Values under the detection limits were recorded as half of the detection limit. The urinary concentrations were normalized to urinary creatinine amounts (g-Cr). To enable a statistical analysis comparison, the urinary concentrations were transformed into common logarithmic values because the distributions were not normal. EQPs were defined according to the criteria proposed by Ideno et al. [18 (link)]: log(equol/daidzein) ≥ −1.42 indicates EQP status, and log(equol/daidzein) < −1.42 indicates non-EQP status.
Urinary zinc concentrations between EQP and non-EQP were compared by Student’s t-test. To adjust the potential confounders (age, BMI, etc.), we conducted multivariable regression analyses. Records with missing values were excluded from the main analyses while the missing values (smoking: 19, menstrual status: 13) were processed by single imputation methods (mode imputations) as additional analyses. JMP Pro (ver. 15; SAS Institute, Cary, NC) was used to perform these calculations. The alpha level for all tests was 0.05.
Urinary zinc concentrations between EQP and non-EQP were compared by Student’s t-test. To adjust the potential confounders (age, BMI, etc.), we conducted multivariable regression analyses. Records with missing values were excluded from the main analyses while the missing values (smoking: 19, menstrual status: 13) were processed by single imputation methods (mode imputations) as additional analyses. JMP Pro (ver. 15; SAS Institute, Cary, NC) was used to perform these calculations. The alpha level for all tests was 0.05.
Free full text: Click here
