• osteoarthritis
• knee Osteoarthritis
• arthritis

I read with great interest the ChONdroitin versus CElecoxib versus Placebo Trial (CONCEPT) by Reginster et al.1 I would like to raise some worthwhile issues that need to be clarified.

The authors concluded that 800 mg/day pharmaceutical-grade chondroitin sulfate (CS) is similar to celecoxib in improving pain and function after 6 months in patients with symptomatic knee osteoarthritis (OA).1 However, the interpretation of similarity between these two active treatments was not based on the analysis of the data. Because, the authors did not assess whether these treatments are similar by equivalence analysis. Rather, the authors assessed the superiority between these two active treatments by aiming to detect a difference in intention-to-treat (ITT) populations—although the study might not be powered to be able to detect differences between these two active treatments (type 2 error), since sample size is calculated to show superiority of CS over placebo; and they failed to show a difference between CS and celecoxib. According to the Consolidated Standards of Reporting Trials 2010 statement on reporting of non-inferiority and equivalence randomised trials, 'Failure to show a difference does not mean they are equivalent. By contrast, …