Intake of soybean protein was associated with a reduced risk of breast cancer in a case–control study. It has also been demonstrated to increase menstrual cycle length in an experimental setting.

To ascertain whether the association of soybean protein intakes with menstrual cycle length persists in an uncontrolled community setting.

Cross-sectional food frequency dietary survey, menstrual cycle survey and prospective collection of menstrual cycle data.

A hospital clinic and a nursing college.

Two hundred menstruating women.

An association (P=0.034) of higher intakes of soybean protein with increased menstrual cycle length, as recorded by self report and by prospectively recording three consecutive cycles, was observed. The risk of menstrual cycle length being greater than the median, when comparing the upper quartile (8.7–35.2 g day−1) of soybean intake and the lowest quartile (0.1–3.3 g day−1) was double, and this approached statistical significance (OR=2.02, 95% CI=0.88−4.64 and OR=1.93, 95% CI=0.82−4.56 for self-reported cycle length and cycle length as recorded by diary, respectively). In terms of the absolute association with cycle length, subjects in the upper quartile of soybean intake demonstrated a cycle length 1–2 days longer than did subjects in the lowest quartile.

It is likely that the association between dietary intake of soybean protein and length of menstrual cycle prevails in the community setting. This is shown using both self-reported cycle length and cycle length as recorded in a prospective diary.