Skip to main content

Table 5 Improvement in predicting prediabetes by adding the indices to the conventional risk factors

From: Elevated triglyceride-glucose (TyG) index predicts incidence of Prediabetes: a prospective cohort study in China

Prediabetes

C statistic

IDI

NRI

Estimate (95% CI)

P

Estimate(95% CI), %

P

Estimate (95% CI), %

P

Conventional model

0.584 (0.564–0.603)

 

Ref.

 

Ref.

 

+ BMI

0.599 (0.579–0.619)

0.0049

0.61 (0.36–0.86)

<  0.0001

17.16 (10.21–24.11)

<  0.0001

+ WC

0.598 (0.579–0.618)

0.0076

0.60 (0.35–0.85)

<  0.0001

15.46 (8.51–22.41)

<  0.0001

+ TG

0.609 (0.589–0.628)

<  0.0001

1.03 (0.68–1.38)

<  0.0001

24.70 (17.89–31.50)

<  0.0001

+  1/HDL-C

0.599 (0.579–0.619)

0.0054

0.61 (0.35–0.86)

<  0.0001

19.76 (12.82–26.69)

<  0.0001

+ TG/HDL-C

0.603 (0.584–0.622)

<  0.0001

0.73 (0.43–1.02)

<  0.0001

25.44 (18.72–32.15)

<  0.0001

+ TyG

0.622 (0.603–0.641)

<  0.0001

1.89 (1.44–2.33)

<  0.0001

28.76 (21.84–35.67)

<  0.0001

+ Mets-IR

0.613 (0.593–0.633)

0.0001

1.35 (0.97–1.73)

<  0.0001

21.93 (15.00–28.87)

<  0.0001

+ FPG

0.620 (0.604–0.642)

<  0.0001

1.68 (1.29–2.08)

<  0.0001

27.67 (20.82–34.52)

<  0.0001

  1. Conventional risk factors included age, gender, family history of diabetes, cigarette smoking, alcohol drinking, education level, hypertension and CVD
  2. IDI integrated discrimination improvement, NRI net reclassification index