Skip to main content
. 2015 Apr 13;10(4):e0121705. doi: 10.1371/journal.pone.0121705

Table 2. Cox proportional hazards model for outcomes in incident hemodialysis patients subsequently with long bone fracture, non-long bone fracture, and no fracture.

Variable LB Fx N = 3008 No Fx 1 N = 15040 NLB Fx N = 2070 No Fx 2 N = 10350 NLB Fx N = 2070 LB Fx N = 3008
N (%) N (%) HR P N (%) N (%) HR P N (%) N (%) HR P
Mortality 1045 (34.7) 3501 (23.3) 1.59 <0.001 660 (31.9) 2298 (22.2) 1.52 <0.001 660 (31.9) 1045 (34.7) 0.94 0.196
Stroke 823 (27.4) 3930 (26.1) 1.09 0.028 603 (29.1) 2619 (25.3) 1.19 <0.001 603 (29.1) 823 (27.4) 1.08 0.157
CAD 922 (30.7) 4675 (31.1) 0.99 0.898 707 (34.2) 3204 (31.0) 1.13 0.003 707 (34.2) 922 (30.7) 1.13 0.012
ACS 290 (9.6) 1624 (10.8) 0.92 0.181 221 (10.7) 1105 (10.7) 1.03 0.687 221 (10.7) 286 (9.5) 1.12 0.206
PAOD 236 (7.8) 1177 (7.8) 1.06 0.382 210(10.1) 778 (7.5) 1.41 <0.001 210 (10.1) 236 (7.8) 1.27 0.011

ACS, acute coronary syndrome; CAD, coronary artery disease; HR, hazard ratio; PAOD, peripheral arterial occlusive disease; LB Fx, long bone fracture; No Fx-1, no fracture group matched with the LB Fx group; NLB Fx, non-long bone fracture; No Fx-2, no fracture group matched with the NLB Fx group.

Adjustments were made for age, sex, diabetes, hypertension, hyperlipidemia, and COPD.