Table 2.
Specificity | |||
---|---|---|---|
Model | 97.5% | 95% | 90% |
Median time to progression (years [95% CIs]) (10 tests) | |||
Hi-linear | 6.5 [6.2, 6.8] | 5.7 [5.4, 5.9] | 5.1 [4.9, 5.3] |
Hi-censored | 6.2 [6.0, 6.5] | 5.6 [5.4, 5.9] | 5.1 [4.9, 5.2] |
Hi-HSK | 6.3 [6.0, 6.6] | 5.6 [5.3, 5.8] | 5.0 [4.8, 5.2] |
Simple linear | 11 [10.7, 11.4] | 9.3 [9.1, 9.7] | 7.5 [7.2, 7.8] |
PoPLR (p-value) | 10.1 [9.8, 10.4] | 8.6 [8.3, 8.9] | 6.9 [6.7, 7.2] |
PoPLR (S) | 7.4 [7.1, 7.7] | 6.7 [6.4, 6.9] | 5.8 [5.6, 6] |
Hit-rate (% [95% CIs]) (10 tests) | |||
Hi-linear | 53 [51, 53] | 57 [55, 57] | 61 [59, 61] |
Hi-censored | 54 [51, 53] | 57 [55, 57] | 61 [59, 61] |
Hi-HSK | 52 [51, 52] | 55 [54, 55] | 59 [57, 59] |
Simple linear | 32 [29, 32] | 37 [35, 37] | 45 [43, 45] |
PoPLR (p-value) | 35 [33, 36] | 41 [39, 41] | 50 [47, 49] |
PoPLR (S) | 49 [47, 49] | 52 [50, 52] | 56 [55, 56] |
Hi = Hierarchical; HSK = Heteroskedastic.
The 95% confidence intervals (CIs) were obtained from the Kaplan Meier estimator for the median time to progression and via bootstrap for the hit-rate.