Abstract The anomalous and opposing trends in the published data dealing with turbulent flow friction factor and heat transfer coefficient in microchannels, and their apparent disagreement with macroscale correlations, are discussed. It is shown that the modification of turbulent eddy diffusivities, consistent with the way suspended particles may modify turbulence, can explain the observed higher-than-expected heat transfer coefficients in some data. It is thus suggested that suspended microscopic particles may be a major contributor to the aforementioned inconsistencies and disagreements in some of the published data.