Hormone Therapy Dementia Risk: What 1 Million Women Show
Women come to me afraid of two things when hormone therapy comes up: cancer and dementia. The cancer conversation has shifted considerably in the last decade, and I’ve addressed it at length in other posts. The dementia conversation has been murkier, louder, and less honest. Women have been warned that estrogen could accelerate mental decline, leave them vulnerable to Alzheimer’s, and push them into dementia earlier than they would otherwise arrive. Those warnings have kept a lot of symptomatic women away from treatment that could have helped them.
Here is what I’ve never been satisfied with in that conversation: most of those dementia warnings trace back to the same flawed data we’ve been relitigating for twenty years. The Women’s Health Initiative Memory Study enrolled women aged 65 and older and gave them oral synthetic combined hormone therapy. It found elevated dementia risk. Medicine took that finding and applied it broadly, without asking whether the age of initiation mattered, whether hormone type mattered, or whether delivery route mattered. Nobody connected those dots.
In December 2025, The Lancet Healthy Longevity published the largest review of hormone therapy dementia risk ever conducted: a systematic review and meta-analysis covering ten studies and 1,016,055 women. I’ve read it carefully. The headline is more complicated than either side wants to admit, and that matters enormously for every woman deciding whether to pursue HRT.
What the Lancet Review Found About Hormone Therapy Dementia Risk
Researchers from University College London, the University of Melbourne, Trinity College Dublin, and the University of Exeter analyzed every high-quality study published between January 2000 and October 2025 examining the relationship between hormone therapy and incident mild cognitive impairment or dementia.
Here is what the data showed:
The only randomized controlled trial in the analysis was the Women’s Health Initiative Memory Study. This trial enrolled women aged 65 and older who started combined hormone therapy at that age. The result: approximately 7 additional dementia cases per 1,000 women compared to placebo. That is a real signal. But it reflects women who began hormone therapy at 65 or older, which is not how most clinicians prescribe it and not when most symptomatic women want it.
The nine observational studies told a different story. When researchers pooled five studies covering 672,195 women on estrogen-alone therapy, the pooled risk ratio was 1.10, translating to roughly 9 additional dementia cases per 1,000 women. For combined therapy across four studies covering 693,412 women, the pooled risk ratio was 1.12, roughly 10 additional cases per 1,000 women. Both results carried confidence intervals wide enough to include both potential benefit and potential harm. In statistical terms: the evidence does not confirm a meaningful effect in either direction.
One signal was notably different. Long-term estrogen-alone use for more than 10 years produced a pooled risk ratio of 0.93, hinting at a possible small reduction in dementia risk. The authors rated that finding as very low certainty. But it points in the direction many clinicians have suspected for years.
Why the Evidence Still Cannot Answer the Question Women Are Actually Asking
The Lancet authors did something that takes scientific courage: they graded the certainty of most of their results as “very low,” with only one analysis rated “moderate certainty.” That is not a failure of the review. It is an honest assessment of the underlying data.
Here is the core problem. The observational studies that make up most of this evidence cannot fully account for who chooses hormone therapy and who does not. Women who seek out HRT may differ from women who avoid it in ways that independently affect dementia risk: education level, health engagement, cardiovascular status, smoking history. Researchers can adjust for many of these factors, but not all of them.
On top of that, most studies used prescription records to define HRT exposure without confirming whether women actually took their medications. Dementia diagnoses often came from administrative health records rather than specialist evaluations, which means misclassification is likely in both directions.
The Menopause Society and other clinical bodies have long noted that the evidence quality on HRT and cognition has lagged behind clinical need. This Lancet Healthy Longevity meta-analysis confirms that gap clearly.
What the data cannot tell us is what happens when a woman in her late 40s starts hormone therapy at menopause onset and stays on it for 15 years. That study does not exist. The one randomized trial enrolled 65-year-olds. The observational studies mix women who started at various ages without cleanly separating age of initiation from duration of use.
What the Hormone Therapy Dementia Risk Data Mean for Decisions Today
I want to be direct here, because this is where many clinicians get evasive. The hormone therapy dementia risk conversation has two honest conclusions, and women deserve both of them.
First: the evidence does not support using dementia fear as a reason to withhold HRT from symptomatic women in their late 40s and 50s. The one trial showing real harm enrolled women who started at 65 and older. The observational data are too uncertain to carry the weight of a clinical recommendation against treatment in younger symptomatic women. Telling a 49-year-old woman in perimenopause that she risks dementia if she takes estrogen is not supported by what this review found.
Second: the evidence does not support promising HRT will prevent dementia. The long-term estrogen signal is intriguing. The mechanistic case for estrogen’s role in brain health is real. But the clinical trial evidence at this point does not license making that promise to patients.
That is an uncomfortable middle. It is also the honest one. Too many women make decisions based on either unfounded fear or unfounded promise. Neither serves them.
