Validity of self-reported menstrual cycle length

PURPOSE: Self-reported menstrual cycle length has been associated with host and environmental factors and chronic disease risk. The purpose of this study is to evaluate the validity of self-reported cycle length. METHODS: The authors assessed the agreement between a woman's self-reported "usual" cycle length at study onset with the mean of her observed cycle lengths from prospective daily diaries for 398 women aged 19 to 41 years in the Mount Sinai Study of Women Office Workers (1990 to 1994). RESULTS: Forty-three percent of women self-reported usual cycle lengths more than 2 days different from their mean length. When self-reported cycle length was categorized (<26, 26 to 35, and >35 days) and compared with mean cycle length, 21% of women were misclassified. Women who were older, married, and with higher income were more likely to have accurately reported their menstrual cycle length. Women who had short or long mean cycle lengths (lowest and highest quintile of length) were less likely to self-report accurately, and accuracy decreased monotonically with increasing cycle variability. CONCLUSIONS: These findings show considerable measurement error in self-reported cycle length, as well as describe population subgroups that report menstrual cycle length with the greatest accuracy.