OBJECTIVES: Cyclo-oxygenase (COX)-2 inhibitor rofecoxib and valdecoxib were withdrawn from the market because of their association with heart problems. There is a lack of information on the impacts of COX-2 withdrawal on the utilization of related drug classes. The objective is to evaluate to what extent prescriptions and expenditures of non-selective non-steroidal anti-inflammatory drugs (NSAIDs) and proton pump inhibitors (PPIs) changed after the removal of two COX-2 drugs. METHODS: Prescription records from January 1, 2004 through November 30, 2005 were obtained from pharmacy claims database in a pharmacy benefit management (PBM) organization. Clients continuously enrolled in the PBM but not enrolled in COX-2, PPI, or Prilosec OTC Step Care programs during the study period were included. Number of prescriptions per thousand eligible members per month, per member per month (PMPM) total costs for non-selective NSAIDs and PPIs were calculated and compared between the pre withdrawal (January 1, 2004 through September 30, 2004) and the post withdrawal (May 1, 2005 through November 30, 2005). RESULTS: This study included 536,569 patients. After the withdrawal of the two COX-2 inhibitors, the average non-selective NSAID prescriptions per thousand eligible members per month and PMPM total costs increased by 37.72% (from 13.94 to 19.20) and 75.56% (from $0.52 to $0.91). PPI prescriptions and PMPM total costs increased by 7.21% (from 25.73 to 27.58) and 12.32% (from $4.38 to $4.92). In contrast, the average prescriptions and PMPM total costs for COX-2 inhibitors dropped by 70.99% (from 16.52 to 4.79) and 66.16% (from $2.07 to $0.70). CONCLUSIONS: After the withdrawal of COX-2 inhibitor rofecoxib and valdecoxib, there has been a large increase in non-selective NSAID prescriptions and only slight increase of PPIs. Given the safety concerns with the NSAIDs, further studies are warranted regarding the health outcomes associated with the increased use of non-selective NSAIDs.