"A more homogeneous subgroup with regression between 18 and 36 months (n = 48) had higher rates of intellectual disability, epilepsy, and special education, more socially restrictive educational settings, and more severe ASD [autism spectrum disorder] communication deficits and schizophrenia spectrum symptoms than non-regressed youth (n = 136)."
So said the findings reported by Kenneth Gadow and colleagues  taking on one of the more important issues in relation to autism: developmental regression (see here). I say 'more important' in that last sentence because whilst it has taken autism research and science quite a while to accept that regression in skills/behaviour can occur in cases of autism, there is now a realisation that such regression can impact on various other areas of investigation, not least the discussions on the potential [very] early detection of autism (see here). And still researchers continue to talk about very early autism detection...
Gadow (a name not unfamiliar to this blog in terms of autism and schizophrenia spectrums potentially colliding) et al set out to examine "the psychiatric and clinical correlates of loss of previously acquired skills (regression) as reported by parents of youth with autism spectrum disorder (ASD)." They relied to quite a large extent on data derived from the Child and Adolescent Symptom Inventory partly created by Dr Gadow, that assesses for the presence of various behaviours linked to various developmental and psychiatric labels. Some 213 6- to 18-year old children and adolescents with ASD were included for study; 77 (36%) of them defined as showing some parent-reported regression in skills coincidental to the presentation of their autism.
As per that opening sentence, there did appear to be some differences in clinical presentation when using regression as a characterising feature. A few additional details: 37% of those "whom parents indicated had regressed (whether or not they indicated age of regression and type of skill)" fell into the intellectual/learning disability category vs. 18-19% of non-regressed participants. Significantly more participants who regressed were 'ever hospitalised' (48% vs 29%); a statistic that probably ties into the rates of epilepsy/seizure disorder also described in this cohort (18% vs 5-6%). Although the authors talk about 'a more severe ASD communication deficit' being linked to reports of regression, the data obtained from both parent and teacher ratings in relation to the presentation of autistic traits were mainly 'more severe' in the regression group when examined as a whole: social deficits, communication deficits and perseverative behaviours; albeit not significantly so.
Then to the suggestion that schizophrenia spectrum symptoms were more pronounced in the regression group. Appreciating that there is some history when it comes to autism and schizophrenia (think Mildred Creak for example) but also some controversy, this is something potentially quite important, not least when one considers the rise and rise of research interest into something like psychosis and [some] autism (see here and see here) and also how [some] regressive autism might not be a million miles away from a condition like anti-NMDA-receptor encephalitis (see here) typically encompassing a psychotic element to it. I also wonder whether the talk about regression accompanying autism might set the tone for further regression potentially accompanying the onset of something like schizophrenia for example. Could a life of 'regression in skills' be a feature of this group?
Of course further investigations are required to independently verify the findings reported by Gadow and colleagues; such work should really also include more professionally-derived ratings (bearing in mind parents are generally the experts on their own children). I wonder too if it's time to start including a few more 'biological' measures when it comes to looking at regression vs. no regression in the context of autism to see if genetics/epigenetics/biochemistry variables might also tie up with some of the behavioural findings. Indeed, this should really be a must.
Finally: "A brief parent report of developmental regression may be a useful clinical indicator of later general functioning." I agree that this might be a rather good idea based on the current data; perhaps allied to some further chatter about ESSENCE and autism also (see here).
 Gadow KD. et al. Parent-Reported Developmental Regression in Autism: Epilepsy, IQ, Schizophrenia Spectrum Symptoms, and Special Education. J Autism Dev Disord. 2017 Jan 10.
Gadow KD, Perlman G, & Weber RJ (2017). Parent-Reported Developmental Regression in Autism: Epilepsy, IQ, Schizophrenia Spectrum Symptoms, and Special Education. Journal of autism and developmental disorders PMID: 28074354