A bottom-up approach to autism research

Mirroring the quite extraordinary increase in the numbers of people diagnosed as being on the autism spectrum in recent years, so autism research has similarly witnessed an explosion in research output. Given such volumes of research being funded, undertaken and published, it is surprising that we still know relatively little about autism, its underlying aetiology and developmental course, outside of just description and diagnosis.

There may be several reasons for this continued lack of not knowing. One of the primary candidates put forward is the continued reliance on the singular diagnosis of autism as a research starting point, not necessarily being conducive to identifying genetic or biochemical markers for such a heterogeneous (varied) condition. The continued reconceptualisation of autism potentially reflective of a more plural condition – the so-called ‘autisms’ – set against wider moves blurring the edges and overlap of other behavioural or psychiatrically defined conditions, offers a potential answer to this state of not knowing.

The ‘proof of principle’ paper by Unwin and colleagues* provides autism research a new way of organising knowledge based on a bottom-up approach. Their suggestion and initial examples relied on a system aimed at identifying biological subtypes of ‘the autisms’ in order to focus research efforts on more homogeneous (similar) subgroups on the autism spectrum.

Their paper focused on two groups where suggested risk factors pertinent to receiving a diagnosis of autism offered a research starting point, based on (a) gestational exposure to a class of anti-depressant medication (SSRIs) and (b) low birth weight (defined as less than 2.5 kgs). Analysis was then undertaken on a variety of behaviours and complaints, both classically associated with autism and others more peripherally linked to autism.

Their findings suggested that whilst the presentation of autism did not dramatically differ between groups which fell into these new research categories and those that did not (e.g. those presenting with low birth weight and those who did not), differences were observed in other measures such as the presence of gastrointestinal (GI) disorders or sleeping problems.

Whilst the results are preliminary and constrained by the relatively low numbers of participants included for study, this alternate way of undertaking research based on the analysis of subgroups on the autism spectrum may yet provide greater insight into the nature of the autism spectrum.

* Unwin LM. et al. A “bottom-up” approach to aetiological research in autism spectrum disorders. Front Hum Neurosci. 2013 Sep 19;7:606.

Further commentary on this study can be found at: http://questioning-answers.blogspot.com/2013/11/a-bottom-up-approach-to-autism-research.html