Saturday, 14 June 2014

Optimal outcome and autism: a role for intervention?

Optimal outcome and autism.

Not here you don't @ Wikipedia 
I'm sure most people with an eye on autism research and practice will have come across this issue. The idea is that in amongst the various developmental trajectories which are being realised in these days of autisms over autism, there may be some children previously diagnosed with an autism spectrum disorder (ASD) who move outside of the diagnostic boundaries of the condition.

There is still lots of debate about the hows and whys of this scenario [1], and in particular, whether leaving the autism diagnostic label behind means everything about autism is left behind (see here). But certainly it is getting more and more difficult to dispute the fact that [some of] the signs and symptoms of autism are anything but static [2] and for some at least, the diagnostic label of autism might not be as lifelong as previously thought. Even the recent IMFAR conference saw some continued interest in this issue (see here).

Today I want to introduce a couple of papers which add to the literature on optimal outcome (OO) and autism, and in particular move away from just description to looking at what factors might potentially influence optimal outcome and specifically a possible role for intervention. You might consider this post an extension of an entry not-so-long-ago (see here) talking about the findings from Deborah Anderson and colleagues [3].

The paper by Nahit Motavalli Mukaddes and colleagues [4] (open-access) adds to the increasing literature on optimal outcome and autism with their assertion that: "High IQ and the development of communicative and language skills at an early age could be the most powerful factors contributing to an optimal outcome". The paper is open-access but a few points are noteworthy:

  • Based in Turkey, this was a descriptive study of some 39 children "who previously received a diagnosis of ASD and who did not meet the criteria for any ASD in the final examination". Importantly, the authors held pretty detailed records on this group based on their initial assessment and diagnosis for autism including "an in-depth psychiatric examination of the child" which meant plenty of face-to-face contact accompanied by an analysis of medical history. 
  • They also followed children "every 3-4 months" following their referral to an education program "inspired by Pivotal Response Training (PRT)". Most children followed PRT although 2 children "were able to attend ABA [Applied Behaviour Analysis] programs with frequencies ranging from as low as 8 hours per week up to the recommended 20 hours per week".
  • Optimal outcome was specifically defined in this study based on the paper by Helt and colleagues [5] slightly modified. This included both abatement of autistic symptoms and IQ also being measured in the 'normal' range.
  • Results: Well, aside from talking about communication and IQ as being potentially important factors related to optimal outcome: "The time from baseline to optimal outcome was 2.71 ± 1.76 years (range: 0.5-8 years)". The mean age at optimal outcome was round about 5 years old although there was some variation around these variables.

The authors provide quite a bit more detail on their study and why their results might be considered credible. I note they also suggest that comorbid medical disorders such as epilepsy were largely absent from this group and might "be another factor that influences outcomes". I'd chime in here and agree with that sentiment in light of what we know about the autism-epilepsy relationship (see here) and specifically the growing realisation that IQ or cognitive ability seems to be an important factor, confirmed by Jokiranta and colleagues [6]. The issue of intervention type and specifically whether something like ABA might play a role in optimal outcomes leads me on to the next study.

The paper by Alyssa Orinstein and colleagues [7] adds even further to the collected literature in this area, daring again to talk about what intervention might show involvement with a small group of optimal outcomers compared with those categorised as being high-functioning autism (HFA). Deborah Fein, the godmother of optimal outcome is also a co-author on this study.


  • "The current study examined intervention histories in 25 individuals with OO and 34 individuals with HFA (current age, 8-21 years), who did not differ on age, sex, nonverbal intelligence, or family income. Intervention history was collected through detailed parent questionnaires". 
  • Those letters A-B-A were again mentioned in the results: "Substantially more children with OO than HFA received applied behavior analysis (ABA) therapy, although for children who received ABA, the intensity did not differ between the groups".
  • Likewise that issue of comorbid conditions likely rears its head in this study too: "Children in the HFA group were more likely to have received medication, especially antipsychotics and antidepressants".
  • One final quote to make at the risk of plagiarising the whole abstract: "There were no group differences in the percent of children receiving special diets or supplements".

