Skip to main content
. 2021 May 13;31:11. doi: 10.1186/s12610-021-00129-5

Table 3.

Univariate analysis and multivariate analysis for presence of ED

Items UVA MVA1a,c MVA2b,c
OR 95% CI P-value OR 95% CI P-value OR 95% CI P-value
TT, ng/mL 0.987 0.968–1.006 0.135
BMI, kg/m2 1.173 1.046–1.300 0.022 1.201 1.114–1.288 0.005 1.221 1.120–1.323 0.001
MetS (Yes vs. No) d 1.213 1.197–1.231 0.005 1.323 1.265–1.383 0.001 1.354 1.272–1.436 0.002
NO, μmol/L 0.679 0.603–0.756 0.027 0.693 0.611–0.712 0.035 0.671 0.626–0.720 0.007
E-selectin, ng/mL 2.022 1.895–2.147 0.000 1.765 1.631–1.898 0.000 1.739 1.682–1.800 0.002
Diet (PDI)e 0.779 0.721–0.837 0.001 0.869 0.795–0.943 0.004
Diet (hPDI)e 0.602 0.546–0.661 0.001 0.784 0.690–0.878 0.000

Abbreviations: BMI body mass index, CI confidence interval, ED erectile dysfunction, hPDI healthful plant-based diet index, MetS metabolic syndrome, MVA multivariate analysis, NO nitric oxide, OR odds ratio, PDI overall plant-based diet index, TT total testosterone, UVA univariate analysis

aMultivariate analysis included PDI

bMultivariate analysis included hPDI

c Adjusting for age, partner age, duration of partnership, frequency of intercourse, residence, occupational and educational status, income, and lifestyle

d MetS was defined following the criterion shown in Supplementary Table 2, in Additional file 1

e PDI and hPDI were converted from FFQ following the method described in the method section