The title of this post mirrors that published by Rebecca Grzadzinski and colleagues  looking at an intriguing phenomenon whereby "some children with ASD [autism spectrum disorder] may show behavioral improvements during fever."
I had expected these results to be published given their appearance at this years IMFAR conference (see here) as one of my 'ones to watch' picks. Indeed, the results do not disappoint as the words 'discovery sample' and 'replication sample' are used as per other instances in the peer-reviewed research literature of this methodology (see here for example).
Drawing on data from the Simons Simplex Collection where over 1500 parents with children with autism were asked "whether and in which areas their child improved during fever", researchers randomly assigned participants to either the discovery or replication sample. They took various information - "demographics, medical history, ASD symptoms, adaptive skills, and presence of de novo ASD-associated mutations" - for each child and compared 'fever improvers' with 'fever non-improvers' across those measures for the discovery set.
Results: "Parent reports of 17% of children indicated improvements during fever across a range of domains." Some people might quibble with the reliance on parent report in this study but not me as per other work indicating 'sensitivity to an issue if not necessarily expertise' (see here). That almost a fifth of children with ASD might potentially show some clinical improvement during a fever episode is not to be sniffed at (see here) and hints that previous chatter about 'hot baths and autism' (no really, see here) might not be just some odd idea (albeit with the requirement for further experimental study and certainly with no endorsement from me at this time).
Then: "Discovery and replication analyses revealed that the Improve Group had significantly lower non-verbal cognitive skills (NVIQ) and language levels and more repetitive behaviors." Such results, whilst requiring replication, hint at some interesting future studies to be done with certain parts of the autism spectrum. Lower NVIQ and language levels shouldn't however be just interpreted as just meaning those towards the more 'severe' end of the autism spectrum; even those with spoken language on the autism spectrum can be quite severely disabled by their autism.
"Understanding the profiles of children who improve during episodes of fever may provide insights into innovative treatments for ASD." I find myself in agreement with such sentiments, particularly when set in the context of possible "immunologic and neurobiological pathways, intracellular signaling, and synaptic plasticity" being potentially affected by fever onset  in the context of autism. I of course understand that for some people, the whole notion of fever being tied into behavioural presentation is gonna be met by the same old cynicism (of course, autism is lifelong and immutable!) but science is science...
Music to close: Hüsker Dü - Makes No Sense At All (no reflection of the current research I might add).
 Grzadzinski R. et al. Children with autism spectrum disorder who improve with fever: Insights from the Simons Simplex Collection. Autism Res. 2017 Aug 31.
 Curran LK. et al. Behaviors Associated With Fever in Children With Autism Spectrum Disorders. Pediatrics. 2007; 120: e1386-92.