Introduction It is not clear what impact the decline of COX-2 selective inhibitors has had on the use of non-selective non-steroidal anti-inflammatory drugs (ns-NSAIDs) or on the incidence of non-variceal upper gastrointestinal bleeding (NV-UGIB). Also, proton pump inhibitors (PPIs), recommended for NSAID-ulcer prophylaxis, have been linked to Clostridium difficile infection.

Methods We aimed to measure the incidence [cases per 100 000 population per annum] of NV-UGIB and C difficile infection as well as the use of ns-NSAIDs and PPIs in a well-defined population served by a single centre. Data were collected over three calendar years, 2007–2009. Significance was assessed using the χ² test for trend.

Results As shown in Abstract PWE-158 table 1, the use of COX-2 selective and newer NSAIDs continues to decline, while that of ns-NSAIDs continues to rise. This has been coupled with a steady rise in the use of PPIs. On the other hand, the incidence of both NV-UGIB and C difficile infection has been falling.

Conclusion The incidence of NV-UGIB has been falling despite the increasing use of ns-NSAIDs. This could be, at least in part, due to the rise in PPI use. Also, the incidence of C difficile infection has been falling despite more PPIs being used: this might suggest a poor link between these two factors, or might reflect the outcome of other measures being deployed against this infection. While these falls appear promising, both NV-UGIB and C difficile infection continue to represent significant clinical challenges.

Competing interests None declared.