Skip to main content

Table 3 Associations of urinary BPA, BPS, and BPF levels with NAFLD

From: Serum glycolipids mediate the relationship of urinary bisphenols with NAFLD: analysis of a population-based, cross-sectional study

 

Tertile 1

Tertile 2

Tertile 3

Per Log unit increase

 

OR(95%CI)

p

OR(95%CI)

p

OR(95%CI)

p

BPA

 Model 1

1 (ref)

1.25(0.912,1.714)

0.166

1.693(1.231,2.327)

0.001

1.531(1.158,2.025)

0.003

 Model 2

1 (ref)

1.49(1.045,2.125)

0.028

1.786(1.255,2.542)

0.001

1.568(1.147,2.144)

0.005

 Model 3

1 (ref)

1.43(0.999,2.047)

0.051

1.581(1.100,2.274)

0.013

1.391(1.007,1.921)

0.045

BPS

 Model 1

1 (ref)

1.339(0.975,1.838)

0.071

1.988(1.444,2.738)

 < 0.001

1.573(1.253,1.974)

 < 0.001

 Model 2

1 (ref)

1.346(0.948,1.909)

0.096

1.970(1.381,2.811)

 < 0.001

1.576(1.225,2.026)

 < 0.001

 Model 3

1 (ref)

1.253(0.876,1.792)

0.216

1.799(1.246,2.597)

0.002

1.476(1.139,1.913)

0.003

BPF

 Model 1

1 (ref)

1.054(0.779,1.425)

0.733

1.200(0.871,1.655)

0.265

1.227(0.976,1.541)

0.079

 Model 2

1 (ref)

1.131(0.808,1.582)

0.473

1.344(0.937,1.929)

0.108

1.280(0.989,1.656)

0.061

 Model 3

1 (ref)

1.056(0.749,1.489)

0.755

1.192(0.823,1.726)

0.353

1.165(0.895,1.517)

0.256

  1. Model 1 is adjusted urine creatinine (tertiles); Model 2 is adjusted for ethnicity (non-Hispanic white, non-Hispanic black, non-Hispanic Asian, Mexican American, Other Hispanic), education (less than high school, high school or equivalent, college or above), drinking (no or unknown, yes), hypertension (no, yes), diabetes (no, yes), log-transformed levels of triglyceride, high-density lipoprotein cholesterol, glucose, glycosylated hemoglobin A1c and urine creatinine (tertiles), Model 3 is adjusted variables in Model 2 plus log-transformed concentration of BPA and/or BPS. BPA bisphenol A, BPS bisphenol S, BPF bisphenol F, OR odds ratio, CI confidence intervals, p < 0.05 suggests significant differences