70

Hormone Replacement Therapy Associated With Lower Dementia Risk in Postmenopausal Women

1-minute summary

A large prospective cohort analysis of 183,450 postmenopausal women from the UK Biobank found that use of hormone replacement therapy (HRT) for at least 1 year was associated with a 10% lower relative risk of all-cause dementia and a 16% lower relative risk of Alzheimer disease over a mean follow-up of 13.3 years. (Alzheimer’s Association)

The association was stronger in several subgroups, particularly women who had undergone surgical menopause, carried an APOE ε4 allele, or had relatively lower lifetime exposure to naturally produced estrogen. The strongest associations were observed among women who initiated HRT between approximately 46 and 56 years of age. (Alzheimer’s Association)

Importantly, this was an observational cohort study, not a randomized clinical trial. Therefore, the results show an association between HRT use and dementia incidence but cannot establish that HRT prevents dementia or causes the observed reduction in risk.

---

Why this matters

Dementia disproportionately affects women, and Alzheimer disease is the most common form of dementia. The researchers investigated whether HRT use was associated with dementia risk and whether characteristics such as type of menopause, lifetime estrogen exposure, APOE genotype, and age at HRT initiation modified that association. (Alzheimer’s Association)

The study is particularly relevant because it moves beyond asking whether HRT is associated with dementia risk overall and examines whether some groups of women may have different associations with HRT use.

However, these findings should not be interpreted as evidence that HRT should be prescribed specifically to prevent dementia. The authors themselves note that further research is required to establish causality.

---

Study at a glance

The published study reports that the analysis adjusted for multiple potential confounding factors, including age, education, smoking, blood pressure, ethnicity, BMI, cholesterol, socioeconomic deprivation, diabetes and use of cholesterol-lowering or antihypertensive medications. (MedRxiv)

---

What the researchers found

Among the 183,450 women, 3,948 developed dementia during follow-up. Of these cases, 1,993 were Alzheimer disease, while 1,955 were classified as non-Alzheimer or unspecified dementia. (Alzheimer’s Association)

Overall dementia risk

HRT use for at least 1 year was associated with:

The all-cause dementia association was statistically significant, with an HR of 0.90 (95% CI, 0.84-0.96). (Alzheimer’s Association)

Important interpretation

A hazard ratio of 0.90 does not mean that HRT prevents exactly 10% of dementia cases. It indicates that, after statistical adjustment for measured factors, HRT users had approximately 10% lower relative hazard of developing dementia than the comparison group.

---

Subgroups that appeared to have stronger associations

The researchers examined whether the association differed according to biological and reproductive characteristics.

The association was particularly pronounced among women who had experienced surgical menopause. (Alzheimer’s Association)

The study therefore raises the possibility that the relationship between HRT and dementia may differ according to lifetime estrogen exposure and genetic background.

---

Age at HRT initiation

Age at the start of HRT also appeared to modify the association.

The supplied report highlighted the following age groups:

The researchers reported that initiation between approximately 46 and 56 years was associated with the strongest reductions in dementia risk. (PubMed)

This finding is consistent with the concept that the timing of exposure to menopausal hormone therapy may be relevant to its relationship with later-life neurological outcomes. However, because this was not a randomized trial, the age-related differences cannot establish that starting HRT at a particular age causes a reduction in dementia risk.

---

What did the study actually measure?

An important point for clinicians is that the researchers evaluated HRT use as an exposure, rather than assigning women to HRT.

In other words:

HRT users → observed over time → dementia diagnoses compared with nonusers

rather than:

Women randomized to HRT vs placebo → followed prospectively

Consequently, women who used HRT could differ from nonusers in numerous ways that are difficult to completely measure or eliminate statistically.

---

Clinical significance

The study adds to a growing body of observational evidence suggesting that the relationship between menopausal hormone therapy and brain health may be more nuanced than a simple overall benefit-or-harm model.

