(P<0.05; 0.00) seems incidental with the study size and honestly I can't see how could they smell the blood type.
(I'm not saying they can't, I'm saying I would like to know how.)
I'm saying it may be incidental because the paper doesn't define if the population from where mosquitoes fed had a higher or lower O-type density, nor their distribution.
I read the claim about the correlation between mosquito bites and blood type in a news article where this paper was linked as the source. This teaches me (again) to not blindly trust any news articles without verifying the information.