Cardiovascular disease and non-steroidal anti-inflammatory drug prescribing in the midst of evolving guidelines

Rationale, aims and objectives Responding to safety concerns, the American Heart Association (AHA) published guidelines for non‐steroidal anti‐inflammatory drug (NSAID) use in patients with pre‐existing cardiovascular disease (CVD) during 2005 and revised them in 2007. In the revision, a stepped approach to pain management recommended non‐selective NSAIDs over highly selective NSAIDs. This research evaluated NSAID prescribing during and after guideline dissemination. Method A cross‐sectional sample of 8666 adult, community‐based practice visits with one NSAID prescription representing approximately 305 million visits from the National Ambulatory Medical Care Survey (NAMCS) from 2005 to 2010 was studied. Multivariable logistic regression controlling for patient, provider and visit characteristics assessed the associations between diagnosis of CVD and NSAID type prescribed during each calendar year. Visits were stratified by arthritis diagnosis to model short‐term/intermittent and long‐term NSAID use. Results Approximately one‐third (36.8%) of visits involving a NSAID prescription included at least one of four diagnoses for CVD (i.e. hypertension, congestive heart failure, ischaemic heart disease or cerebrovascular disease). Visits involving a CVD diagnosis had increased odds of a prescription for celecoxib, a highly selective NSAIDs, overall [adjusted odds ratio (AOR) = 1.29, 95% confidence interval (CI): 1.06–1.57] and in the subgroup of visits without an arthritis diagnosis (AOR = 1.45, 95% CI: 1.11–1.89). Results were not statistically significant for visits with an arthritis diagnosis (AOR = 1.10, 95% CI: 0.47–2.57). When analysed by year, the relationship was statistically significant in 2005 and 2006, but not statistically significant in each subsequent year. Conclusion National prescribing trends suggest partial implementation of AHA guidelines for NSAID prescribing in CVD from 2005 to 2010.