Methods: We identified all Partners Healthcare outpatients prescribed one of five prescription-only NSAIDs (diclofenac, celecoxib, indomethacin, nabumetone, or piroxicam) between January 1, 2004 and September 30, 2012. We identified ADRs to the prescribed NSAID within one year of prescription using Partners’ integrated longitudinal allergy database. We used reaction details to classify ADRs as likely HSRs and compared patient demographics and allergy history between HSR and non-HSR groups.
Results: Of 83,439 patients prescribed at least one prescription-only NSAID, 1,318 (1.58%; 95% CI 1.49-1.66%) reported an ADR. We identified 279 patients (0.33%; 95% CI 0.30-0.37%) with one or more likely HSRs representing 21.2% (95% CI 19.0-23.4%) of NSAID ADRs. The most common HSRs were rash (n=118; 42.3%), hives (n=52; 18.6%), swelling (n=50; 17.9%), itching (n=45; 16.1%), and angioedema (n=17; 6.1%). Females were more likely than males to report an NSAID HSR (RR=2.00; 95% CI 1.54-2.61). Patients with a history of another NSAID HSR were also more likely to report an HSR to the newly prescribed NSAID (RR=3.64; 95% CI 1.81-7.31).
Conclusions: NSAIDs commonly cause ADRs, with HSRs comprising about 1 in 5 ADRs. Consistent with prior knowledge, NSAID HSRs most commonly involve skin and mucous membranes. Further data are needed to validate EHR classification of NSAID HSRs.