Information Technology Reference
In-Depth Information
Table 14. APRDRG mortality index and hospital
Table of DSHOSPID by APRDRG_Risk_Mortality
DSHOSPID
APRDRG_Risk_Mortality(All Patient Refined DRG: Risk of Mortality Subclass)
Total
Frequency
Row Pct
Col Pct
1
2
3
4
1
122
37.54
27.79
125
38.46
22.08
54
16.62
16.22
24
7.38
18.75
325
2
6
54.55
1.37
5
45.45
0.88
0
0.00
0.00
0
0.00
0.00
11
3
24
21.43
5.47
47
41.96
8.30
25
22.32
7.51
16
14.29
12.50
112
4
81
32.02
18.45
109
43.08
19.26
49
19.37
14.71
14
5.53
10.94
253
5
15
17.05
3.42
27
30.68
4.77
37
42.05
11.11
9
10.23
7.03
88
6
47
26.86
10.71
55
31.43
9.72
57
32.57
17.12
16
9.14
12.50
175
7
78
27.56
17.77
98
34.63
17.31
73
25.80
21.92
34
12.01
26.56
283
8
13
28.89
2.96
23
51.11
4.06
7
15.56
2.10
2
4.44
1.56
45
9
26
33.33
5.92
34
43.59
6.01
11
14.10
3.30
7
8.97
5.47
78
10
27
28.13
6.15
43
44.79
7.60
20
20.83
6.01
6
6.25
4.69
96
Total
439
566
333
128
1466
opposed to 3612-3614, still has the second highest proportion of patients in the highest mortality index;
in contrast, hospital #10 with a much higher proportion of more difficult surgeries, has a small percentage
of under 5% in the highest mortality category. Hospital #7 has most of its patients in the more difficult
surgeries and has the highest proportion in the most severe mortality category. For severity, hospitals
#1 and #7 again have the highest proportions in the most severe categories in spite of performing very
different types of surgery. There appears to be some definite shifting of patients into more severe cat-
egories, especially by hospital #1.
Figure 26 shows the results of the predictive model we used to predict mortality. It shows that the
optimal model is rule induction; however, the decision tree is very similar (Figure 27)
For the decision tree, congestive heart failure is first followed by the APRDRG mortality index.
Search WWH ::
Custom Search