It is noteworthy that the cautious and accelerated eating groups have
very different male to female ratios. This difference is statistically
significant (P = 0.01, Fisher's exact test). I therefore wonder how robust
the randomisation procedure was. Was it possible that some of the study
participants cheated, and that the boys resorted to nefarious means to
ensure that they were included in the accelerated group?
In any case, given the large difference in the sex ratio between the
groups, the possibility exists that the result was due to confounding if
boys were more susceptible to ice-cream headache than girls.
Competing interests: No competing interests