. 2011 Jun 24;6(6):e21447. doi: 10.1371/journal.pone.0021447

Table 3. Fully adjusted association between initiation of an antidepressant agent (exposed versus unexposed) and hospitalization for gastrointestinal bleeding in warfarin users.

Variables	0 to 29 days			30 to 59 days			60 to 119 days
	(1^st prescription)^*			(2^nd prescription)^*			(3^rd or 4^th prescription)^*
	OR	Lower	Upper	OR	Lower	Upper	OR	Lower	Upper
		95% CI	95% CI		95% CI	95% CI		95% CI	95% CI
Citalopram	1.73	1.25	2.38	1.05	0.63	1.73	1.24	0.84	1.85
Escitalopram	1.19	0.82	1.71	1.01	0.63	1.63	0.95	0.64	1.41
Fluoxetine	1.63	1.11	2.38	0.80	0.41	1.57	0.72	0.38	1.35
Paroxetine	1.64	1.27	2.12	1.29	0.90	1.87	1.07	0.77	1.49
Sertraline	1.18	0.90	1.56	1.23	0.87	1.75	1.19	0.89	1.60
Venlafaxine	1.43	0.88	2.31	0.59	0.24	1.45	1.31	0.79	2.17
Amitriptyline	1.47	1.02	2.11	0.82	0.42	1.59	1.61	1.03	2.53
Nortriptyline	1.45	0.68	3.12	No data^†			No data^†
Mirtazapine	1.75	1.30	2.35	0.64	0.36	1.13	0.95	0.63	1.42

*Adjusted for age, gender, race, number of prior warfarin prescriptions filled on the index date, nursing home, use of dementia, liver disease, prior gastrointestinal bleed, renal disease, use of acetaminophen, use of levofloxacin, and proton pump inhibitors.

^†

Insufficient number of exposed cases or controls to analyze.