Phyto-oestrogen intake in Scottish men: use of serum to validate a self-administered food-frequency questionnaire in older men

Objective: To study dietary intake and serum concentrations of isoflavones in order to provide relative validation of isoflavone intake estimates from the Scottish Collaborative Group - Food-Frequency Questionnaire (SCG-FFQ). Design: Validation study. Setting: Southern Scotland. Method: Dietary intake of isoflavones was estimated using the semiquantitative SCG-FFQ and rank correlation and Kappa statistics were used for the relative validation of intakes against serum isoflavone concentrations in 203 male participants who were population controls in a case - control study of diet and prostate cancer. Results: The median intake of isoflavones ( daidzein and genistein) was 1.0 mg/day (I-QR 0.6 - 1.8). The median serum concentration of genistein was 33.79 nmol/ l ( I-QR 14.12 - 64.93), nearly twice that of daidzein ( 18.00 nmol/ l, I- QR 8.26 - 29.45). Equol was detected in 49% of subjects; in these subjects the median was 0.67 nmol/ l ( I- QR 0.34 - 1.51). Isoflavone intake was significantly correlated with serum concentrations of daidzein (rho = 0.24, P = 0.001), genistein (rho = 0.26, P<0.001) and total isoflavonoids ( sum of daidzein, genistein and equol) (rho = 0.27, P<0.001). Whereas values for weighted Kappa ranged from 0.16 ( P = 0.002) for daidzein and equol combined to 0.22 (P<0.001) for genistein. Conclusions: These results demonstrate the suitability of the SCG-FFQ to rank usual isoflavone intakes in older Scottish men, a population observed to have low consumption of soy foods.