August 26, 2026

expert reaction to HRT and dementia risk among postmenopausal women

A study published in Alzheimer’s and Dementia looks at HRT and the risk of dementia in postmenopausal women.

Prof Aimee Spector, Professor of Clinical Psychology of Ageing, Director of the Menopause Mind Lab, University College London (UCL), said:

“The results are promising in that they analysed data from the UK Biobank using large samples. The novelty of their findings was timing of HRT initiation, with those who started HRT between the ages of 46-56 conferring the greatest benefit. This supports the ‘window of opportunity’ hypothesis, suggesting that HRT use during the menopause transition phase, where symptoms are typically more problematic; is important. (In contrast, many other studies have looked at HRT use in older women, who may have fewer troublesome symptoms and may be less likely to be using it in practice).

“The important caveat, like all observational studies, is that association does not imply causation. It is well established that people who take HRT are generally less exposed to known dementia risk factors such as low education, poor physical health and poverty. On looking at the two groups in this study, I noticed that the dementia group (compared to the no dementia group) was less educated (11% fewer in higher education), potentially in poorer health (for example 6% more were diabetic), and 5 years older (average age 65 versus 60, highly relevant with age being the biggest risk factor for dementia). However, the groups appear fairly well matched overall, which is promising.

“The authors correctly highlight the need for Randomised Controlled Trials, the ‘gold standard’ methodology needed to categorically establish whether HRT can directly reduce dementia risk.”

Dr Sarah James, Senior Research Fellow at the MRC Unit for Lifelong Health and Ageing, University College London (UCL), said:

“This paper seeks to address an extremely important question around the relationship between HRT and dementia. It is also a very difficult question to answer. We need very large studies with long-term follow-up to understand whether what happens around menopause relates to brain health later in life.

“The UK Biobank is a valuable resource for this because its scale, long-term follow-up and linkage to healthcare records allow researchers to track outcomes decades later. Although participants tend to be generally healthier than the average UK population, this resource has already enabled important research into menopause, HRT and brain health. This study therefore adds to a growing body of evidence suggesting that the relationship between HRT and brain health is more nuanced than simply asking whether HRT is “good” or “bad”.

“This is a particularly carefully conducted, thoughtful and transparent study, led by researchers with substantial expertise in this area. The authors have made good use of the available data and have carefully considered a range of factors that could influence the findings. The results around surgical menopause and earlier HRT initiation are particularly interesting and provide compelling evidence that timing and individual circumstances matter for the observed relationship with dementia risk.

“As the authors themselves recognise, this is an observational study, so it shows an association rather than proving that HRT prevents dementia. Some closely related factors, including surgical menopause, age at menopause, and likelihood of HRT use, cannot be fully disentangled here, and we need to understand more about the biological mechanisms that may drive these associations.

“The next step is to build on this work: we need to see these findings replicated in more representative populations and with better information about HRT formulation, dose, route, duration and timing. We also need to understand other ways of supporting brain health for women who cannot or choose not to take HRT.

“Overall, I see this as a valuable contribution to an important and evolving field. It strengthens the case for a more personalised understanding of HRT and brain health, while also showing why we need better evidence to understand who might benefit, when and under what circumstances. That is important for giving women accurate information to make informed decisions.”

Prof Ian Maidment, Professor of Clinical Pharmacy, Aston University, said:

“This observational cohort study found that HRT was associated with a reduced risk of dementia. Overall, HRT was linked to a 10% reduced risk of dementia.

“The picture was quite complicated with some differences between groups. For example, the statistical evidence indicated that for women who had experienced natural menopause, HRT was not linked to a reduced risk of dementia, whereas it was linked to reduced risk for women who underwent surgical menopause. Age was also important and HRT only had an effect in women who started HRT when aged between 46 and 56 years old.

“Limitations include lack of information on the type of HRT (e.g., oestrogen only or oestrogen combined with progesterone) and the dose. We also need more research in diverse communities; the data mainly included women from more affluent areas and of European descent.

