Table 2.

Univariate and multivariable analysis of grade 2+ gastrointestinal hemorrhage by treatment and clinical risk factors.

Risk factor	Cause-specific hazard ratio (95% CI) P-value
	Clinical cohort (N = 1008)		GWAS cohort (N = 616)
	Univariate	Multivariable	Univariate	Multivariable
Bevacizumab treatment	2.50 (1.47–4.26) <0.001	2.88 (1.64–5.06) <0.001	4.67 (1.94–11.21) <0.001	5.70 (2.17–14.97) <0.001
Agea	1.05 (1.02–1.08) 0.001	1.05 (1.02–1.09) (0.002)	1.05 (1.01–1.10) 0.01	1.06 (1.01–1.12) 0.01
History of peptic ulcer disease	2.44 (1.11–5.36) 0.03	2.20 (0.98–4.90) 0.055	2.97 (1.04–8.46) 0.04	2.61 (0.87–7.83) 0.09
History of hemorrhageb	1.16 (0.36–3.71) 0.80	0.97 (0.30–3.14) 0.96	1.60 (0.38–6.70) 0.52	0.98 (0.22–4.33) 0.98
Antiplatelet/ anticoagulant use	1.34 (0.81–2.21) 0.26	1.29 (0.76–2.19) 0.35	1.25 (0.63–2.51) 0.52	1.14 (0.54–2.42) 0.72
History of smoking	1.27 (0.77–2.09) 0.34	1.31 (0.79–2.17) 0.29	1.22 (0.61–2.44) 0.57	1.32 (0.65–2.66) 0.45
Hemoglobina	0.94 (0.82–1.09) 0.45	0.93 (0.79–1.08) 0.34	0.92 (0.75–1.13) 0.45	0.94 (0.75–1.17) 0.57

^aAge and hemoglobin as continuous variables.

^bPrior history of hemorrhage includes gastrointestinal/genitourinary hemorrhage and/or gastrointestinal perforation.