Table 3.
Net reclassification improvement for adding VACS Index values at HI and 6M and their interaction to the index value at one year after HI. Cells show numbers of patients reclassified in each group, using inverse probability weighting to account for censoring (blue indicates correctly reclassified, red incorrectly reclassified patients, green indicates no change).
VACS Index at 1Y alone | VACS Index at HI, 6M, 1Y and VACS Index_HI*VACS Index_6M | ||||
---|---|---|---|---|---|
Tertiles of (Cox) Predicted Mortality | <2.28% | 2.28-4.36% | >4.36% | Total (%) | |
Died during follow-up | <2.28% | 8.2 | 5.9 | 0 | 14.1 (29%) |
2.28-4.36% | 0 | 2.5 | 12.6 | 15.1 (32%) | |
>4.36% | 0 | 0 | 18.6 | 18.6 (39.0%) | |
Total (%) | 8.2 (17.2%) | 8.4 (17.6%) | 31.2 (65.2%) | 47.8 (100%) | |
Did not die | <2.28% | 734.4 | 73.6 | 6.9 | 814.9 (39%) |
2.28-4.36% | 16.7 | 55.5 | 81.9 | 154.1 (33%) | |
>4.36% | 1.4 | 4.2 | 68.0 | 73.6 (28%) | |
Total (%) | 751.5 (72.1%) | 133.3 (12.8%) | 157.8 (15.1%) | 1042.6 (100%) |