To date, no reports of hypersensitivity reactions (HSRs) among nonsteroidal anti‐inflammatory drugs (NSAIDs) according to cyclo‐oxygenase (COX) selectivity and chemical groups have been published in a single study. The present study assessed the reporting frequency of HSRs for NSAIDs based on their relative inhibitory potency toward COX enzymes and chemical groups, including the presence/absence of a functional sulfonamide group, in strata observed 5 years after market authorization. A case/noncase study was performed among individual case safety reports (ICSRs) with NSAIDs as suspected drugs in VigiBase, the WHO spontaneous reporting database. Cases were ICSRs mentioning angioedema and anaphylactic/anaphylactoid shock conditions, while noncases were ICSRs without HSRs. NSAIDs were categorized into (i) NSAIDs with high COX‐2 selectivity (coxibs), (ii) noncoxib NSAIDs with COX‐2 preference, (iii) NSAIDs with poor selectivity, or (iv) NSAIDs with unknown selectivity. Chemical groups were defined based on the Anatomical Therapeutic Chemical classification system and the presence/absence of a functional sulfonamide group. Reporting odds ratios (RORs) and 95% confidence intervals (95% CIs) were calculated using logistic regression analysis. We identified 13 229 cases and 106 444 noncases. In the first 5 years after marketing, poor‐selectivity NSAIDs and acetic acid derivatives were associated with the highest ROR of HSRs (age‐ and sex‐adjusted ROR 2.12, 95% CI 1.98–2.28; and ROR 2.21, 95% CI 1.83–2.66, respectively) compared with coxibs, and sulfonamide NSAIDs were associated with the highest ROR of HSRs compared with nonsulfonamide NSAIDs (age‐ and sex‐adjusted ROR 1.38, 95% CI 1.29–1.47). After the first 5 years of marketing, most of the RORs returned to approximately 1.