Skip to main content

Table 3 Applying logistic regression to assess the significance of relationship between Independent demographic and genetic variables and metabolic syndrome

From: Evaluating machine learning-powered classification algorithms which utilize variants in the GCKR gene to predict metabolic syndrome: Tehran Cardio-metabolic Genetics Study

Variables

B

Odds ratio (OR)

P value

Age

0.025

1.025

 < 0.001

Gender (female = 0)

0.864

2.373

 < 0.001

Schooling years

− 0.021

0.978

0.009

BMI

0.207

1.230

 < 0.001

Physical activity

0.0001

1.000

0.005

Smoking status

Current smoker(reference)

   

Never smoker

0.005

1.005

0.962

Former smoker

0.072

1.075

0.698

Second hand

0.646

1.066

0.593

Marital status

Divorced (reference)

   

Married

− 0.087

0.916

0.808

Single

− 0.198

0.820

0.595

Widowed

− 0.010

0.989

0.981

rs1260326

CC(reference)

   

TC

0.206

1.229

0.347

TT

0.472

1.603

0.133

rs780094

CC(reference)

   

TC

0.149

1.161

0.664

TT

− 1.211

0.298

0.008

rs780093

CC(reference)

   

TC

− 0.122

0.884

0.664

TT

1.066

2.903

0.002

  1. BMI Body Mass Index; logistic regression is used to predict the metabolic syndrome status of the participants in TCGS. The metabolic syndrome was significantly associated with age, gender, schooling years, BMI, physical activity, rs780094, and rs780093 (P < 0.05)