Table 2.

Characteristics associated with hospice use in patients who withdrew from dialysis in the USRDS 2001 to 2002 deceased cohort, n = 25,075a

Characteristicn%Hospice Use (%)PbAdjusted OR(95% CI)Pc
YesNo
All41.958.1
Age (yr)<0.001<0.001
    0 to 447002.828.171.9Reference
    45 to 64420716.834.265.81.29 (1.07 to 1.56)
    65 to 74744729.741.758.31.72 (1.43 to 2.07)
    75 to 84962138.444.855.21.87 (1.55 to 2.24)
    ≥85310012.447.252.82.09 (1.71 to 2.54)
Gender0.1280.935
    female13,13752.441.558.5Reference
    male11,93847.642.457.61.00 (0.95 to 1.06)
Race<0.001<0.001
    white18,32673.144.155.9Reference
    black412116.434.865.20.72 (0.67 to 0.78)
    Hispanic18827.543.156.90.80 (0.72 to 0.90)
    other7463.025.574.50.46 (0.38 to 0.55)
Time on dialysis (yr)<0.0010.905
    <316,05264.042.857.21.00 (0.94 to 1.06)
    ≥3902336.040.459.6Reference
Modalityd0.7110.146
    hemodialysis23,70594.541.958.10.92 (0.81 to 1.03)
    peritoneal dialysis13675.542.457.6Reference
Dependence for ADLe<0.001<0.001
    yes13215.337.162.90.78 (0.69 to 0.88)
    no23,16392.442.457.6Reference
Primary cause of renal failure<0.0010.007
    diabetes11,04744.141.258.8Reference
    hypertension720228.743.656.40.95 (0.88 to 1.01)
    glomerulonephritis17146.841.258.80.93 (0.84 to 1.05)
    cystic kidney disease4621.845.754.31.08 (0.89 to 1.33)
    urologic5162.146.153.91.08 (0.89 to 1.31)
    other302112.041.858.20.98 (0.90 to 1.07)
    unknown or missing11134.436.663.40.77 (0.67 to 0.88)
Cause of dialysis withdrawal<0.001<0.001
    access failure4711.942.357.70.95 (0.78 to 1.17)
    transplant failure300.126.773.30.54 (0.22 to 1.34)
    failure to thrive10,76542.948.551.51.05 (0.98 to 1.12)
    medical complication884535.330.769.30.50 (0.46 to 0.54)
    other496419.847.752.3Reference
State of residencef25,075100.0<0.001<0.001
  • a The χ2 test was used for comparisons between categorical variables. ADL, activities of daily living; CI, confidence interval; OR, odds ratio.

  • b P for bivariate comparisons between those with and without hospice.

  • c On logistic regression.

  • d Data missing for three patients.

  • e Data missing for 591 patients. ADL assessment is completed at dialysis initiation.

  • f State category was significant, but there was wide variation in the χ2 results and OR.