Table 4.

Diagnostic accuracy of prognostic markers for progression to ESRD in different populations selected for screeninga

Prognostic MarkerScreening Strategies
Diabetes/HypertensionDiabetes/Hypertension/Age >55 yrUK CKD GuidelinesbEverybody
TPR0.03pAUCTPR0.03pAUCTPR0.03pAUCTPR0.03pAUC
Best clinical model0.1550.6030.1910.6230.2250.6440.3390.704
ACR0.4590.7520.5200.7760.5440.7860.5350.786
eGFR0.4530.7540.4640.7570.5260.7870.6050.821
eGFR + ACR0.5790.8070.5960.8130.6390.8340.6600.844
eGFR + ACR + best clinical model0.6120.8200.6160.8220.6600.8420.6920.858
  • a TPR0.03, true-positive rate (i.e., sensitivity) at a fixed false-positive rate (FPR) of 0.03; pAUC, partial area under the clinically relevant part of the ROC curve (FPR 0.00 to 0.10) transformed to values between 0.5 and 1.0.35 Analogous to ordinary ROC analysis, a perfect test would have pAUC = 1.0, whereas a test with no ability to discriminate between those progressing to ESRD and those not progressing would have pAUC = 0.5. Best clinical model includes age, gender, physical activity, diabetes, systolic BP, antihypertensive treatment, and HDL cholesterol.

  • b British CKD guidelines recommend screening of individuals with hypertension, diabetes, autoimmune diseases, CVD, or postrenal obstruction.4