Comparison of the incidence rates of selected gastrointestinal events reported for patients prescribed celecoxib and meloxicam in general practice in England using prescription-event monitoring (PEM) data

Background. Celecoxib and meloxicam are classified as cyclo-oxygenase (COX)-2 selective inhibitors, and were developed to minimize the risk of gastrointestinal (GI) toxicity commonly associated with non-steroidal anti-inflammatory drugs (NSAIDs). The Drug Safety Research Unit (DSRU) monitored the safety of these drugs immediately after launch in England using the non-interventional observational cohort technique of prescription-event monitoring (PEM). Our objective was to investigate whether there is a clinically relevant difference in incidence of reported symptomatic (acid/peptic) and complicated upper GI conditions (perforations/bleeding) between celecoxib and meloxicam during use in general practice. Methods. Patients were identified from dispensed prescriptions written by general practitioners (GPs) for meloxicam (December 1996 to March 1997) and celecoxib (May to December 2000). Simple questionnaires requesting details of events occurring during/after treatment and potential risk factors (including age, sex, history of upper GI problems, and NSAIDS prescribed within 3 months of treatment) were posted to prescribing GPs at least 6 months after the first prescription for each patient. Incidence rates of the first event were calculated; crude and adjusted rate ratios (RR) obtained using regression modelling. Results. For celecoxib and meloxicam, respectively, 1054 (6.0%) and 1376 (7.2%) patients had symptomatic (acid/peptic) upper GI events whereas 42 (0.2%) and 67 (0.4%) had complicated upper GI conditions (perforations/bleeding). A higher proportion of the celecoxib cohort had an indication for osteoarthritis (28.1 vs 23.2%), were female (68.3 vs 67.1%), were aged 60 yr or more (59.5 vs 55.0%), were prescribed NSAIDs within 3 months of starting treatment (49.4 vs 47.9%), and had a past medical history of upper GI conditions (54.7 vs 29.2%) than those prescribed meloxicam. This suggests differential channelling of groups at higher risk of such events on to celecoxib compared with meloxicam. There was no difference between the two drugs in the time to occurrence of either group of event. The RR over the 270-day study period for celecoxib compared with meloxicam were 0.77 (95% CI 0.69, 0.85) and 0.56 (95% CI 0.32, 0.96) for symptomatic (acid/peptic) upper GI events and complicated upper GI conditions (perforations/bleeding), respectively, adjusted for age, age2, sex, indication, medical history of upper GI problems and whether NSAIDs were prescribed with