Thread by @Richard_Autism

So the more I reflect on the assumption that PDA is ASD/ autism subtype/ subgroup/ profile; the more rediculous it seems to me.

I am reflecting on the agenda of O'Nions et al (2016), where they seek to find PDA DISCO questions that can identify what the authors think is, in the autistic population. This is while being mistaken about Newson's PDA research.

Newson was not trying to find PDA in autistic persons, she was trying to show PDA is significantly different to Kanner's autism & Asperger's Syndrome, and thus PDA is needed. Newson included in non-autistic persons in her sample.

Removed cases of PDA with autism features from her database. Said PDA is not autism and should not be viewed as such. Newson was also trying to find what PDA profile fully looked like.

More recently, it has been acknowledged that not all of Newson's sample would be diagnosed as autistic (by some). I would argue majority of them would not under strict DSM-5 autism criteria. Tangent.

Back to the approach to O'Nions et al (2016). So they had preconceived ideas of PDA, aimed to make it a "meaningful" autism subgroup. Arbitrarily choosing a threshold of 30% or less.

O'Nions et al (2016) argue their research allows for an honest picture of PDA to be represented in their sample.

Problem is, it an arbitrary threshold for PDA and they have arbitrarily chosen to view PDA as autism.

Considering that others view PDA as being seen in non-autistic persons, this is supported by limited evidence, including Newson's research (as noted above). So point is O'Nions et al (2016) cannot assume PDA is a meaningful autism subgroup.

The author's opinions are not sufficient to make the assumption that PDA is an autism subgroup either.

What is making me think it is even more ridiculous, is that different clinical populations will have different characteristics associated to them. We know that many autism comorbids present differently outside of autism.

So O'Nions et al (2016) have tried to produce a set of PDA DISCO questions, to make PDA an autism subgroup. Including changing its wording to be more autism-like. Such as "Apparently manipulative behaviour". These eleven questions are widely avaliable.

The article is open access and has been accessed about 25000 times last time I checked.

This is a problem as people looking to do PDA research in non-autistic populations probably have 11 PDA DISCO questions that are not designed to differentiate PDA in said populations.

There is nothing to say PDA features are more common than 30% non-autistic populations. Which PDA features are more or less common is predicted PDA populations. Or if said PDA features present slightly different to autism, (which they likely do).

So what use are these 11 revised PDA DISCO questions for searching PDA in non-autistic population samples?

I would point out that the less than 30% of autistic population cut-off was chosen as O'Nions et al (2016) say that features commonly associated to autism are unlikely to produce viable subgroups.

If a feature is seen in less than 30% of autistic population it is probably related to autism comorbids instead of triad/ dyad of impairment (what underpins autism dx criteria).

I.e. using non-autistic features to divide autistic persons. Then by definition PDA is not autism.

I have this point out elsewhere.

Sorry, this is a rant. It is a valid rant. Essentially one can use PDA tools designed to find PDA in autistic persons, to research and diagnose PDA in non-autistic persons.

*can not use*