Is there a non-linear relationship between dietary protein intake and prostate-specific antigen: proof from the national health and nutrition examination survey (2003–2010)

Table 3 Nonlinearity addressing by weighted two-piecewise linear model

Outcome	PSA(ng/ml) log2 transform β (95% CI)
Fitting by weighted linear regression model	0.02 (-0.02, 0.06) 0.2837
Fitting by weighted two-piecewise linear regression model
Inflection point	18.18
< 18.18	-0.03 (-0.09, 0.02) 0.2721
≥18.18	0.22 (0.07, 0.36) 0.0040
Log likelihood ratio test	0.002

Noted:Independent variable is dietary protein intake per 10 change and dependent variable is PSA(ng/ml log2 transform
Covariates involved in this model was the same as GAM model presented in Table 2

ISSN: 1476-511X