New longitudinal evidence suggests reproductive history may shape women’s cognitive aging, offering insights into possible protective pathways while highlighting the need for cautious interpretation.
Study: Pregnancy and breastfeeding are associated with less later-life cognitive decline in a longitudinal, prospective cohort. Image Credit: ibragimova / Shuttertock
A recent study published in the journal Alzheimer’s & Dementia found that breastfeeding and pregnancy are associated with higher later-life cognitive scores over time rather than necessarily slower cognitive decline, although the clinical significance of these modest differences remains uncertain.
Background: Pregnancy, Lactation, and the Female Brain
Female mammals, including humans, undergo structural and functional brain changes during pregnancy and lactation that prepare them for motherhood. However, the long-term consequences of these changes for neurodegenerative disease risk or cognitive resilience remain unclear. Females often experience cognitive deficits, particularly in verbal memory, during pregnancy, which may persist for up to 32 weeks postpartum.
Despite these short-term effects, long-term outcomes of motherhood appear more favourable for later-life cognitive trajectories, although findings across studies are mixed. Postmenopausal females who breastfed have been reported to show superior cognitive performance compared with those who never breastfed for at least one month. Rodent studies also demonstrate both short- and long-term cognitive benefits of motherhood that accumulate across pregnancies and persist into later life. Conversely, some observational studies have reported detrimental cognitive effects associated with breastfeeding, highlighting ongoing uncertainty.
Study Design and Data Sources
In this study, researchers examined whether greater exposure to pregnancy and breastfeeding was associated with better later-life cognitive trajectories rather than definitively slower decline. Data were drawn from the Women’s Health Initiative Memory Study (WHIMS) and the Women’s Health Initiative Study of Cognitive Aging (WHISCA).
WHIMS enrolled dementia-free postmenopausal females aged 65 to 79 years between 1995 and 1998. WHISCA included participants aged 66 to 84 years, drawn from WHIMS, who underwent annual assessments of mood and cognitive function. Participants with fewer than two cognitive assessments or who developed dementia within one year of baseline were excluded. Reproductive history data were obtained through retrospective baseline interviews.
Exposure Measures: Pregnancy and Breastfeeding History
The primary exposures were total time pregnant and total time breastfeeding. Secondary exposures included the breastfeeding-to-pregnancy ratio (BF:PREG), gravidity (ever versus never pregnant), and breastfeeding history (ever versus never). Total time pregnant was defined as the cumulative lifetime months of pregnancy, including estimated durations for non-livebirth pregnancies based on standardized assumptions. Total time breastfeeding was calculated as the sum of months of breastfeeding across the lifespan.
The BF:PREG ratio was calculated as the number of children breastfed divided by the number of pregnancies and was treated as a continuous variable. Values may exceed 1, reflecting variability in breastfeeding duration across pregnancies rather than a strict proportionality.
Cognitive Outcomes and Statistical Analysis
Outcomes included global cognition, verbal memory, and visual memory. Global cognitive function was assessed in WHIMS using the modified Mini-Mental State Examination (3MS). Verbal and visual memory were assessed in WHISCA using the California Verbal Learning Test long-delay free recall and the Benton Visual Retention Test, respectively.
Linear mixed-effects models were used to examine associations between reproductive exposures and cognitive outcomes. Minimally adjusted models included age and education, while fully adjusted models additionally accounted for race, income, apolipoprotein E ε4 carrier status, estrogen therapy use, and combined estrogen-progesterone therapy. Reciprocal adjustment for parity or breastfeeding history was applied to estimate independent effects.
Participant Characteristics and Follow-Up
The analysis included 6,083 participants from WHIMS and 1,935 from WHISCA, with follow-up of 10.4 and 8.4 years, respectively. Most participants were White, married or partnered, and not Hispanic or Latino. Approximately 7.1 percent had never been pregnant, and 9 percent had never experienced a full-term pregnancy. The average number of pregnancies was 3.8, corresponding to a mean total pregnancy duration of 30.5 months.
Breastfeeding history varied substantially. Forty-four percent of participants never breastfed for more than 1 month; among those who did, the average total duration of breastfeeding was 11.6 months.
Key Findings: Pregnancy, Breastfeeding, and Cognitive Scores
Pregnancy and breastfeeding were associated with higher cognitive scores across follow-up rather than with slower rates of cognitive decline. In fully adjusted models, each additional month of pregnancy was associated with a 0.01-point increase in global cognition scores.
Total time pregnant was not associated with visual or verbal memory. In contrast, each additional month of breastfeeding was associated with 0.01-0.02-point increases in global cognition, visual memory, and verbal memory scores. Ever-pregnant participants had a 0.6-point higher global cognition score compared with those who were never pregnant.
Participants who breastfed for at least one month had 0.19- and 0.27-point higher global cognition and verbal memory scores, respectively, than those who never breastfed. Associations with visual memory did not reach statistical significance. Each unit increase in the BF:PREG ratio was associated with 0.24-, 0.38-, and 0.54-point higher global cognition, visual memory, and verbal memory scores, respectively. Although statistically significant, these effect sizes were modest.
Conclusions and Interpretation
A greater number of pregnancies, longer breastfeeding duration, and higher BF:PREG ratios were associated with better global cognitive performance in postmenopausal females. Associations with memory domains were less consistent, with positive findings primarily linked to breastfeeding and the BF:PREG ratio rather than pregnancy alone.
Given the observational design, these findings should not be interpreted as causal. Residual confounding and reverse causation remain possible. The predominantly White, United States-based cohort limits generalisability, and the results should not inform reproductive decision-making. Future research should investigate biological mechanisms linking reproductive history to later-life cognitive health and determine whether these modest score differences translate into clinically meaningful protection against dementia.
MRI scans reveal human brain growth from pregnancy through birth