I was just looking through the protocol for:
http://www.bristol.ac.uk/media-library/sites/ccah/migrated/documents/smprotv6final.pdf
They say that:
Were the results for this primary outcome ever released?
In this paper they say that they're going to change their primary outcome for SMILE:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4235039/
I can't see where they released the results for this primary outcome measure though. [There was no significant difference between group for it in the recent paper, but I think that did not include those from this pilot?]
How dodgy is it to run a 'pilot' trial, and then use that to change your primary outcome for the proper trial? They're presenting it as if this change was patient led, but if they started with null results for their pilot, shouldn't that have been reported? This whole thing seems so weird that I don't know how to fairly complain about it (I feel like I should get all these papers out, and go through with a highlighter and biro... but don't want to!). Any views? I don't know what's seen as normal/acceptable/respectable for this sort of pilot trial.
Also, as Valentijn pointed out, in their original protocol it seemed that they were planning to check attendance rates with the school, rather than relying on potentially biased self-report:
It seems that was changed but I couldn't find disucussion of it.