At Living Well Dallas, I look at each woman’s full picture: symptom burden, cardiovascular status, family history, metabolic health, and the complete hormone profile. At Menrva Health, we apply the same approach through telehealth for women in all 50 states. The dementia question is part of that conversation, not a door that closes it.
Key Takeaways
- The largest-ever analysis of hormone therapy dementia risk covered 1,016,055 women in 10 studies and found no consistent, statistically confirmed effect in either direction.
- The only randomized trial (WHIMS) found 7 additional dementia cases per 1,000 women, but it enrolled women who started combined HRT at age 65 or older.
- Observational pooled data showed risk ratios of 1.10 and 1.12 for estrogen-alone and combined therapy, but confidence intervals were too wide to confirm meaningful harm.
- Long-term estrogen-alone use (more than 10 years) produced a pooled risk ratio of 0.93, a possible small risk reduction, rated very low certainty.
- Most evidence was rated “very low certainty” by the Lancet authors, meaning the existing data cannot reliably answer the dementia question for women who start HRT in their 40s or 50s.
Frequently Asked Questions
Does hormone therapy increase dementia risk in women? The current evidence does not confirm a clear increase or decrease for most women. The one randomized trial showing increased risk enrolled women who started at 65 or older, which is not typical clinical practice. Observational studies in larger populations showed risk ratios of 1.10 to 1.12, but confidence intervals were too wide to confirm meaningful harm. For women who start HRT at menopause onset in their late 40s or early 50s, no high-quality randomized data show an increased risk.
What did the Women’s Health Initiative Memory Study find? The WHIMS enrolled women aged 65 and older and gave them combined estrogen-progestin therapy. Women on combined therapy had approximately 7 additional dementia cases per 1,000 compared to placebo. Estrogen-alone therapy increased mild cognitive impairment risk in this older cohort. These findings apply to women who start HRT well past menopause, not to women who begin treatment in their late 40s or early 50s at perimenopause onset.
HRT Timing and Brain Health
Is there a critical window for hormone therapy to protect the brain? The critical window hypothesis suggests that starting HRT close to menopause may offer brain protection that starting later does not. This Lancet review found no statistical confirmation of that timing effect, but the authors noted that the evidence base is too limited to rule it out. The biology behind the theory remains compelling: estrogen receptors in the brain are most active when estrogen levels first begin to decline.
If HRT does not clearly prevent dementia, should I still consider it? Yes, if you have significant menopausal symptoms affecting your quality of life. Hormone therapy treats hot flashes, sleep disruption, vaginal atrophy, and other symptoms with solid evidence behind it. Dementia prevention is not an established indication at this point. That uncertainty should not be the reason physicians deny treatment to symptomatic women in their 40s and 50s.
What Research Still Needs to Happen
Why is the evidence on hormone therapy dementia risk still uncertain after decades of study? Most studies used prescription records rather than confirmed medication use, and dementia diagnoses came from administrative records rather than specialist evaluations. Women who choose HRT also differ from those who do not in ways that are difficult to control for statistically. The one randomized trial enrolled 65-year-olds who are not representative of most HRT candidates. A decades-long trial in younger women simply has not happened.
What would better research on this question look like? It would enroll women in their late 40s at menopause onset, assign them to specific formulations and delivery routes rather than oral synthetic hormones only, follow them for 15 to 20 years, and use clinical dementia diagnosis rather than insurance codes. That study does not exist. Until it does, clinicians and patients should make decisions based on symptoms, individual risk profiles, and the evidence that actually exists rather than waiting for certainty that may be a decade away.
Dr. Betty’s Bottom Line
I’ve spent years watching women get scared away from hormone therapy by dementia arguments that were never as solid as they sounded. The original WHIMS data applied to 65-year-olds on oral synthetic hormones, and medicine used those results to frighten 50-year-olds away from transdermal bioidentical estrogen. That is not how science is supposed to work.
This Lancet review is the most honest document on this question I’ve seen: the evidence does not confirm that HRT raises dementia risk in women who start at menopause, and it does not confirm that HRT prevents dementia either. What it confirms is that women deserve a much more precise conversation than the one they have been getting. A woman who is sleeping three hours a night because of hot flashes, whose memory is slipping because her brain never reaches restorative sleep, whose mental clarity is declining from cumulative estrogen loss: she deserves better than a physician waving a speculative dementia risk flag based on 65-year-old trial participants.
Come in. Let us look at your whole picture. The brain health conversation is part of it, not a reason to end it.
In-person care at Living Well Dallas is available for patients in the Dallas area. Menrva Health offers comprehensive menopause evaluation and hormone therapy through telehealth in all 50 states.
Source: Melville M, He L, Desai R, Nyamayaro P, Fox C, Kothari KU, Condron P, Miao M, Hickey M, Spector A. Menopause hormone therapy and risk of mild cognitive impairment or dementia: a systematic review and meta-analysis. Lancet Healthy Longev. 2025 Dec 22. doi: 10.1016/j.lanhl.2025.100803. PMID: 41448220.