Table 3: Classification analysis using Hosmer-Lemeshow goodness-of-fit test for EuroSCORE II of the n=621 Greek cardiac surgery patients who participated in the study. Columns present observed and expected cases according to the estimated risk, divided in 10 percentiles groups (1: 0-10%, 2: 11-20%, etc).
Estimated Risk Class |
Patients Who Died |
Patients Who Were Alive |
Total Patients |
|
Observed |
Expected1 |
Observed |
Expected1 |
|
0-10% |
0 |
1.20 |
63 |
61.80 |
63 |
11-20% |
0 |
1.22 |
63 |
61.78 |
63 |
21-30% |
1 |
1.24 |
62 |
61.76 |
63 |
31-40% |
0 |
1.24 |
62 |
60.76 |
62 |
41-50% |
1 |
1.26 |
61 |
60.74 |
62 |
51-60% |
0 |
1.29 |
62 |
60.71 |
62 |
61-70% |
1 |
1.34 |
61 |
60.66 |
62 |
71-80% |
2 |
1.40 |
59 |
59.60 |
61 |
81-90% |
4 |
1.53 |
57 |
59.47 |
61 |
91-100% |
9 |
6.27 |
53 |
55.73 |
62 |
Overall, n
|
18 |
17.99 |
603 |
603.01 |
621 |
1Note: the expected proportion of cases is computed using the coefficients from the applied logistic regression; the Hosmer-Lemeshow test value was 10.9, (p = 0.21), suggesting good fit of the observed to the expected values (Ho: Null hypothesis).