To the Editor: At Pfizer, we are very concerned about major conclusions drawn with regard to cost-effectiveness of COX-2- specific inhibitors (CSI) in the article by Segal and colleagues;1 these conclusions are based on an entirely inadequate and inappropriate database.

The CSI cost-effectiveness analysis is based on two publications described by Segal et al as “seminal”,2,3 and on a British Medical Journal editorial described as “a report by the US FDA [United States Food and Drug Administration]”.4 The first of the two “seminal” publications was a pivotal study for registration purposes;2 the second was neither pivotal nor “seminal”.3 The purported “report by the FDA” presents the authors’ opinion of a single study5 which does not reflect the overall body of gastrointestinal (GI) safety data for either CSI or non-steroidal anti-inflammatory drugs. The issues with this study and the inaccuracies in the editorial have been reported.6

Segal et al1 have apparently ignored at least 20 other publications (references available on request) comparing the safety and efficacy of celecoxib with NSAIDs; we would consider these relevant to any valid analysis of cost-effectiveness. There are probably a similar number of relevant publications for the second CSI, rofecoxib, which should also be considered if an analysis of the CSI class is to be rigorous.

The body of evidence (previously summarised7) shows that CSIs are associated with a 50% or lower incidence of serious GI complications than non-specific NSAIDs, which results in lower mortality. Any cost-effectiveness analysis that ignores this GI safety advantage of CSIs over NSAIDs is incomplete. Therefore, the conclusions by Segal et al that “Non-specific NSAIDs and COX-2 NSAIDs were found to perform similarly in terms of outcomes and side effects . . .” and “ . . . In most scenarios, COX-2 NSAIDs are dominated by non-specific NSAIDs”1 are not true reflections of either the trial data nor real-world clinical practice, and may mislead the medical community as to the relative safety of these agents.

It is important to note that the full body of celecoxib data has been submitted by Pfizer as a cost-effectiveness analysis to the Pharmaceutical Benefits Advisory Committee (PBAC). Contrary to the conclusions of Segal et al, in every scenario in the PBAC submission celecoxib was more cost-effective than non-specific NSAIDs. Celecoxib is currently listed on the Pharmaceutical Benefits Scheme and this indicates that it was considered cost-effective. An abstract has been published regarding the cost-effectiveness of celecoxib in the Australian setting.8

We consider that every statement made in the article by Segal et al with regard to the cost-effectiveness of COX-2-specific inhibitors1 is inaccurate and misleading, as it is not based on the available evidence. We request a retraction statement from the authors.