Nope. I can't find a PDF of the original study, but in the review, they mention that one quadrant of the mouth was professionally flossed. This presumably lets

Nope. I can't find a PDF of the original study, but in the review, they mention that one quadrant of the mouth was professionally flossed. This presumably lets them do within-subject comparisons: are there more/less cavities (or whatever) within the flossed portion? This should remove confounds like 'ate more sugary things'. It's possible, of course, that kids brushed more on the flossed side (or the unflossed side), but you could presumably quantify that too.