Does the fact that this was a meta-analysis of multiple studies maybe account for differences in particular types of honey?