Menopausal Hormone Therapy: Why Mortality Outcomes Are ‘Vital’

JoAnn E. Manson, MD, DrPH


September 12, 2017

Hello. I am Dr JoAnn Manson, professor of medicine at Harvard Medical School and Brigham and Women's Hospital in Boston, Massachusetts. I want to discuss a newly released report in JAMA[1] on the Women's Health Initiative (WHI) hormone therapy trials that looked at all-cause and cause-specific mortality during the intervention phases and cumulative follow-up for an average of 18 years.

The two parallel trials included more than 27,000 women. One trial looked at estrogen plus progestin and the other at estrogen alone. During the follow-up, more than 7000 deaths occurred. This is more than twice as many deaths as included in our earlier analyses.

Why is all-cause mortality an important endpoint to examine? With an intervention such as hormone therapy, with a complex pattern of benefits and risk, all-cause mortality provides a critically important summary measure, showing the net effect of the intervention on serious and life-threatening health outcomes. Hormone therapy has known benefits, including a reduction in symptoms and in hip fracture and other fractures. Hormone therapy has also been linked to risks such as venous thromboembolism, stroke, and, with estrogen plus progestin, increased risk for breast cancer.

What have we found in terms of mortality? In the overall study population, which comprised women ages 50-79 years (mean age, 63 years), we found a neutral effect of hormone therapy in both trials during both the intervention phase and the cumulative 18-year follow-up period for all-cause mortality and for mortality from cardiovascular disease and cancer.

We used a preplanned analysis to look at the results by age group, however, and in women who were 50-59 years of age, we saw a signal for reduced mortality with hormone therapy during the intervention phase of both trials. In the pooled analysis of the two trials, the hazard ratio was 0.69 with a P value of 0.01 for differences across age groups. This suggested significant differences depending upon age, with more favorable results in the younger women. With the cumulative 18-year follow-up, however, the test for trend by age groups was attenuated and no longer statistically significant.

There were a few surprises in our study. First, for dementia mortality—deaths from Alzheimer disease and other forms of dementia—we found a significantly reduced risk in the estrogen-alone trial, with a hazard ratio of 0.74 and a P value of .01. We consider this to be exploratory and need to delve more deeply into these findings. With estrogen plus progestin, we found a neutral effect on dementia mortality.

Also surprising, results were neutral for total cancer mortality in both trials separately and in the trials together. For breast cancer mortality, the results diverged between the two trials, with an increased risk with estrogen plus progestin and a decreased risk with estrogen alone. Overall, the results were neutral or counterbalancing for other cancers, and thus, the results for total cancer mortality ended up being neutral for both formulations over the full 18 years of follow-up.

What are the clinical implications of these findings? Many professional societies have guidelines and position statements that endorse the use of hormone therapy for treatment of moderate to severe menopausal symptoms. Certainly, our findings would support that practice, especially in the younger women who, during the treatment phase, tended to have a lower risk for mortality. That would suggest that these guidelines are quite appropriate in women in early menopause who do not have contraindications to treatment.

Our findings did not support the use of hormone therapy for the prevention of cardiovascular disease, other chronic diseases, or in an effort to reduce mortality, given that the results for all-cause mortality were neutral. Especially for women in their 60s and 70s, we found no signal for a reduction in all-cause mortality over the cumulative 18-year follow-up.

What are our next steps? We certainly need more research on the formulations of hormone therapy that are currently in common use. The WHI tested the most common formulations that were in use at the time the trial started. Currently, it is common to use lower doses and different formulations, such as estradiol, micronized progesterone, and transdermal routes of delivery. It would be very important to have additional research and understanding of the balance of benefits and risks of these formulations commonly used in current clinical practice.

Thank you so much for your attention. This is JoAnn Manson.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.