Skip to main content

Table 4 Reclassification and discrimination statistics for predicting CKD by adding baseline TyG index, TyG index variability and cumulative TyG index

From: Association of long-term triglyceride-glucose index patterns with the incidence of chronic kidney disease among non-diabetic population: evidence from a functional community cohort

 

C-statistic

P

NRI

P

IDI

P

Men

      

Clinical risk model a

0.610 (0.590, 0.629)

     

+baseline TyG index

0.610 (0.590, 0.630)

0.690 b

-0.130 (-0.236, -0.025)

0.015 b

-0.17% (-0.36%, 0.01%)

0.060 b

+baseline TyG index and TyG index variability

0.633 (0.614, 0.653)

0.008 c

0.240 (0.135, 0.345)

< 0.001 c

0.85% (0.36%, 1.35%)

< 0.001 c

+baseline TyG index and cumulative TyG index

0.629 (0.609, 0.648)

0.006 c

0.097 (-0.009, 0.202)

0.072 c

0.42% (0.02%, 0.81%)

0.039 c

+baseline TyG index, TyG index variability and cumulative TyG index

0.649 (0.629, 0.668)

< 0.001 c

0.309 (0.204, 0.414)

< 0.001 c

1.50% (0.88%, 2.12%)

< 0.001 c

 

0.007 d

0.088 (-0.017, 0.194)

0.101 d

0.42% (0.00%, 0.83%)

0.048 d

 

0.010 e

0.243 (0.138, 0.348)

< 0.001 e

0.95% (0.45%, 1.45%)

< 0.001 e

Women

      

Clinical risk model a

0.612 (0.593, 0.630)

     

+baseline TyG index

0.615 (0.597, 0.633)

0.318 b

0.143 (0.039, 0.246)

0.007 b

0.17% (0.02%, 0.32%)

0.027 b

+baseline TyG index and TyG index variability

0.629 (0.611, 0.647)

0.019 c

0.189 (0.084, 0.294)

< 0.001 c

0.38% (0.12%, 0.64%)

0.004 c

+baseline TyG index and cumulative TyG index

0.623 (0.605, 0.641)

0.107 c

0.156 (0.051, 0.261)

0.004 c

0.22% (0.02%, 0.42%)

0.035 c

+baseline TyG index, TyG index variability and cumulative TyG index

0.636 (0.617, 0.653)

0.004 c

0.233 (0.129, 0.338)

< 0.001 c

0.64% (0.29%, 0.99%)

< 0.001 c

 

0.152 d

0.178 (0.073, 0.283)

< 0.001 d

0.26% (0.04%, 0.48%)

0.023 d

  

0.026 e

0.192 (0.087, 0.297)

< 0.001 e

0.42% (0.14%, 0.70%)

0.003 e

  1. CKD: chronic kidney disease, TyG: triglyceride-glucose, NRI: net reclassification improvement, IDI: integrated discrimination improvement
  2. a The clinical risk model for incident CKD include age, SBP, DBP, WC, FBG, serum gamma-glutamyl transferase, ALT, AST, TC, HDL-C, LDL-C (only for men), TG, HbA1c, eGFR, BMI, alcohol intake, smoking status, physical activity, past medical history (heart disease, stroke, hypertension, diabetes mellitus (only for women), and hyperlipidemia), and family history (heart disease, stroke, and hypertension). When no information of predictors was available in this cohort, a fixed value for the “missing variables” (i.e., “0” for categorial variables and a fixed number for continuous variables) was used for analysis
  3. b Compared to clinical risk model
  4. c Compared to clinical model with baseline TyG index added
  5. d Compared to clinical model with baseline TyG index and TyG index variability added
  6. e Compared to clinical model with baseline TyG index and cumulative TyG index added