The effect of hormone replacement therapy on the survival of UK women: a retrospective cohort study, 1984 - 2017
A major new IFoA-commissioned study by the University of East Anglia (UEA) concludes that for healthy women taking HRT, oestrogen-only therapy does not impact the risk of death from all causes, while taking combined HRT is associated with an average 9% reduction in death from all causes.
The study is the first of its kind to look at the impact of HRT on overall life expectancy using UK primary care data. It followed 105,199 healthy women aged 46 to 65 years at first HRT prescription over up to 32 years with an average follow-up of 13 years, and compared their outcomes with 224,643 non-users of the same age and GP practice.
This study supports the emerging consensus that, for most women, the benefits of HRT outweighs the harm. We hope this research will help to inform the debate as the private member’s bill is considered in Parliament and also, support women deciding whether to start or continue with HRT.
“Delving into these complex, long-term, multi-faceted issues is where the IFoA’s Actuarial Research Centre can really add value. It provides a valuable lens for exploring some of the most pressing social and policy challenges of the 21st century through credible, evidence-based insight and analysis in the public interest.
Louise Pryor, IFoA President
Whereas most previous studies adjusted for demographic and/or lifestyle factors only, this research also adjusted for type 2 diabetes, hypertension and its treatments, coronary heart disease and oophorectomy/hysterectomy status, as well as body mass index, smoking and deprivation status.
The research provides reassurance to women concerned about recent reports linking HRT with an increased risk in breast cancer. In 2019, a major analysis reported that using combined HRT for 10 years increased the risk of breast cancer substantially. The Medicines and Healthcare Products Regulatory Agency issued an alert (2019) that the risk of breast cancer was increased during use of all types of HRT and that an excess risk of breast cancer persisted for longer after stopping HRT than previously thought.
Although the UEA study observed a significant increased incidence of breast cancer, this did not translate into increased all-cause mortality in HRT users. This may be due to the higher likelihood of death from cardiovascular disease, osteoporosis and dementia, meaning that benefits in these conditions can easily offset risks from rarer conditions.
The researchers used a Weibull-Double-Cox regression model to estimate the risk of all-cause mortality at different ages. Taking combined HRT was associated with a 9% reduction overall in all-cause mortality compared to non-users. For those starting treatment at ages 51 to 55 this reduction was 13%, for ages 56 to 60 12% and for ages 61 to 65 8%. There was no significant reduction in all-cause mortality for combined HRT users who started the treatment at age 46 to 50. Importantly, there was no significant effect on all-cause mortality for women taking oestrogen-only HRT in any age group.
It’s exciting that this new research found that combined hormone replacement therapy (HRT) use was linked to an overall lower risk of death, and that oestrogen-only HRT was not linked to an increased risk of death. HRT use has been controversial for many years, as it offers symptomatic relief to many women but there have been conflicting reports about the long-term risk of breast cancer, as well as possible benefits to cardiovascular health. UK primary care data has now enabled long term follow-up of thousands of women in the UK, comparing the overall risk of death over many years for those using HRT with those not using it.
Nick Steel, Clinical Professor in Public Health, Norwich Medical School
The main limitation of using observational data is that there may be some differences between women using HRT and non-users that were not accounted for. Randomised trials are better at accounting for all differences between groups, and interestingly the data from aggregated trials found a reduced risk of breast cancer in HRT users. The main strengths of this study are the long period of follow-up over up to 32 years with all-cause mortality as the key outcome, and the use of a large primary healthcare database that enabled HRT users to be matched with controls with information on co-existing illnesses and sociodemographic factors.
This new information on the long-term risks and benefits of HRT can be used to inform women deciding whether to start or continue with HRT. This research strengthens the emerging consensus that for most women the benefits of long-term HRT outweigh the harms, but every woman’s situation is different and the risks and benefits from different conditions should be considered individually.
This research was part of the Actuarial Research Centre’s Big Health and Actuarial Data programme. Further information on the programme can be found on the IFoA programme page.
The peer-reviewed preprint is available below.