The evidence linking the use of non-steroidal anti-inflammatory drugs (NSAIDs) with increased atherothrombotic risk is controversial, particularly in older patients. This population is consistently underrepresented in epidemiological studies. Moreover, several confounding factors such as co-morbidities, polypharmacy, and institutionalisation might affect the interpretation of studies on the real association between NSAID use and cardiovascular risk. These issues are herewith discussed together with a proposed mechanism to explain the results of recent studies demonstrating a relatively low atherothrombotic risk associated with NSAIDs in older patients. Suggestions for future research directions are also provided.