Although the association between the number of deliveries and cognitive impairment has been previously examined, the influence of subclinical cerebrovascular diseases (SCDs), such as silent cerebrovascular lesions and carotid atherosclerosis, on this association remains unclear. This cross-sectional study aimed to examine whether SCDs mediated the association between the number of deliveries and cognitive impairment. Among 627 Japanese women with a mean age of 73 years, the number of deliveries was collected in the 1998 survey and classified into four groups (0-1, 2, 3, ≥ 4), with two deliveries as the reference. At the annual comprehensive medical examinations, cognitive function was assessed using the Mini-Mental State Examination (MMSE), and SCDs were evaluated using brain magnetic resonance imaging and ultrasonography. Each participant's latest data on these variables and covariates between 1992 and 2018 were used. MMSE scores were divided into three ordinal categories: ≥ 28 (normal), 24-27 (mild cognitive impairment; MCI), and ≤ 23 (severe cognitive impairment). Ordinal logistic regression models were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for cognitive impairment. The ORs for cognitive impairment associated with the number of deliveries were 2.13 (95% CI, 1.21-3.76) in the lowest (0-1) group and 1.45 (0.95-2.23) in the highest (≥ 4) group. These association estimates remained similar after adjusting for SCDs but were weaker in the more recent birth year group. We demonstrated a U-shaped association between the number of deliveries and cognitive impairment, independent of SCDs, and the cohort effect confounded the association.
Keywords: atherosclerosis; cerebrovascular disease; cognitive impairment; delivery; women’s health.