Table 2.

Crude and adjusted odds ratios for association between use of antithrombotic drug and serious upper gastrointestinal bleeding

	Cases (exposed/unexposed)	Controls (exposed/unexposed)	Crude odds ratios (95% CI)	Adjusted odds ratios (95% CI)*
Single drug regimens, current use
Aspirin alone	196/1063	4123/50 498	2.4 (2.0 to 2.8)	1.8 (1.5 to 2.1)
Clopidogrel alone	12/1063	203/50 498	3.1 (1.7 to 5.6)	1.1 (0.6 to 2.1)
VKA alone	56/1063	1227/50 498	2.2 (1.7 to 3.0)	1.8 (1.3 to 2.4)
Dipyridamole alone	36/1063	738/50 498	2.4 (1.7 to 3.4)	1.9 (1.3 to 2.8)
Two drug regimens, current use
Aspirin and clopidogrel	13/1063	49/50 498	12.6 (6.6 to 24)	7.4 (3.5 to 15)
Aspirin and VKA	16/1063	114/50 498	6.4 (3.7 to 11)	5.3 (2.9 to 9.5)
Dipyridamole and aspirin	44/1063	737/50 498	3.1 (2.2 to 4.2)	2.3 (1.7 to 3.3)
Past use
Aspirin	108/886	3990/44 968	1.5 (1.2 to 1.8)	0.9 (0.7 to 1.2)
Clopidogrel	4/990	111/48 847	1.8 (0.7 to 5.1)	0.8 (0.3 to 2.2)
Dipyridamole	2/992	152/48 806	0.8 (0.2 to 2.9)	0.4 (0.1 to 1.6)
VKA	48/946	1028/47 930	2.4 (1.8 to 3.3)	1.8 (1.3 to 2.4)

VKA=vitamin K antagonist.

^*Adjusted for previous discharge diagnosis of peptic ulcer, peptic ulcer bleeding, chronic obstructive lung disease, ischaemic heart disease, alcohol related diagnosis or drug use, or liver cirrhosis or renal failure; for past Helicobacter pylori eradication; and for concurrent use of non-steroidal anti-inflammatory drugs, antiulcer drugs, nitrate vasodilators, selective serotonin reuptake inhibitors, or systemic corticosteroids.