Table 4.
Performance of the ICD-9-CM-based and alternative models for identifying CABG SSIs (training data: medical center A)
| Sensitivity | Specificity | PPV | NPV | Accuracy | |
|---|---|---|---|---|---|
| ICD-9-CM-based model |
37.50% (9/24) |
96.27% (956/993) |
19.57% (9/46) |
98.46% (956/971) |
94.89% (965/1017) |
| Model 1 |
4.17% (1/24) |
99.90% (992/993) |
50.00% (1/2) |
97.73% (992/1015) |
97.64% (993/1017) |
| Model 2 |
54.17% (13/24) |
96.78% (961/993) |
28.89% (13/45) |
98.87% (961/972) |
95.77% (974/1017) |
| Model 3 |
100.00% (24/24) |
3.22% (32/993) |
2.44% (24/985) |
100.00% (32/32) |
5.51% (56/1017) |
| Model 4 |
100.00% (24/24) |
94.56 (939/993) |
30.77% (24/78) |
100.00% (939/939) |
94.69% (963/1017) |
| Model 5 | 87.50% (21/24) | 99.40% (987/993) | 77.78% (21/27) | 99.70% (987/990) | 99.12% (1008/1017) |
% (numerator/denominator).
ICD-9-CM: The International Classification of Diseases, Ninth Revision, Clinical Modification; PPV: positive predictive value; NPV: negative predictive value. Model 1: the classification algorithms (strict); Model 2: the classification algorithms (moderate); Model 3: the classification algorithms (loose); Model 4: the multivariable regression model: Model 5: data mining-decision tree model.