In the largest randomized trial ever done (WHI), hormone therapy over its full 18-year follow-up was not linked to a change in all-cause mortality, for better or worse.

There is a more encouraging pattern in women who started younger, and several meta-analyses that pool smaller trials of younger women do show lower mortality on hormone therapy. But the WHI's own younger-age subgroup, which is the best individual randomized evidence, did not reach a clearly significant mortality reduction over the long term. The fair statement is: no proof of mortality harm, and a real but not fully proven signal of possible benefit in younger initiators.

What to know about this answer

The disagreement is about how much weight to give pooled meta-analyses of many smaller trials versus the single largest randomized trial's own subgroup, which did not reach statistical significance over the long term.

What the research shows, and which source each finding rests on

WHI 18-year cumulative follow-up: pooled all-cause/CV/cancer mortality HR 0.99 (0.94-1.03); CEE-alone HR 0.94; CEE+MPA HR 1.02. The ratio of mortality hazard ratios for ages 50-59 vs 70-79 was 0.61 during the active intervention phase, attenuating to 0.87 (CI touching 1.00) over 18 years.

B[1]

Meta-analyses restricted to younger populations (mean age under 60) show lower mortality: OR 0.61 (0.39-0.95) in one meta-analysis, and RR 0.73 (0.52-0.96) in a Bayesian synthesis of 19 RCTs (mean age 55).

A[2]A[3]

Cochrane 2015: women who started hormone therapy less than 10 years after menopause had lower mortality (RR 0.70, 0.52-0.95) and lower coronary heart disease (RR 0.52, 0.29-0.96), still with increased VTE (RR 1.74).

A[4]

In WHI's estrogen-alone trial, outcomes were consistently more favorable in the youngest age group: 19 fewer adverse global-index events per 10,000 women per year for ages 50-59 versus 51 excess events at ages 70-79.

B[5]

Where the real disagreement is

Each position is shown with its own sources. A position listed here is not necessarily the accepted view.

Yes, mortality benefit in younger initiators

meta-analysts including Dr. Salpeter, and field clinicians such as Dr. Hodis and Dr. Bluming

Multiple independent meta-analyses restricted to younger or recently menopausal women consistently find lower mortality, with one Bayesian analysis describing the probability that hormone therapy reduces total mortality in younger women as close to certain.

A[2]A[3]A[4]

Not established; no mortality harm, but benefit unproven

WHI investigators

In the WHI itself, the only very large randomized cohort, 18-year all-cause mortality was neutral overall, and the younger-women signal seen during the active treatment years attenuated toward the null over 18 years, with the confidence interval touching 1.00. Meta-analyses aggregate many smaller trials not originally designed to measure mortality, with low event counts.

B[1]

What would settle it: An individual-participant-data meta-analysis restricted to women who started within 10 years of menopause with a mortality analysis planned in advance, or the same large hard-endpoint trial described in the heart-disease question.

What is really going on here

  1. Is that finding a relative risk reduction, or an absolute number - how many fewer deaths per 1,000 women per year does it actually mean?
  2. Does that finding come from the WHI itself, or from a meta-analysis pooling smaller trials?

Questions for your doctor

  1. Is there RCT evidence specific to someone my age and years-since-menopause on mortality, or mainly pooled data from smaller trials?
  2. What does the honest uncertainty here mean for how long I might stay on therapy?
  3. What other proven longevity factors (exercise, blood pressure, not smoking) should I prioritize alongside this decision?

Questions with no answer yet

Does hormone therapy started in the 50s actually reduce a woman's lifetime risk of dying, or does the large trial's own younger-age subgroup simply not have enough deaths to detect a real but modest effect?

Partly studied

What is known: Meta-analyses pooling many smaller trials in younger women consistently find lower mortality, but the single largest randomized trial's own younger subgroup showed a favorable pattern during active treatment that faded toward the null over 18 years of follow-up, with the confidence interval touching the point of no effect. Low event counts in that trial's youngest stratum leave open whether it was simply underpowered to detect what the pooled smaller trials found.

What would test it: An individual-participant-data meta-analysis restricted to women who started within 10 years of menopause with a mortality analysis planned in advance, or the same large hard-endpoint trial described in the heart-disease question.

Is a mortality benefit in younger initiators a real drug effect, or does it partly reflect that healthier women were the ones enrolled in, or who stayed in, the trials that found it?

Not directly studied yet

No research does not mean it does not work, and it does not mean it works. It means nobody has tested it yet.

What is known: This question has not been specifically separated from healthy-user and survivorship effects within the mortality literature, even though these biases are well documented elsewhere in the hormone therapy evidence base.

What would test it: Target-trial-emulation methods designed to detect and adjust for healthy-user and survivorship bias, applied specifically to the mortality outcome in younger initiators.

Sources

  1. 1.Menopausal Hormone Therapy and Long-term All-Cause and Cause-Specific Mortality: The Women's Health Initiative Randomized Trials.Manson JE, et al. Menopausal hormone therapy and long-term all-cause and cause-specific mortality: the WHI randomized trials. JAMA. 2017;318(10):927-938. · Randomised trial
  2. 2.Mortality associated with hormone replacement therapy in younger and older women: a meta-analysis.Salpeter SR, et al. Mortality associated with hormone replacement therapy in younger and older women: a meta-analysis. J Gen Intern Med. 2004;19(7):791-804. · Systematic review or guideline
  3. 3.Bayesian meta-analysis of hormone therapy and mortality in younger postmenopausal women.Salpeter SR, et al. Bayesian meta-analysis of hormone therapy and mortality in younger postmenopausal women. Am J Med. 2009;122(11):1016-1022. · Systematic review or guideline
  4. 4.Hormone therapy for preventing cardiovascular disease in post-menopausal women.Boardman HM, et al. Hormone therapy for preventing cardiovascular disease in post-menopausal women. Cochrane Database Syst Rev. 2015;(3):CD002229. · Systematic review or guideline
  5. 5.Health outcomes after stopping conjugated equine estrogens among postmenopausal women with prior hysterectomy: a randomized controlled trial.LaCroix AZ, et al. Health outcomes after stopping conjugated equine estrogens among postmenopausal women with prior hysterectomy. JAMA. 2011;305(13):1305-14. · Randomised trial

The letter grades the type of study behind a claim, not the answer itself.

This page is general information based on published research and public sources. It is not medical advice and not a personal assessment, and it does not replace a conversation with your doctor. Peritale is a wellness product, not a medical device, and is not intended to diagnose, treat, cure or prevent any disease. For any medical question, speak with a licensed clinician.

You are not imagining it

In about seven minutes on your phone, Peritale reads five signal systems and maps your symptoms, so you can see that what you feel is real, follow your hormonal transition over time, and walk into your doctor ready to ask for the care you deserve. Your first check is free, no credit card.

Start Free Mapping

Related questions