Subclinical autistic traits affecting adolescent sleep patterns?

There was something rather intriguing about the results published by Liisa Salmela and colleagues [1] (open-access available here) observing that: "Elevated levels of autistic traits were significantly associated with shorter weekday sleep duration" in a cohort of adolescents from Helsinki in Finland. The fact that researchers also concluded that "autistic traits remained an independent predictor of short sleep duration when comorbid psychiatric symptoms were controlled for" added to the intrigue.

I was impressed with some elements of the Salmela study; notably the use of actigraphy as an objective measure of sleep accompanying self-report data derived from the Pittsburgh Sleep Quality Index (PSQI). The fact that actigraphs were worn "continuously for an average of 8.36 nights (SD = 1.76; range 4–17)" also meant that researchers had access to quite a bit of data from their 150+ participant study group when it came to activity cycles covering sleep.

What was the measure of autistic traits used I hear you ask? Well, we are told that: "Autistic traits were assessed using the Autism Spectrum Quotient (AQ)" which is an OK measure I suppose, although not without some shortcomings in terms of what is being specifically measured (see here). I should also mention that for the most part, the Salmela study was a study of non-autistic adolescents (two participants were reported to have "scored at or above the clinical cut-off score of 32" on the AQ).

"Continuous autistic traits significantly predicted weekday total sleep time" and "having elevated autistic traits as measured by AQ (Autism Spectrum Quotient) increased the risk for short sleep duration." Mindful that correlation is not necessarily the same as causation, details were important to the Salmela findings as we are also told that (a) as a group, boys tended to score higher on the AQ and (b) again as a group "boys had significantly shorter weekday total sleep time" than girls. It's perhaps no surprise therefore that: "Sex had a statistically significant main effect (p = 0.032) on total sleep time."

Next question: why? Why should autistic traits potentially "increase the risk for short sleep duration in a general adolescent population"? Is there something about the presentation of autistic traits that affects sleep either through psychological/cognitive processes or more physiological processes? Well, those are questions that still need answering. And alongside we seem to have yet another example where core autistic features may very well impact on so-called comorbid issues to add to the collection (see here and see here). This also has potential implications for intervention too...

----------

[1] Salmela L. et al. Autistic traits and sleep in typically developing adolescents. Sleep Med. 2018 Oct 29;54:164-171.

----------