This appears to be the research paper in question: https://ehjournal.biomedcentral.com/articles/10.1186/s12940-025-01208-0
It's a lot to wade through right now, but it does suggest that this should be investigated further.
Given the highly genetic and heriditary nature of ASD and ADHD, I really wish studies for prenatal exposure included assessments of the mother at minimum and ideally both parents. This would help clarify two important points:
-
Does the child in question already have the genetic potential for ASD or ADHD and if so, does the prenatal exposure to certain medication increase the likelihood of it being exhibited?
-
How much of the correlation between maternal use of certain medications and ASD/ADHD diagnoses is a result of the mother being neurodivergent (but potentially undiagnosed) and hence taking medication for one of the many medical conditions that are statistically more likely to be comorbid with ASD & ADHD? (See https://allbrainsbelong.org/all-the-things/ for details)
Given how genetically complex both ASD and ADHD are (and are clearly multi-factorial in nature) I would expect both of the above points to explain the majority of any apparent link.
It's not a linear scale from "mild*" to "debilitating". A spectrum just groups closely-related things (e.g. consider the visible light spectrum; red is not more of a colour than blue). The whole point of it was to highlight that autism is a complex condition that has a wide range of presentations.
There is also a good argument that so-called "severe" autism (learning disabilities, etc.) are conditions that are simply more likely to be comorbid with autism (just like many other medical conditions) and are not an intrinsic part of autism itself.
*mild really should just be considered high-masking and indicates that those people have learnt how to "fit in" to a neurotypical society.