Casting out the truth: why ITT alone fails in trials that compare surgery to casting

Abstract

Randomisation and intention-to-treat (ITT) analysis are pillars of trial methodology – and for many questions they are rightly so. But when the interventions under study are epistemologically dissimilar, ITT stops answering the clinical question clinicians and patients actually care about. In trials that compare a discrete, one-off surgical procedure to a prolonged therapeutic pathway of casting, the standard ITT framework produces a categorical mismatch: it analyses assignment rather than completed treatment. That mismatch is not a minor technicality – it is a fundamental design error.