SOS

The Safety Of non-Steroidal anti-inflammatory drugs (SOS) proposal aims to assess the relative cardiovascular (CVD) and gastrointestinal (GI) safety of non-steroidal anti-inflammatory drugs (NSAIDs). The NSAIDs are divided in traditional NSAIDs (tNSAIDs) and the newer COX-II inhibitors(coxibs). The aim will be fulfilled by a two-phase approach comprising systematic reviews and synthesis of CVD and GI risk information from clinical trials and published observational studies, followed by the design and conduct of a multi-country study in existing health care databases in the UK, Netherlands, Germany and France, comprising medical information on at least 35 million persons. A data ware house will be constructed that will contain all pre-specified and locally elaborated anonimized data from inception cohorts of NSAID users. Data elaboration is standardized through a common protocol that will be designed on the basis of information and knowledge gaps observed in the systematic literature reviews, plus information requirements for the statistical and decision models. The database study will yield risk estimates for CVD and GI bleeding for each individual NSAID by dose and duration and by other important effect modifiers (e.g. aspirin use). Separate models will be built for children since the indications and dosages differ and little is known on the safety of NSAIDs in this group as they are often prescribed off-label. Special emphasis will be put on the assessment and evaluation of methodological issues, such as confounding by indication, and outcome validity as these constitute the most important threats to the interpretation, robustness and perceived validity of observational studies. The results of the literature reviews, analysis of observational databases and re-analysis of published studies will feed into a decision model for clinicians to support treatment decisions and a decision model for regulatory authorities that will focus on the public health risk.