A multicentre, double-blind, randomized study was conducted in patients with osteoarthritis (OA) of the hip or knee in order to compare the efficacy and safety of the new cyclooxygenase-2 (COX-2) inhibitor, meloxicam, with diclofenac sodium, a conventional treatment for this condition. Three hundred and thirty-six patients were treated with oral meloxicam 7.5 mg once daily or diclofenac 100 mg slow release once daily for 6 months. There were no significant differences between the treatment groups with respect to overall pain, pain on movement, global efficacy or quality of life scores at the end of treatment, all of which showed good levels of improvement. Sixty-six patients were withdrawn after the start of the double-blind phase due to adverse events (n = 21, meloxicam; n = 31, diclofenac) or to lack of efficacy (seven in each group). The median of dose paracetamol taken concomitantly was statistically significantly lower in the meloxicam group than in the diclofenac group (185 vs 245 mg/day; P = 0.0123) with a comparable proportion of patients taking concomitant paracetamol therapy in both groups. Both drugs were well tolerated, although severe adverse events, treatment withdrawal and clinically significant laboratory abnormalities were more common with diclofenac than with meloxicam. Thus, meloxicam 7.5 mg is a safe and effective treatment for OA of the hip and knee which demonstrates a trend towards an improved safety profile compared with diclofenac.