Skip to main content

Table 4 Evaluation of the predictive power of smoking and alcohol variables on a high TyG index

From: Concurrent smoking and alcohol consumers had higher triglyceride glucose indices than either only smokers or alcohol consumers: a cross-sectional study in Korea

 

KNHANES (2013–2018)

P

KOICA (2012–2016)

P

Multivariate model

Multivariate model

Multivariate model

Multivariate model

Without S and Aa

With S and Ab

Without S and Ac

With S and Ad

AUC (95 CI)

0.817 (0.805–0.829)

0.829 (0.818–0.841)

<.001

0.822 (0.811–0.834)

0.826 (0.815–0.838)

<.001

NRI (95 CI)

Reference

0.040 (0.014–0.050)

<.001

Reference

0.025 (0.007–0.043)

.006

IDI (95 CI)

Reference

0.017 (0.006–0.013)

<.001

Reference

0.005 (0.003–0.008)

<.001

  1. aWithout S and A model = age + sex + final education + SBP + WC + TC + HDL
  2. bWith S and A model = age + sex + final education + SBP + WC + TC + HDL + S and A
  3. cWithout S and A model = sex + SBP + BMI + TC + HDL + regular exercise
  4. dWithout S and A model = sex + SBP + BMI + TC + HDL + regular exercise + S and A
  5. The high TyG index group comprises subjects with a TyG index ≥8.8
  6. KNHANES Korean National Health and Nutrition Examination Survey, KOICA Korean Initiatives on Coronary Artery Calcification, S and A Smoking and alcohol consumption, AUC The area under the receiver operating characteristic (ROC) curve, NRI Net reclassification improvement, IDI Integrated discrimination improvement, WC Waist circumference, BMI Body mass index, TC Total cholesterol, HDL High density lipoprotein