This study compared the analgesic efficacy and safety of the cyclo-oxygenase-2 specific inhibitor celecoxib with the non-specific non-steroidal anti-inflammatory drug, naproxen, in patients with acute shoulder pain. Adult patients with shoulder pain, onset within the previous 14 days and shoulder pain of ≥ 40 mm on a 100 mm visual analogue scale (VAS), were treated with oral doses of celecoxib or naproxen for 14 days and followed by a visit at day 42 in a double-blind, randomized study. The primary efficacy assessment was change in maximum pain at rest on a 100 mm VAS at day 14 compared with baseline. In addition, secondary efficacy pain and functional assessments were analysed at baseline, day 14 and day 42. A total of 202 patients were included in the trial (99 celecoxib 400 mg/day; 103 naproxen 1 g/day). The difference in change from baseline at day 14 in maximum pain at rest was not statistically significant between the two treatment groups, but was numerically higher for celecoxib than for naproxen (−47.9 ± 2.5 versus −42.3 ± 2.5, respectively). According to the limits of the 95% confidence interval of the difference between groups (−12.52; 1.38), celecoxib appeared to be at least as effective as naproxen. All secondary efficacy measures followed the same pattern, showing similarity between the two treatments with a trend in favour of celecoxib. The incidences of adverse events were similar for both groups. Fewer patients experienced epigastric pain with celecoxib (seven patients versus 14 with naproxen). This adverse event led to discontinuation in two patients receiving celecoxib and five receiving naproxen. Celecoxib 400 mg/day was at least as effective as naproxen 1 g/day in managing pain in this condition.