The association between soya consumption and serum thyroid-stimulating hormone concentrations in the Adventist Health Study-2

Objective Consumers may choose soya foods as healthful alternatives to animal products, but concern has arisen that eating large amounts of soya may adversely affect thyroid function. The present study aimed to examine the association between soya food consumption and serum thyroid-stimulating hormone (TSH) concentrations in North American churchgoers belonging to the Seventh-day Adventist denomination that encourages vegetarianism. Design Participants completed six repeated 24 h dietary recalls within a 6-month period. Soya protein and soya isoflavone intakes were estimated, and their relationships to TSH concentrations measured at the end of 6 months were calculated using logistic regression analyses. Setting Calibration sub-study of the Adventist Health Study-2. Subjects Women ( n 548) and men ( n 295) who were not taking thyroid medications. Results In men, age and urinary iodine concentrations were associated with high serum TSH concentrations (>5 mIU/l), while among women White ethnicity was associated with high TSH. In multivariate models adjusted for age, ethnicity and urinary iodine, soya isoflavone and protein intakes were not associated with high TSH in men. In women higher soya isoflavone consumption was associated with higher TSH, with an adjusted odds ratio (highest v . lowest quintile) of 4·17 (95 % CI 1·73, 10·06). Likewise, women with high consumption of soya protein (midpoint of highest quintile, 11 g/d) v . low consumption (midpoint of lowest quintile, 0 g/d) carried increased odds of high TSH (OR=2·69; 95 % CI 1·34, 5·30). Conclusions In women high consumption of soya was associated with elevated TSH concentrations. No associations between soya intake and TSH were found in men.