Age at Menopause and Dementia Risk: Blood-Based Biomarker Evidence from a US National Study
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Importance
Earlier age at menopause associates with increased later life dementia risk, but the biological links that underlie this relationship remain poorly understood in population-based samples.
Objective
To evaluate the relationship between history of menopausal timing and AD/ADRD-related blood-based biomarkers in postmenopausal women.
Design
Observational study using the U.S. Health and Retirement Study. AD/ADRD-related biomarkers were measured from venous blood draws conducted in 2016 and released in 2025. We estimated linear and nonlinear associations between spontaneous menopause timing and biomarkers, accounting for sociodemographic and health covariates. We evaluated effect modification by APOE 4 carrier status. Secondary analyses evaluated associations with hysterectomy history.
Setting
Population-based.
Participants
Community-dwelling post-menopausal women aged 65 years or older.
Exposure
Primary: Self-reported spontaneous age of menopause among women with no history of hysterectomy. Secondary: Self-reported history of hysterectomy (oophorectomy history was not assessed).
Main Outcomes and Measures
Neurofilament light chain (NfL) and glial fibrillary acidic protein (GFAP) measured in plasma, and phosphorylated tau (p-tau181) and growth differentiation factor 15 (GDF-15), measured in serum, by Simoa or ELLA.
Results
Among 1449 women ≥ 65 years, 38.4% self-reported a hysterectomy. Among 893 postmenopausal women without a history of hysterectomy (mean [SD] age at biomarker collection, 75.2 [7.2] years; mean age at menopause, 50.2 [5.2] years), menopausal timing was nonlinearly associated with NfL. Between menopausal ages 45–50, menopause timing was associated with higher NfL (0.010 SD units/year, [95% CI: −0.007, 0.027]), while from menopausal ages 50–55, the association reversed ( b = −0.031, [95% CI: −0.051, −0.011]). Associations with GFAP or pTau-181 were not significant, but were directionally similar to those observed for NfL. Later menopausal timing linearly associated with lower GDF-15 ( b = −0.015, [95% CI: −0.027, −0.004]). Self-reported hysterectomy was associated with higher GDF-15 ( b = 0.118, [95% CI: 0.015, 0.220]), but not other biomarkers. Associations did not significantly differ by APOE ε4 status.
Conclusions and Relevance
In a nationally representative study, earlier age at menopause in midlife was associated with later life AD/ADRDs biomarkers reflecting neuroaxonal degeneration and inflammation, underscoring the importance of understanding reproductive aging in dementia risk.
KEY POINTS
Question
What is the association between menopausal timing and blood-based Alzheimer’s disease and related dementias (AD/ADRD) biomarkers among postmenopausal women 65 years and older in the United States?
Findings
Menopausal timing is associated with higher NfL in women who experienced menopause before age 50, but lower NfL in women with menopausal ages after 50. Earlier menopausal timing is linearly associated with higher GDF-15. Secondary analyses showed associations between hysterectomy history and higher GDF-15.
Meaning
Younger menopausal timing associates with higher AD/ADRD-related biomarkers measured decades later in a population-based cohort, underscoring a link between reproductive aging and future neurodegenerative risk.