Differential impact of NSAIDs on rate of adverse events that require hospitalization in high-risk and general veteran populations: a retrospective cohort study.

BACKGROUND: Many observational studies in the general population have demonstrated an increased risk of adverse events associated with NSAIDs, including gastrointestinal bleeds, congestive heart failure, acute renal failure, hypertension and acute myocardial infarction. Few studies, however, have explored outcomes in populations considered to be more vulnerable to their effects.
OBJECTIVE: To determine the rate of adverse events requiring hospitalization that are associated with NSAIDs in two high-risk veteran populations and the general veteran population.
METHODS: In this retrospective cohort study, we identified veterans being dispensed medicines for diabetes mellitus (diabetes cohort), those receiving renin-angiotensin system medicines and frusemide (furosemide) concurrently (ACE inhibitors/angiotensin II type 1 receptor antagonists [angiotensin receptor blockers; ARBs] and frusemide cohort), or at least one other medicine (general population/reference cohort). The primary endpoint was hospitalization with a primary diagnosis of congestive heart failure, gastrointestinal ulcer, acute renal failure, acute myocardial infarction or hypertension. Hospitalization rates during the period of non-exposure and the 30-day period after a subject was first dispensed an NSAID were compared using Poisson regression.
RESULTS: There was a significant increase in risk of all hospitalizations of interest in the exposed period compared with the unexposed period in the diabetes cohort (incidence rate ratio [IRR] 1.31; 95% CI 1.08, 1.60), ACE inhibitor/ARB and frusemide cohort (IRR 1.34; 95% CI 1.13, 1.58) and reference cohort (IRR 1.47; 95% CI 1.30, 1.66). The incidence rates demonstrate that for every 10,000 veterans treated for 30 days with NSAIDs, there were 20 extra hospitalizations in the diabetes population, 30 additional hospitalizations in the ACE inhibitor/ARB and frusemide cohort and 6 extra hospitalizations in the reference population compared with those not treated with NSAIDs.
CONCLUSIONS: NSAID use is associated with an increased risk of hospitalization in all groups, with similar risk ratio estimates. However, the clinical implications were greater in the high-risk populations, in which more hospitalizations were observed. Consideration may need to be given to differential presentation of risk information to clinicians.