The investigators suggested that factors such as:

• type of menopause;
• APOE ε4 genotype;
• lifetime endogenous estrogen exposure; and
• age at HRT initiation

may influence the association between HRT and subsequent dementia risk. (Alzheimer’s Association)

Anne-Marie Minihane, PhD, one of the study authors, described the broader implication as follows:

“Our findings contribute to growing evidence that hormone therapy’s effects on brain health are complex and influenced by individual biological factors.”

She also noted that HRT has traditionally been prescribed primarily for menopausal symptoms, while the findings raise the possibility of a relationship with longer-term cognitive health in some women. (University of Exeter News)

But an important clinical distinction

Association ≠ prevention.

The study does not establish that prescribing HRT to a postmenopausal woman who is otherwise not taking it will prevent dementia.

The authors’ findings should therefore be considered hypothesis-generating and potentially relevant to individualized risk assessment, rather than evidence for HRT as a dementia-prevention therapy.

---

Limitations

1. Observational design

The principal limitation is that this was not an RCT. Although the researchers adjusted for numerous confounders, residual confounding remains possible. (Alzheimer’s Association)

2. HRT formulation was not fully characterized

The researchers acknowledged limitations related to the absence of detailed analysis of:

• HRT formulation;
• dose;
• route of administration; and
• potentially different treatment regimens.

Therefore, the results cannot establish whether one particular HRT formulation, dose or route is responsible for the observed association.

3. Healthy-user and socioeconomic effects

HRT users and nonusers may differ in health behaviors, healthcare utilization, socioeconomic circumstances and other characteristics that are difficult to completely capture in statistical models.

4. UK Biobank population

UK Biobank participants are not necessarily representative of every postmenopausal population. Therefore, the findings may not apply equally to women of all ethnicities, socioeconomic groups or healthcare systems.

5. Dementia diagnosis as an outcome

The study examined incident dementia diagnoses rather than performing standardized prospective cognitive testing in every participant.

6. No causal conclusion

The study cannot determine whether HRT itself produced the lower dementia incidence.

---

What doctors should know

1. HRT should not currently be considered a proven dementia-prevention treatment

The observed HR of 0.90 for all-cause dementia is clinically interesting, but it comes from an observational analysis.

2. Timing may matter

The association appeared strongest among women initiating HRT in the 46–56-year range, but this should not be interpreted as a recommendation to initiate HRT solely for dementia prevention.

3. Surgical menopause deserves attention

Women with surgical menopause demonstrated a particularly strong association, with HR 0.74 (95% CI, 0.65–0.84).

4. APOE ε4 may modify the association

APOE ε4 carriers showed an HR of 0.87 (95% CI, 0.80–0.95) among HRT users, suggesting that genetic background may influence the observed relationship. (Alzheimer’s Association)

5. Individualized HRT decisions remain important

The study supports further investigation of whether menopausal history, endogenous estrogen exposure and genetic factors could eventually contribute to more individualized approaches to menopausal hormone therapy.

For current clinical decisions, however, the established indications, contraindications, risks, benefits and patient preferences for HRT remain central.

---

Key numbers at a glance

---

Bottom line

In this large UK Biobank cohort of 183,450 postmenopausal women, HRT use for at least 1 year was associated with a lower incidence of dementia, including Alzheimer disease, during an average of 13.3 years of follow-up. The association was stronger among women with surgical menopause, APOE ε4 carriage, lower lifetime endogenous estrogen exposure, and HRT initiation between approximately 46 and 56 years. (Alzheimer’s Association)

However, this was an observational study, so it does not establish that HRT prevents dementia. The findings also do not establish that a particular HRT preparation, dose, route or initiation age should be used specifically for dementia prevention. Further randomized and mechanistic research is needed.

Clinical takeaway: HRT was associated with lower dementia risk in this cohort, but the evidence is not sufficient to prescribe HRT solely for dementia prevention.

---

Original research / DOI

Squires S, Saleh RNM, Pilling LC, et al. Hormone replacement therapy and dementia risk among postmenopausal women: identifying responsive subgroups in the UK Biobank. Alzheimer’s & Dementia. 2026;22(8):e71679. Published August 26, 2026. (DOI)

DOI: 10.1002/alz.71679 (DOI)