BACKGROUND: Menstrual cycle characteristics are potential indicators of hormonal exposures and may also signal cardiovascular disease risk factors, both of which are relevant to cognitive health. However, there is scarce epidemiological evidence on the association between cycle characteristics and cognitive function. OBJECTIVE: We studied the associations of menstrual cycle characteristics at 3 stages of a woman's reproductive lifespan with cognitive function in midlife. STUDY DESIGN: We studied participants from the Nurses' Health Study II, an ongoing longitudinal cohort of female nurses initially enrolled in 1989. Exposures were cycle regularity at 14 to 17 and 18 to 22 years, and cycle length (the interval between 2 consecutive cycles) at 18 to 22 years (all retrospectively reported at enrollment), and current cycle regularity and length at 29 to 46 years (reported in 1993). Outcomes were composite z scores measuring psychomotor speed/attention and learning/working memory obtained with 1 self-administered Cogstate Brief Battery assessment, measured among a subset of participants in 2014 to 2022. We included 19,904 participants with data on at least 1 menstrual cycle characteristic and a cognitive assessment. We estimated mean differences (β, 95% confidence intervals) using linear regression models adjusted for age at cognitive assessment, race and ethnicity, participants' education, wave of cognitive assessment, parental education and occupation, neighborhood socioeconomic status, age at menarche, adiposity, oral contraceptive use, and lifestyle factors (smoking, alcohol intake, physical activity, diet quality). RESULTS: In the analytical sample, the mean (standard deviation [SD]) age at cognitive assessment was 62.0 (4.9) years. Women with irregular cycles at 29 to 46 years scored lower in learning/working memory (β, -0.05 SD
95% confidence interval, -0.08 to -0.01) than those with very regular cycles. We did not observe associations for cycle regularity at 14 to 17 or 18 to 22 years. Women with cycle length ≤25 days at 18 to 22 years scored lower in learning/working memory in later life (β, -0.05 SD
-0.09 to -0.02) than those with cycles 26 to 31 days. We did not observe associations of cycle length at 29 to 46 years with later cognitive function. In a secondary analysis, women whose cycles were regular at 14 to 17 or 18 to 22 years but became irregular by 29 to 46 years also had lower learning/working memory scores, compared to women whose cycles remained regular across time points. CONCLUSION: In this large longitudinal study, cycles ≤25 days at 18 to 22 years and irregular cycles at 29 to 46 years were associated with lower performance in learning/working memory. Future studies in other populations should confirm our findings and investigate the biological processes underlying these associations.