“Overall, HRT appeared to have a potential modest effect with the available evidence indicating less impact than physical activity.

“This was an observational study. An observational study involves researchers following a group of people over a period of time. They do not change how the patient is treated, but study how something that happens in the real world, like a GP prescribing a medication, affects the patient’s health.

“One key limitation with observational studies is so-called “confounding variables”. These are hidden factors, which the researchers do not identify that explain the apparent link between the event (e.g., GP prescribing a medication) and the seen outcome. One example would be: children being taller is associated with a higher IQ. Here, the hidden variable is age – as children get older both their IQ and height increases.

“Researchers always try and identify them (the confounding variables) and control for them, as they did in this study. However, there is always the risk that some remain hidden and are not controlled for.”

From our friends at the All-Ireland SMC:

Prof Andrea Kwakowsky, Professor of Pharmacology and Therapeutics, School of Pharmacy and Medical Sciences, University of Galway, comments: 

“The literature on HRT and its link to dementia is controversial. The available evidence indicates that estrogens play a significant role in influencing dementia risk, with studies demonstrating both beneficial and detrimental effects of HRT.  

“The strength of this new study by Squires and colleagues is its large sample size, and the analysis of different groups of participants. The study provides further evidence that the protective effects of HRT vary depending on the type of menopause, the age and stage that women start treatment, and their genetic risk for developing dementia,  

“The attempt to capture the duration of HRT exposure is important but not fully addressed, as HRT is a binary variable in the study; people who are current users and those who have used HRT for at least 1 year in the past have been put in the same category. There is a wide range in the duration of HRT exposure and in the types of HRT used, which are not captured by this study design.  

“Based on the current literature, it is clear that more large-scale multinational studies considering as many confounding factors as possible are required to get more clarity on the relationship between HRT and dementia, as studies vary widely in their results from showing a positive correlation to a negative one to none at all.”

“The use of different forms of HRT and treatment regimens may partially explain the reported discrepancies, and capturing this information is critical. Given the many confounding factors involved, including genetic, socio-economic, and environmental influences, considering these is also essential to better understand the effects of HRT on dementia risk.”

Dr Yana Vinogradova, Principal Research Fellow Centre for Academic Primary Care in the School of Medicine, University of Nottingham, said:

“This observational study suggests a decreased risk of developing Alzheimer’s disease associated with HRT use. The study has, however, a number of limitations, which limit its general applicability. Its findings relate primarily to a small proportion of women with removed ovaries or wombs who had used oestrogen-only therapy. Biological evidence supports a protective effect of oestrogen-only MHT (aka HRT) on the brain but not for the combined oestrogen/progestogen treatments prescribed to most women. Early menopause is also associated with generally weakened health and the negative side effects of MHT/HRT can be more risky for such women. Bearing in mind these and other study limitations – in particular no detailed information on HRT use, this study should not be used as a possible basis for prophylactic prescribing of MHT/HRT to reduce risks of developing dementia.”

Prof Tara Spires-Jones, Professor of Neurodegeneration at the University of Edinburgh, Division Lead in the UK Dementia Research Institute, said:

“This is a well conducted study looking at whether HRT is associated with dementia risk in UK Biobank participants.  Scientists looked at data from over 180,000 women, 3,948 of whom developed dementia post-menopause.  They observed that HRT use was associated with approximately 10% lower rates of any type of dementia and a 16% lower rate of Alzheimer’s disease, the most common cause of dementia.  Interestingly, there was not a significant association in women who underwent natural menopause, but there was a 26% reduction in dementia risk with HRT use in women who underwent surgical menopause, a 16% risk reduction with HRT in women who were exposed to natural estrogen for less than 37 years (the time between staring their period and the menopause), and  a 16% decrease in risk with HRT in women who inherited the Alzheimer’s risk gene APOE4. The strongest association with decreased risk occurred when HRT was started between age 46 and 55.  Together, these data indicate that using HRT may reduce risk of dementia particularly in people who have shorter exposure to natural estrogen due to surgical menopause or a shorter time between starting their period and menopause. However, there are important limitations to this study.  This type of observational study cannot prove that the HRT was the cause of lower dementia risk.  For example, women with access to HRT can differ from non-users in their access to healthcare, educational attainment, ethnicity and lifestyle factors that affect dementia risk.

