The correct use of a drug is determined by several important factors. The most significant of these is a correct diagnosis for the choice of the appropriate therapeutic approach, followed by the physicianis awareness of each drugis product characteristics sigma such as indications, contraindications and warning eta and by a careful evaluation of the patient in order to consider possible risk factors, concomitant pathologies and treatments. An adequate knowledge of pharmacological therapy and of the patientis history and disease states could in fact prevent most of the adverse drug reactions attributable to an inappropriate prescription. It is clear that the physicianis role in the correct use of a drug is extremely important. Observational studies and surveys evaluating doctorsi prescribing habits could be a very useful instrument in identifying medication prescribing errors and the consequent occurrence of adverse events. An open-label, multicentre, observational cohort study, recently performed in Ireland and involving more than 9000 patients, investigated the use of non-steroidal anti-inflammatory drugs (NSAIDs) in general practice, and in particular the use of diclofenac, nimesulide and ibuprofen. These three drugs were shown to be the NSAIDs most frequently prescribed (80% of total prescriptions) by general practitioners in Ireland. The study was designed and powered to detect differences in the general safety proffle of the three NSAIDs and to test prescribing physiciansi compliance with each drugis Summary of Product Characteristics (SmPC). The study was performed as closely as possible to the normal prescribing habits of the participating physicians. The three drugs were shown to be well tolerated. In the nimesulide group, the percentage of patients who experienced treatment-related adverse events was lower than that of the diclofenac group and similar to that observed with the use of ibuprofen. A more favourable gastrointestinal safety profile was evident for nimesulide in comparison to diclofenac. The data showed that the study physicians prescribed the drugs with a good degree of compliance (84% of prescriptions) with the information reported in each of the SmPCs. This supported the fact that the adverse events were markedly limited.