Connections between reproductive health and cognitive aging among women enrolled in the HCHS/SOL and SOL‐INCA

Abstract

AbstractBackgroundLifetime risk for Alzheimer’s disease is higher among women than men, and reproductive health factors may contribute to differential risk. Reproductive health and cognitive aging among Hispanic women/Latinas is understudied, so we examined such relationships in the Hispanic Community Healthy Study/Study of Latinos (HCHS/SOL) and the SOL‐Investigation of Neurocognitive Aging (SOL‐INCA).MethodThe sample included Hispanic/Latina post‐menopausal women (n = 1,903, baseline age 44‐74 years). Exposures included oral contraceptive use, number of live births (parity), number of pregnancies, age at menarche, age at menopause, and reproductive span (years with menses). Outcomes included visit 2 cognitive performance (7‐years after baseline on average), 7‐year cognitive change (reliable change index), and NIA‐AA criteria mild cognitive impairment (MCI) prevalence. Cognitive measures included verbal learning and memory, verbal fluency, executive functioning (Trail B, higher scores = worse performance), and processing speed (Digit Symbol Substitution = DSS; Trail A, higher scores = worse performance), and a global composite (average of all individual z‐scores). Survey linear regressions were used to separately model the associations between reproductive health measures and outcomes, controlling for baseline age, education, language preference, Hispanic/Latina background, field center, income, birthplace, marital status, insurance status, and body mass index.Result Performance at visit 2: Oral contraceptive use was associated with higher global cognitive performance (β = 0.129 [SE = 0.043], p<0.01), learning (β = 0.197 [SE = 0.062], p<0.01), and processing speed (βDSS = 0.147 [SE = 0.049], p<0.01 and βTrailA = ‐0.142 [SE = 0.071], p<0.05) than non‐use. Higher parity was associated with worse executive functioning (β = 0.035 [SE = 0.017], p<0.05) and processing speed (βDSS = ‐0.054 [SE = 0.015], p<0.001). 7‐year (average) cognitive change: Oral contraceptive use, higher parity, and higher pregnancies were each protective against change in learning (β = 0.147 [SE = 0.064], β = 0.040 [SE = 0.019], and β = 0.034 [SE = 0.014], respectively; all ps<0.05). Older age at menarche was protective against change in global cognition (β = 0.045 [SE = 0.016], p<0.01), learning (β = 0.035 [SE = 0.017], p<0.05) and memory (β = 0.037 [SE = 0.017], p<0.05). MCI: No exposures were associated with MCI.ConclusionReproductive health was linked to later life cognition among diverse Hispanic/Latina women. Future work should examine potential hormonal (e.g., estrogen) mechanisms. Although not associated with MCI In the present study, reproductive health may be linked to MCI status as the cohort ages and more women transition to MCI or dementia.