Skip to main content

Table 3 Exploratory analysis

From: Palaeolithic diet decreases fasting plasma leptin concentrations more than a diabetes diet in patients with type 2 diabetes: a randomised cross-over trial

 

Pearson’s correlation ra

Spearman’s correlation rb

Adjusted R2

p c

Leptin (ng/ml) versus

 Fasting insulin (ng/ml)

 

0.555

 

0.049

 Fat (g)

 

−0.665

 

0.013

 Fat (%)

 

−0.555

 

0.049

 SAF (g)

 

−0.593

 

0.033

 Fatty acid C16:0 (g)

 

−0.571

 

0.041

 Fatty acid C18:0 (g)

 

−0.555

 

0.049

Glucagon (pg/ml) versus

 AUC insulin0-120 (nmol/l min)

0.946

 

0.383

0.015

 Stimulated AUC insulin0-120 (nmol/l min)

0.558

 

0.249

0.047

 Fasting plasma insulin (pmol/l)

0.637

 

0.352

0.019

 Satiety quotient for glycaemic index per meal (RS)

−0.562

 

0.254

0.045

 Glycaemic load (g)

0.63

 

0.342

0.021

 Glycaemic index

0.731

 

0.491

0.005

 Fatty acid C20:5, n-3, EPA (g)


0.581

 

0.277

0.037

 Fatty acid C22:6, n-3, DHA (g)

0.575

 

0.27

0.04

 Vitamin B-12 (µg)

0.571

 

0.265

0.041

Adipsin (ng/ml) versus

 Satiety quotient for GL per meal (RS/kg)

0.581

 

0.277

0.037

GIP (pg/ml) versus

 Fasting plasma insulin (pmol/l)

 

0.555

 

0.049

GLP-1 (pg/ml) versus

    

 Stimulated AUC insulin (nmol/l min)

 

0.654

 

0.015

 AUC insulin (nmol/l min)

 

0.67

 

0.012

Resistin (ng/ml) versus

 Fasting plasma insulin (pmol/l)

−0.728

 

0.451

0.041

 Satiety quotient for GL per meal (RS/kg)

0.810

 

0.598

0.015

Visfatin (ng/ml) versus

 Fasting plasma glucose (mmol/l)

−0.557

 

0.248

0.048

  1. Exploratory analysis was conducted to check for significant correlations between the outcome variables presented in Table 2 and outcome variables related to glucose homeostasis and satiation. Bivariate correlations were also performed between the outcome variables presented in Table 2 and dietary variables. This analysis consisted in bivariate Pearson or Spearman’s (for non normally distributed variables) correlation between within-subject differences in outcome and dietary variables. Normally distributed outcomes that were significant in Pearson’s correlation were entered into simple linear regression
  2. aPearson’s correlation for normally distributed variables
  3. bSpearman’s correlation for non normally distributed variables
  4. c p value for bivariate correlation and simple linear regression