Table 4.
Prediction performance of the IIgANPT with or without race applied to the analytic cohort alone, or in combination with microRNA (miR)-204
| C-stat (95% CI) | Δ C-stat (95% CI) | Event NRI (95% CI) | Nonevent NRI (95% CI) | IDI (95% CI) | |
|---|---|---|---|---|---|
| Model with race | 0.81 (0.76, 0.85) | Ref | Ref | Ref | Ref |
| Plus miR-204 | 0.90 (0.82, 0.96) | 0.09 (0.04, 0.15) | 0.64 (0.17, 0.96) | 0.54 (0.17, 0.82) | 0.35 (0.25, 0.44) |
| Model without race | 0.81 (0.76, 0.85) | Ref | Ref | Ref | Ref |
| Plus miR-204 | 0.90 (0.82, 0.95) | 0.10 (0.04, 0.15) | 0.64 (0.17, 0.96) | 0.54 (0.17, 0.82) | 0.32 (0.22, 0.41) |
CI, confidence interval; IDI, integrated discrimination improvement; IIgANRPT, International IgAN Risk Prediction Tool; NRI, net reclassification improvement; Ref= reference model.
The change (Δ) in C-statistic compares the IIgANRPT model with the biomarker to the IIgANRPT model alone. Prediction performance was evaluated using the 4-year risk of a 50% decline in estimated glomerular filtration rate or end-stage kidney disease. Significant net reclassification improvement and integrated discrimination improvement values are those whose 95%CI do not include zero.