cover some of those unusual ICD codes relevant to autism?

Of course, as the various media on this paper point out, this was a study based on case notes or rather "clinician review of coded behaviors documented in children’s medical and educational evaluations" so one has to be a little bit cautious of making too many judgements. According to the Autism Speaks website there is further work coming to publication soon looking at more "in-person diagnostic evaluations" across both DSM-IV and DSM-V (see here). Perhaps some more concrete answers might emerge from that data; hopefully slightly more optimistic than the results from Wilson and colleagues**.

I'm gonna finish with a link to an interesting blog post by Allen Frances, who was previously Chair of the DSM-IV taskforce, titled: "Will the DSM-5 Reduce Rates of Autism?". According to the Maenner results, Dr Frances is correct in his assertion "Yes, by a lot" (or at least by about 20%). The arguments will no doubt rumble on about DSM-V (and don't even mention RDoC).

Update: 27 January 2014. And then came the paper by Kim and colleagues*** and some chatter about their findings. It seems that the SCD diagnostic coding would serve as a catch-all after all, for most of those not fulfilling the DSM-V criteria for ASD. All well and good I suppose but the questions about SCD remain in terms of what a diagnosis will actually mean for access to services and required provisions.


* Maenner MJ. et al. Potential Impact of DSM-5 Criteria on Autism Spectrum Disorder Prevalence Estimates. JAMA Psychiatry. January 22, 2014. doi:10.1001/jamapsychiatry.2013.3893

** Wilson CE. et al. Comparison of ICD-10R, DSM-IV-TR and DSM-5 in an adult autism spectrum disorder diagnostic clinic. J Autism Dev Disord. 2013 Nov;43(11):2515-25.

*** Kim YS. et al. A Comparison of DSM-IV PDD and DSM-5 ASD Prevalence in an Epidemiologic Sample. Journal of the American Academy of Child & Adolescent Psychiatry. 2014. 23 Jan.

---------- Matthew J. Maenner, Catherine E. Rice, Carrie L. Arneson, Christopher Cunniff, Laura A. Schieve, Laura A. Carpenter, Kim Van Naarden Braun, Russell S. Kirby, Amanda V. Bakian, & Maureen S. Durkin (2014). Potential Impact of DSM-5 Criteria on Autism Spectrum Disorder Prevalence Estimates JAMA Psychiatry : 10.1001/jamapsychiatry.2013.3893