Table 1.

Baseline characteristics of cases with PUB and controls exposed to current use of conventional NSAIDs or selective COX‐2 inhibitors

		Cases (n = 2634)	Controls (n = 5074)	p‐value
Age	Mean (year ± SD)	68.75 ± 15.6	69.28 ± 14.6	0.135
Sex	Women, n (%)	1576 (59.8)	3084 (60.8)	0.420
Concomitant drug(s) use at the index date	Acid‐lowering drugs, n (%)†	164 (6.2)	187 (3.7)	0.000*
	Vitamin K antagonists, n (%)‡	399 (15.1)	244 (4.8)	0.000*
	Platelet aggregation inhibitors, n (%)§	707 (26.8)	999 (19.7)	0.000*
	Glucocorticoids, n (%)	188 (7.1)	234 (4.6)	0.000*
	Serotonin selective reuptake inhibitors, n (%)	132 (5.0)	205 (4.0)	0.048*
History of drug(s) use	Conventional NSAIDs, n (%)	192 (7.3)	502 (9.9)	0.000*
	Selective COX‐2 inhibitors, n (%)	409 (15.5)	619 (12.2)	0.000*
	Conventional NSAIDs + selective COX‐2 inhibitors, n (%)	0 (0.0)	0 (0.0)	NA
	Acid‐lowering drugs, n (%)¶	1444 (54.8)	2432 (47.9)	0.000*

NSAIDs, nonsteroidal anti‐inflammatory drugs; COX‐2, cyclooxygenase‐2; PUB, perforation, ulcers, or bleeding; NA, not applicable; SD, standard deviation.

^† Acid‐lowering drugs (antacid and H2‐receptor antagonists).

^‡ Vitamin K antagonists (phenprocoumon and acenocoumarol).

^§ Platelet aggregation inhibitors (clopidogrel, acetyl salicylic acid, dipyridamole, and prasugrel).

^¶ Acid‐lowering drugs (antacid, H2‐receptor antagonists, and proton pump inhibitors).

^* Statistically significant (p < 0.05).