Table 2:

Univariate and adjusted proportional HRs of all-cause mortality for BP, PP and SDR per unit increase.

	AF or HF (n = 196)		No AF or HF (n = 362)		All (N = 558)
Model	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value
SBP	0.976 (0.965–0.986)	<.001	1.010 (0.997–1.023)	.13	–e	–e
SBPb	0.978 (0.967–0.989)	<.001	1.016 (1.004–1.028)	.01	–e	–e
SBPc	0.978 (0.967–0.990)	<.001	1.015 (1.003–1.027)	.02	–e	–e
SBPd	–a	–a	–a	–a	–a,e	–a,e
PP	0.978 (0.962–0.994)	.008	1.049 (1.033–1.065)	<.001	–e	–e
PPb	0.972 (0.955–0.990)	.002	1.032 (1.014–1.049)	<.001	–e	–e
PPc	0.973 (0.955–0.990)	.003	1.032 (1.015–1.050)	<.001	–e	–e
PPd	1.002 (0.970–1.034)	.92	1.045 (1.013–1.078)	.005	–e	–e
SDR	1.322 (0.978–1.787)	.07	1.475 (1.278–1.702)	<.001	1.356 (1.196–1.538)	<.001
SDRb	1.320 (0.974–1.789)	.07	1.413 (1.168–1.710)	<.001	1.264 (1.075–1.486)	.005
SDRc	1.361 (1.005–1.843)	.05	1.413 (1.165–1.714)	<.001	1.283 (1.096–1.503)	.002
SDRd	1.314 (0.972–1.777)	.08	1.388 (1.126–1.711)	.002	–e	–e

^aAdjustment not performed since 24-hour SBP used as risk predictor. Adjustment for

^bage, sex, diabetes mellitus and serum albumin (model A);

^cmodel A plus UFV and log-transformed dialysis vintage (model B);

^dmodel B plus 24-hour SBP (model C).

^eModel and adjustment (model C) not calculated since proportional assumption violated for 24-hour SBP and PP (see Hametner and Wassertheurer [14]).