Skip to main content

Table 4 Multiple logistic regression analysis models for the association between hyperuricemia (dependent variable) and insulin resistance surrogates (independent variables)

From: The association between hyperuricemia and insulin resistance surrogates, dietary- and lifestyle insulin resistance indices in an Iranian population: MASHAD cohort study

Variable

TyG quartiles

TyG-BMI quartiles

Q1

Q2

Q3

Q4

p-trend

Q1

Q2

Q3

Q4

p-trend

Model 1

1

2.95 (1.94–4.40) *

5.66 (3.85–8.32)

8.68 (5.96–12.65)

< 0.001

1

2.47 (1.70–3.59)

4.32 (3.03–4.15)

8.17 (5.79–11.52)

< 0.001

Model 2

1

2.45 (1.68–3.86)

4.49

(3.03–6.67)

6.81 (4.62–10.04)

< 0.001

1

2.52 (1.69–3.76)

4.51 (2.94–6.84)

8.53 (5.17–14.05)

< 0.001

Model 3

1

3.34 (1.91–5.87)

5.59 (3.25–9.61)

10.04 (5.85–17.24)

< 0.001

1

3.27 (1.95–5.51)

5.70 (3.25-10.00)

14.49 (7.33–28.63)

< 0.001

Variable

TG/HDL quartiles

METS-IR quartiles

Q1

Q2

Q3

Q4

p-trend

Q1

Q2

Q3

Q4

p-trend

Model 1

1

2.03 (1.42–2.89)

3.48 (2.50–4.84)

5.65

(4.11–7.77)

< 0.001

1

2.11 (1.48–3.01)

3.31 (02.37–4.63)

6.42 (4.66–8.84)

< 0.001

Model 2

1

1.81 (1.26–2.60)

2.98 (2.12–4.19)

4.65 (3.34–6.49)

< 0.001

1

1.95 (1.34–2.85)

2.94 (1.98–4.35)

5.27 (3.33–8.34)

< 0.001

Model 3

1

2.36 (1.44–3.86)

3.59 (2.24–5.74)

6.08 (3.82–9.68)

< 0.001

1

2.02 (1.25–3.28)

3.35 (2.05–5.56)

6.64 (3.64–12.11)

< 0.001

  1. Model1: adjusted for age, and sex; Model2: Model1 + BMI, energy intake, education level, smoking status, physical activity level; Model3: Model2 + chronic diseases including diabetes, hypertension, or dyslipidemia, and estimated glomerular filtration rate. TG/HDLc: triglyceride to high-density lipoprotein cholesterol ratio; TyG: the product of fasting triglycerides and glucose; BMI: body mass index; METS-IR: metabolic score for insulin resistance. Multiple logistic regression analysis was applied. *Shows odds ratio (confidence interval)