ABA seems to do rather well in both the Mukaddes and Orinstein studies in relation to optimal outcome. I know that as an intervention it's not everyone's cup of tea, particularly with some rather chequered history, but there are 'moderates' out there who support the principles of ABA (see here) and the evidence base is not bad compared to lots of other interventions put for improving outcome in relation to autism. Without nailing my colours to any particular mast, I'd echo the sentiments of Orinstein et al and their suggestion of more to do - prospective studies - in this area. I might also draw your attention to other findings on PRT and ABA which might also be pertinent [8].

I'll also pass some comment on the use of the words 'special diets' when it comes to the Orinstein paper. I assume special diets includes things like the gluten- and casein-free (GFCF) diet and goes some way to addressing the notion that the optimal outcome group are not over-represented by those following such an intervention. I'd agree with that sentiment knowing what I think I know about this particular intervention. There is of course another way of looking at this insofar as there being best responders to something like a GFCF diet included in the optimal outcome group... but let's not get too carried away just yet.

'Optimal outcome' with autism in mind still has the ability to divide opinion. Reading again the editorial by Sally Ozonoff (see here) which accompanied the original Fein study and the description from Uta Frith on "a variant that is temporary" (see here) makes me realise that sometimes the bigger job of science is not actually doing and reporting the work, but altering long and often passionately held views and opinions as a result of that new knowledge. As intimated on a previous post talking about the economics of autism (see here), although I am not a great fan of talking about 'what autism costs' and the potentially unhelpful way that headlines like 'Autism costs '£32bn per year' in UK' can be construed, the findings from the Mukaddes and Orinstein studies should perhaps be the topic of further autism research. Not only because they might further highlight who is most likely to fall into the OO category and how they might arrive there but as per the paper by Barrett and colleagues [9] the onward implications to public finances with the aim to: "leverage investment in education and intervention to mitigate aspects of autism spectrum disorder that negatively impact individuals with the disorder and their families".


[1] Bölte S. Is autism curable? Dev Med Child Neurol. 2014 May 20.

[2] Wodka EL. et al. Predictors of phrase and fluent speech in children with autism and severe language delay. Pediatrics. 2014; 131: e1128-e1134.

[3] Anderson DK. et al. Predicting young adult outcome among more and less cognitively able individuals with autism spectrum disorders. J Child Psychol Psychiatry. 2014 May;55(5):485-94.

[4] Mukaddes NH. et al. Characteristics of Children Who Lost the Diagnosis of Autism: A Sample from Istanbul, Turkey. Autism Res Treatment. 2014: 472120.

[5] Helt M. et al. Can children with autism recover? If so, how? Neuropsychol Rev. 2008 Dec;18(4):339-66.

[6] Jokiranta E. et al. Epilepsy Among Children and Adolescents with Autism Spectrum Disorders: A Population-Based Study. J Autism Dev Disord. 2014 May 7.

[7] Orinstein AJ. et al. Intervention for optimal outcome in children and adolescents with a history of autism. J Dev Behav Pediatr. 2014 May;35(4):247-56.

[8] Mohammadzaheri F. et al. A Randomized Clinical Trial Comparison Between Pivotal Response Treatment (PRT) and Structured Applied Behavior Analysis (ABA) Intervention for Children with Autism. J Autism Dev Disord. 2014 May 20.

[9] Barrett B. et al. Comparing service use and costs among adolescents with autism spectrum disorders, special needs and typical development. Autism. 2014 Jun 9.

---------- Mukaddes, N., Tutkunkardas, M., Sari, O., Aydin, A., & Kozanoglu, P. (2014). Characteristics of Children Who Lost the Diagnosis of Autism: A Sample from Istanbul, Turkey Autism Research and Treatment, 2014, 1-10 DOI: 10.1155/2014/472120 Orinstein AJ, Helt M, Troyb E, Tyson KE, Barton ML, Eigsti IM, Naigles L, & Fein DA (2014). Intervention for optimal outcome in children and adolescents with a history of autism. Journal of developmental and behavioral pediatrics : JDBP, 35 (4), 247-56 PMID: 24799263