Is Menopausal Hormone Therapy Safe for All Ages?

Key Highlights

• Hormone therapy (HT) relieved moderate–severe vasomotor symptoms (VMS) and did not significantly affect atherosclerotic cardiovascular disease (ASCVD) risk in women aged 50–59 years with VMS.
• Cardiovascular risk rose with age among women with VMS: ASCVD hazards were increased at ≥70 years with both conjugated equine estrogens (CEE) alone and CEE plus medroxyprogesterone acetate (MPA).
• CEE alone reduced VMS by ~41% across age groups, while VMS reduction with CEE+MPA attenuated with age and was not evident at 70–79 years.
• Findings support HT for symptomatic women aged 50–59 years, call for caution at 60–69 years, and advise avoidance at ≥ 70 years.

In a secondary analysis of two randomized Women’s Health Initiative (WHI) trials, researchers found that hormone therapy (HT) effectively reduced moderate or severe vasomotor symptoms (VMS) and, in younger symptomatic women (50–59 years), did not significantly alter atherosclerotic cardiovascular disease (ASCVD) risk. In contrast, among symptomatic women aged 70–79 years, HT was associated with increased ASCVD risk, with excess events per 10,000 person-years observed for both conjugated equine estrogens (CEE) alone and CEE plus medroxyprogesterone acetate (MPA).

HT remains the most effective therapy for menopausal VMS, yet concerns about adverse cardiovascular outcomes have limited its use. Prior subgroup analyses suggested age-dependent cardiovascular effects, but estimates were imprecise. This analysis was undertaken to clarify ASCVD risk and VMS relief with CEE and CEE+MPA across decades of age among women with VMS.

Investigators analyzed the WHI HT trials conducted at 40 US centers (1993–2012): postmenopausal women aged 50–79 years with hysterectomy were randomized to CEE 0.625 mg/d vs placebo; those with an intact uterus were randomized to CEE 0.625 mg/d plus MPA 2.5 mg/d vs placebo. The primary outcome was ASCVD, a composite of nonfatal myocardial infarction, hospitalization for angina, coronary revascularization, ischemic stroke, peripheral arterial disease, carotid artery disease, or cardiovascular death. Analyses followed intent-to-treat using Cox models with prespecified age strata and adjustment for key covariates. VMS were assessed at baseline and year 1 and categorized as none/mild vs moderate/severe.

Among 27,347 women (mean [SD] age, 63.4 [7.2] years), median follow-up was 7.2 years in the CEE-alone trial and 5.6 years in the CEE+MPA trial. The researchers also studied the effects of therapy on VMS and ASCVD risk by age group (Table 1.)

Table 1. Effects of hormone therapy on vasomotor symptoms and ASCVD risk by age group

Abbreviations: ASCVD, atherosclerotic cardiovascular disease; CEE, conjugated equine estrogens; MPA, medroxyprogesterone acetate; RR, relative risk; HR, hazard ratio.

Limitations include the secondary, exploratory nature of subgroup analyses with some sparse strata and wide confidence intervals; use of composite outcomes not designated as primary trial endpoints; reliance on oral CEE and MPA at fixed doses only; VMS status measured at baseline and year 1; imputation for time-since-menopause in a subset; and no adjustment for multiple testing despite numerous interaction tests.

“Our findings are supportive of current clinical guidelines, which generally recommend initiation of HT for VMS in women aged 50 to 59 years (or within 10 years of menopause),” the study authors wrote. “Some guidelines also recognize continuation of HT use in healthy older women via individualized shared decision-making, while others advise avoidance of HT in older women. Newer US Food and Drug Administration–approved nonhormonal therapies for VMS, such as fezolinetant, are now on the market and are alternatives to HT for this indication, but fezolinetant has not been tested in women older than 65 years.”

Reference:
Rossouw JE, Aragaki AK, Manson JE, et al. Menopausal hormone therapy and cardiovascular diseases in women with vasomotor symptoms: a secondary analysis of the women's health initiative randomized clinical trials. JAMA Intern Med. Published online September 15, 2025. doi:10.1001/jamainternmed.2025.4510