Further to Walter Maksymowych's letter1 about James Wright's article on cyclooxygenase-2 (COX-2) inhibitors,2 we would like to add that disclosure of competing interests and presentation of contrary viewpoints tend to reduce the likelihood of bias contributing to the “tomato effect,” also known as type 4 error. This type of error is an overestimation of risks, which leads to rejection of an efficacious therapy.3
Concerns about the possibility of type 4 error in this case are reinforced by a summary of the Wright article published recently in BMJ,4 which states that “This is an excellent (although non-systematic) review of the benefits and harms of COX-2 inhibitors.” It has been overlooked that Wright, in disregarding systematic reviews and meta-analyses on COX-2 inhibitors, has missed a large body of relevant evidence, including differences between individual NSAIDs.5,6,7,8
Before the publication of Wright´s article, several other authors presented critical views regarding the cardiovascular safety of COX-2 inhibitors under the guise of scientific objectivity.9,10,11 Some argued that use of acetylsalicylic acid (ASA) might change the cost-effectiveness of COX-2 inhibition by reducing gastrointestinal benefit; hence, there would be no justification for prescribing a more expensive therapy.11 However, these authors overlooked the benefits of the combination of ASA and COX-2 inhibition relative to less-expensive options such as ASA combined with a non-ASA NSAID or a non-ASA NSAID alone. These benefits include better gastrointestinal tolerability, sustained inhibition of platelet aggregation and freedom in the dosing regimen.12
Michal R. Pijak Consultant Rheumatologist Frantisek Gazdik Research Fellow Department of Clinical Immunology Institute of Preventive and Clinical Medicine Bratislava, Slovakia