Osteoarthritis trials usually report average changes in visual analogue scale (VAS) pain, and examine the difference between treatment and placebo. We investigated whether dichotomous responder analysis provides a more informative interpretation of drug efficacy.
