Skip to main content

Table 3 Results of MR analyses of lifestyle and glycemic traits

From: Diet/lifestyle and risk of diabetes and glycemic traits: a Mendelian randomization study

Lifestyle

MR Estimates

T2D

HOMA-IR

HOMA-B

INS

GLU

Hba1c

2hGLU

Beta ± SE

p

Beta ± SE

p

Beta ± SE

p

Beta ± SE

p

Beta ± SE

p

Beta ± SE

p

Beta ± SE

p

Carbohydrate

0.051 ± 0.045

0.262

0.000 ± 0.011

0.981

0.000 ± 0.009

0.990

0.001 ± 0.010

0.957

0.011 ± 0.010

0.282

0.003 ± 0.010

0.745

0.029 ± 0.052

0.582

Protein

0.806 ± 0.260

0.002

0.067 ± 0.035

0.046

0.045 ± 0.027

0.092

−0.047 ± 0.033

0.157

0.038 ± 0.029

0.203

0.049 ± 0.041

0.231

0.062 ± 0.141

0.662

Fat

−0.079 ± 0.070

0.261

0.007 ± 0.020

0.722

0.014 ± 0.016

0.400

0.003 ± 0.019

0.881

−0.007 ± 0.018

0.713

−0.020 ± 0.015

0.175

0.062 ± 0.095

0.519

Sleep

0.005 ± 0.003

0.058

0.001 ± 0.001

0.487

0.001 ± 0.001

0.128

0.000 ± 0.001

0.713

−0.001 ± 0.001

0.222

−0.001 ± 0.001

0.106

0.003 ± 0.004

0.534

Smoking cessation

0.559 ± 0.328

0.090

0.165 ± 0.072

0.021

0.064 ± 0.055

0.247

0.132 ± 0.066

0.047

0.132 ± 0.064

0.039

0.093 ± 0.058

0.107

0.314 ± 0.326

0.337

Smoking initiation

0.249 ± 0.380

0.515

0.079 ± 0.085

0.353

0.048 ± 0.071

0.503

0.069 ± 0.081

0.397

0.020 ± 0.077

0.799

0.141 ± 0.073

0.055

−0.541 ± 0.403

0.182

smoking (number)

−0.013 ± 0.014

0.357

0.001 ± 0.004

0.835

−0.003 ± 0.003

0.367

0.000 ± 0.004

0.914

0.002 ± 0.004

0.575

0.003 ± 0.003

0.379

−0.001 ± 0.018

0.944

  1. We pooled β coefficients across SNPs using fixed-effect meta-analysis. After meta-analysis, we used the IV estimators toquantify the strength of the causal association of diet/lifestyle anddiseases. The IV estimator which is identical to that derived by the widely used two-stage least squares method, will be calculated as the β of the regression coefficients SNP-exposure and SNP-outcome