“Previous studies examining HRT and dementia risk including a randomized trial had conflicting results with some showing increased risk of dementia with HRT and others showing decreased risk.  More work is needed to fully understand the links between HRT and dementia risk.”

Dr Susan Kohlhaas, Executive Director of Research & Partnerships at Alzheimer’s Research UK said:  

“Women are more likely than men to develop dementia, but we don’t fully understand why. Hormonal changes around the menopause could be part of the picture, making the relationship between HRT and dementia risk an important question to understand.

“This is a large and carefully conducted study with a long follow-up period. The researchers also carried out a range of additional analyses to test how robust their findings were, including accounting for other factors that could influence dementia risk.

“Overall, women who had used HRT were around 10% less likely to develop dementia and 16% less likely to develop Alzheimer’s disease than women who had never used it. Importantly, this association was not the same for all women. HRT use was associated with a 26% lower risk of dementia in women who had experienced surgical menopause, and the association was also stronger in women with lower lifetime exposure to oestrogen. This suggests that a woman’s hormonal history may be important when trying to understand how HRT use relates to her future risk of dementia.

“As an observational study, this research cannot establish that HRT itself lowered dementia risk. Women who use HRT may differ from women who don’t in other ways that affect their risk of dementia, and although the researchers accounted for many of these factors, some differences may remain. The study also couldn’t reliably examine whether dementia risk differed according to the type, dose or route of HRT, all of which can vary considerably.

“Following women from menopause for long enough to measure dementia outcomes in a randomised trial is challenging, so high-quality long-term studies like this have an important role in building our understanding of whether HRT use in midlife is associated with dementia risk many years later.

“This study alone isn’t enough to tell us whether HRT should be used to reduce dementia risk, but it provides further evidence that the timing of HRT and a woman’s hormonal history may matter. We now need research that can establish whether HRT itself is responsible for these differences in dementia risk and understand which forms of HRT may have an effect.

“Anyone considering starting or stopping HRT should speak to their GP about the potential benefits and risks for them.”

‘Hormone replacement therapy and dementia risk among postmenopausal women: identifying responsive subgroups in the UK Biobank’ by Steven Squires et al. was published in Alzheimer’s and Dementia at 12:00 noon UK Time on Wednesday 26 August 2026.

Declared interests

Prof Aimee Spector: “No DOIs”

Dr Sarah James: “I don’t have any conflicts of interest”

Prof Ian Maidment: “No declarations of interest.”

Prof Andrea Kwakowsky: “I have links with the Alzheimer’s Society Ireland and New Zealand, Dementia Research Network Ireland, and Dementia Ireland, but I do not consider these activities to be COIs (advisory, grant assessment, PPI, event organising). “

Dr Yana Vinogradova: “No conflict of interest.”

Prof Tara Spires-Jones: “I have no conflicts with this study but have received payments for consulting, grant reviews, scientific talks, or collaborative research over the past 10 years from AbbVie, Sanofi, Merck, Scottish Brain Sciences, Jay Therapeutics, Cognition Therapeutics, Ono, Novo Nordisk, Eisai, Boehringer Ingelheim, and Bristol-Meyers Squibb and direct a company Spires-Jones Neuroscience, Ltd to act as a consultant. I am also Charity trustee for the British Neuroscience and serve as scientific advisor to several charities and non-profit institutions.”

Dr Susan Kohlhaas: “Dr Kohlhaas has no declarations of interest. Alzheimer’s Research UK is one of the funders of this study.”