Table 3

Adjusted all-cause mortality 1-year after baseline
Overall (n=1,388) Stratified Sample3
No Hospitalization (n=723) ≥ 1 Hospitalization (n=665)
Dialysis setting: Non-VA (ref: VA) .80 (.48, 1.33) 2.93 (.99, 8.64) .79 (.45, 1.39)
Dual .52* (.31, .86) 1.40 (.40, 4.90) .42** (.24, .74)
Age 1.05*** (1.03, 1.06) 1.05*** (1.03, 1.08) 1.05*** (1.03, 1.07)
Male 2.30 (.29, 18.54)
Race: Non-White (ref: White) .66 (.46, .96)
 Unknown .96 (.40, 2.32)
Marital status: Unmarried (ref: married) 1.34 (.95, 1.89)
 Unknown 1.22 (.43, 3.43)
Copayment required 1.21 (.65, 2.27)
Diagnostic Cost Group 1.04 (.98, 1.10) 1.03 (.93, 1.15) .99 (.92, 1.06)
History of dialysis prior to baseline:
 6 months (ref: none) .84 (.46, 1.55) 1.35 (.53, 3.45) .65 (.28, 1.51)
 1 year .61 (.27, 1.39) .69 (.20, 2.38) .76 (.26, 2.23)
 2 years .92 (.64, 1.33) 1.03 (.55, 1.90) 1.07 (.65, 1.74)
History of hospitalization 1-year prior to baseline 1.30 (.90, 1.8) 1.01 (.56, 1.84) 1.30 (.82, 2.05)
Distance to nearest VA dialysis, miles 1.001 (.997, 1.004) 1.00 (.99, 1.01) 1.00 (1.00, 1.01)
Region: VISN B (ref: VISN A) .94 (.63, 1.40)
C-statistic 0.70 0.71 0.70

1. Notes: 1. * p< .05 **p< .01 ***p<.001.

2. Adjusted odds ratios reported with robust 95% Confidence intervals noted in parens.

3. Due to the limited distribution of deaths in these stratified samples, explanatory variables in the stratified mortality analyses were restricted to several key covariates, shown above. Estimates from this restricted model did not differ from estimates from the unrestricted model.

Wang et al.

Wang et al. BMC Health Services Research 2013 13:26   doi:10.1186/1472-6963-13-26

Open Data