Risk Adjustment Based Upon Resource Utilization - Text Mining Techniques for Healthcare Provider Quality Determination - page 271

SD: 186099
Starting with the next level, race and income quartile are used to make the splits in the tree.
IF Race (uniform) IS ONE OF: 1 2 5
AND Median household income quartile for patient's ZIP Code IS ONE OF: 3 4
AND Admission type < 1.5
AND 2.5 <= Disease Staging: Resource Demand Level < 3.5
THEN
NODE: 62
N: 71128
AVE: 18367.9
SD: 19320.2
IF Race (uniform) IS ONE OF: 3 4 6
AND Median household income quartile for patient's ZIP Code IS ONE OF: 3 4
AND Admission type < 1.5
AND 2.5 <= Disease Staging: Resource Demand Level < 3.5
THEN
NODE: 63
N: 13072
AVE: 21539.1
SD: 24382
IF Median household income quartile for patient's ZIP Code IS ONE OF: 1 2
3
AND Race (uniform) EQUALS 1
AND Admission type < 2.5
AND 3.5 <= Disease Staging: Resource Demand Level < 4.5
THEN
NODE: 64
N: 43987
AVE: 29756.9
SD: 29730.3
IF Median household income quartile for patient's ZIP Code EQUALS 4
AND Race (uniform) EQUALS 1
AND Admission type < 2.5
AND 3.5 <= Disease Staging: Resource Demand Level < 4.5
THEN
NODE: 65
N: 15506
AVE: 33577.4
SD: 33142.4
IF Race (uniform) IS ONE OF: 1 5
AND 3.5 <= Admission source (uniform)
AND 2.5 <= Admission type