> The burden is on the flossers to prove the efficacy beyond a secondary link and prove flossing is better in a controlled study No, you're missing my point -

> The burden is on the flossers to prove the efficacy beyond a secondary link and prove flossing is better in a controlled study No, you're missing my point - the secondary link is the study of flossing w.r.t. oral health outcomes. The primary (causal) links are whether flossing removes food particles, and whether food particles cause tooth decay (neither of which seems to be under contention here). By just studying whether flossing improves oral hygiene, you're introducing a huge number of variables that are difficult to control for (flossing skill, other hygiene factors like flossing or "rinsing", dietary choices, self reporting errors, etc etc etc). (edit: to summarize: A causes B, B causes C, but you don't like saying A causes C because people kind of suck at A, and D, E, and F all also cause C which makes studying the relationship between A and C a challenge)