The association between serious upper gastrointestinal bleeding and incident bisphosphonate use: a population-based nested cohort study

Table 1 Characteristics of the bisphosphonate exposure cohort stratified by upper gastrointestinal bleeding status ^a

	UGIB		Unadjusted odds ratio (95% CI)	p -Value
	n = 117 (0.4)
	≤ 80 years n = 47 (40.2)	> 80 years n = 70 (59.8)
Age (y)
≤ 80	47	-	(reference)
> 80	-	70	2.08 (1.44, 3.01)	0.000
Sex ^c
Female	39 (83.0)	57 (81.4)	(reference)
Male	8 (17.0)	13 (18.6)	1.72 (1.07, 2.76)	0.025
Comorbid conditions ^c
None	2 (4.3)	4 (5.7)	(reference)
1 – 3	9 (19.1)	17 (24.3)	1.29 (0.51, 3.15)	0.571
4 - 6	28 (59.6)	38 (54.3)	1.61 (0.70, 3.71)	0.267
7 - 15	8 (17.0)	11 (15.7)	1.69 (0.55, 3.48)	0.487
Past history of gastric ulcer disease ^b
No	35 (74.5)	58 (82.9)	(reference)
Yes	12 (25.5)	12 (17.1)	2.21 (1.41, 3.46)	0.001
Past history of serious GI bleeding ^b
No	41 (87.2)	62 (88.6)	(reference)
Yes	6 (12.8)	8 (11.4)	2.79 (1.59, 4.90)	0.000
NSAIDs (oral) ^d
No concurrent use	42 (89.4)	65 (92.9)	(reference)
Concurrent use	5 (10.6)	5 (7.1)	0.42 (0.22, 0.80)	0.009
Anti-platelet / anti- coagulant drugs (oral) ^c
No concurrent use	40 (85.1)	64 (91.4)	(reference)
Concurrent use	7 (14.9)	6 (8.6)	0.85 (0.48, 1.51)	0.848
Proton pump inhibitor drugs (oral) ^b
No concurrent use	33 (70.2)	50 (71.4)	(reference)
Concurrent use	14 (29.8)	20 (28.6)	2.03 (1.36, 3.03)	0.001

^a Data are shown as number (percentage) unless otherwise indicated.
^b p < 0.001.
^c Not significant.
^d p < 0.05.
CI = confidence interval; GI = gastrointestinal; NSAID = UGIB = upper gastrointestinal bleed.

ISSN: 1471-2318