Heart health scores drop during perimenopause, a national analysis finds

The conventional account has cardiovascular risk in women rising after menopause — the protective effect of estrogen ends, and the risk curve bends upward. A study published in the Journal of the American Heart Association on May 13, 2026 puts the inflection point earlier, during the transition itself [s1].

What was measured

Researchers analyzed data from 9,248 women aged 18 to 80 who participated in the National Health and Nutrition Examination Survey between 2007 and 2020 [s1]. They scored each woman on Life's Essential 8, the American Heart Association's 100-point composite of diet, physical activity, tobacco use, sleep, blood pressure, cholesterol, body weight, and blood sugar. Scores below 50 count as low, 50 to 79 as intermediate, and 80 or above as ideal [s1].

Median scores fell across menopause stages: 73.3 among premenopausal women, 69.1 among perimenopausal women, and 63.9 among postmenopausal women [s1].

The comparison the study leads with is categorical. Perimenopausal women were twice as likely as premenopausal women to fall into the low overall cardiovascular health band [s1]. Two components drove it: perimenopausal women were 76% more likely to have a low cholesterol score and 83% more likely to have a low blood sugar score [s1]. Diet scored lowest of the eight components across every group [s1].

What the design can and cannot support

NHANES is a cross-sectional survey. Each participant is measured once. That means this study compares different women at different life stages, not the same women over time.

This matters more than it might appear. A cross-sectional gradient across menopause stages is consistent with cardiovascular health declining during the transition, but it is also consistent with age alone, with cohort differences between women surveyed at different points from 2007 to 2020, and with the fact that women who reach menopause earlier differ systematically from those who reach it later. The study can establish that perimenopausal women in this sample had worse cardiovascular health than premenopausal women. It cannot establish that any individual woman's score falls as she enters perimenopause.

The authors list further limitations: menopause stage was assigned from self-reported menstrual history, which introduces misclassification; data on hormone therapy use and surgical removal of the ovaries was incomplete or unavailable; and several Life's Essential 8 components — physical activity, smoking status — rest on self-report [s1].

The framing offered in the accompanying release is that midlife women should treat perimenopause as a "window of opportunity" for reassessing risk [s1]. That is a reasonable inference from the pattern rather than a demonstrated causal claim, and it is worth reading as such.

Why the finding is plausible anyway

Cross-sectional data is weak evidence for a within-person trajectory, but it is not the only evidence on this question, and the mechanism is not mysterious.

A 2025 review in the Methodist DeBakey Cardiovascular Journal summarizes the physiology: estrogen promotes vasodilation and maintains a favorable lipid profile by raising high-density lipoprotein and lowering low-density lipoprotein, so its decline during the menopause transition produces dyslipidemia [s2]. The same review notes that falling estrogen shifts fat storage from peripheral to central depots, increasing visceral fat, which in turn drives release of inflammatory cytokines that impair insulin signaling and raise type 2 diabetes risk [s2].

That is a mechanistic account of exactly the two components the 2026 analysis found most degraded — cholesterol and blood sugar. Concordance between an observed pattern and an independent physiological explanation is not proof, but it raises the prior considerably.

The review also flags a distributional point the national analysis does not address: non-Hispanic Black women have hypertension prevalence rates nearly twice as high as White women, and evidence suggests they go through menopause earlier and more abruptly [s2]. A single median score per menopause stage conceals that.

One further point from the review that has shifted in recent years: vasomotor symptoms — hot flashes — are now recognized as markers of increased cardiovascular risk rather than purely a quality-of-life issue [s2]. That reframes a symptom most women report as potentially informative rather than merely unpleasant.

What would settle it

A longitudinal cohort that measures the same women repeatedly across the transition, with menopause staging based on hormonal markers rather than recalled menstrual history, and with hormone therapy and oophorectomy status recorded. Such cohorts exist; what the 2026 analysis adds is national scope and a standardized composite score, not within-person follow-up.

Until then the accurate statement is narrow: in a nationally representative US sample, perimenopausal women had substantially worse composite cardiovascular health than premenopausal women, concentrated in cholesterol and blood glucose, and the physiology of the transition offers a coherent reason why that might be causal.

This article describes observational research and is informational only. It is not medical advice and does not recommend any test, screening interval, or treatment.

Sources

• [s1] Perimenopause may offer a "window of opportunity" for heart disease prevention in women, American Heart Association Newsroom, 2026-05-13, reporting a study published the same day in the Journal of the American Heart Association.
• [s2] Yousefzai, Amin, Faizan et al., Cardiovascular Health During Menopause Transition: The Role of Traditional and Nontraditional Risk Factors, Methodist DeBakey Cardiovascular Journal, 2025-08-12.