Menopausal Hormone Therapy and Long-term All-Cause and Cause-Specific Mortality

From Wiki Journal Club
Jump to: navigation, search
Manson JE, et al. "Menopausal Hormone Therapy and Long-term All-Cause and Cause-Specific Mortality: The Women's Health Initiative Randomized Trials". JAMA. 2017. 318(10):927-938.
PubMedFull text

Clinical Question

In postmenopausal women ages 50-79, what is the relationship between use of menopausal hormone therapy versus placebo in the Women's Health Initiative randomized trials and all-cause mortality over a median 18 years of follow-up?

Bottom Line

Among postmenopausal women who participated in 2 parallel randomized trials of estrogen plus progestin and estrogen alone, all-cause mortality rates for the overall cohort in the pooled trials were not significantly different for the hormone therapy groups versus the placebo groups.

Major Points

This study provides new data to inform decision making surrounding the concerns of long-term hazards of hormone use, and while previous trials often focus about individual disease outcomes, this paper approaches a broader picture look at overall long-term risk/benefit ratio of hormone therapy. And finds that all-cause mortality rates are not increased with the use of post-menopausal hormone replacement therapy.


Observational follow-up of two Women's Health Initiative randomized trials.

  • N=27347 postmenopausal women ages 50-79
    • 16608 women with a uterus were randomized to receive daily oral conjugated equine estrogens (CEE) and medroxyprogesterone acetate (MPA) versus placebo
    • 10739 women with hysterectomy randomized to receive daily oral CCE versus placebo
  • Setting: 40 US clinical centers
  • Enrollment: 1993-1998
  • Follow-up: through to December 31, 2014 (median 18 years cumulatively)
  • Analysis: Intention to treat
  • Primary outcomes: Mortality determined by surveillance through the National Disease Index (NDI) and reports of next of kin/postal service.
  • Other outcomes evaluated:
    • All cause mortality in overall pooled cohort as well as individual cohorts
    • Intervention all-cause mortality and post-intervention all cause mortality
    • Age stratified analyses
    • Cardiovascular (CV) mortality, cancer mortality and other mortality pooled, un-pooled, intervention, post-intervention, and age stratified


Inclusion Criteria

  • Postmenopausal female volunteers of any race and ethnicity, with or without a uterus and ovaries
  • Ages 50-79 years inclusive at first screening contact
  • Women must reside in the study area for at least 3 years after randomization

Exclusion Criteria

  • Risk of survival of less than 3 years by judgment of a clinic physician
  • Alcohol or drug use disorder, mental illness including severe depression, dementia, active participants in another trial
  • Invasive cancer of any type in past 10 years or breast cancer at any time
  • Baseline mammogram or clinical breast examination findings suspicious of breast cancer
  • Acute myocardial infarction in the past 6 months
  • Stroke or TIA within the past six months
  • Known chronic active hepatitis or severe cirrhosis
  • Severely underweight (BMI <18 or loss of 15 or more pounds in previous 6 months without dieting)
  • Hematocrit under 34% or hemoglobin under 115
  • Platelets less than 75
  • Severe hypertension (systolic greater than 200 or diastolic greater than 105)
  • Current use of oral daily corticosteroids for more than six months
  • BMI greater than 40
  • Endometrial cancer at any time
  • Endometrial hyperplasia at baseline bopsy or endometrial thickness greater than 5mm via ultrasonography
  • Malignant melanoma at any time
  • History of nontraumatic PE, DVT associated with estrogen or OCP use, or DVT or PE in past 6 months
  • Previous osteoporosis-related fracture being treated with HRT
  • Bleeding disorder requiring transfusion
  • Known history of hypertriglyceridemia on baseline blood draw
  • Currently on anticoagulants
  • Currently on tamoxifen
  • Severe menopausal symptoms that would make placebo therapy intolerable to the participant
  • Social situations likely to result in poor retention or adherence


  • Women with a uterus were randomized to receive daily oral CEE 0.625mg + MPA 2.5mg versus placebo. Median treatment of 5.6 years.
  • Women with hysterectomy randomized to receive daily oral CCE 0.625mg versus placebo. Median treatment of 7.2 years. Those on postmenopausal hormones had a 3 month washout phase before starting the intervention.


Primary Outcomes

All cause mortality in the cumulative follow-up phase
Overall mortality 27.1% with hormone therapy vs 27.6% with placebo (HR, 0.99 [95% CI, 0.94-1.03]; P = .60).
Individual trials
26.4% mortality for CEE plus MPA vs 26.0% for placebo (HR, 1.02 [95% CI, 0.96-1.08]; P = .51)
For CEE alone it was 28.3% vs 30.0% for placebo (HR, 0.94 [95% CI, 0.88-1.01]; P = .11)

Secondary Outcomes

All cause mortality
Intervention phase
Pooled cohort: 4.0% with hormone therapy vs 4.0% with placebo (HR, 1.01 [95% CI, 0.90-1.14]; P = .86
CEE plus MPA: HR of 0.97 (95% CI, 0.82-1.16; P = .77)
CEE alone: HR of 1.04 (95% CI, 0.89-1.22; P = .62)
Post-intervention phase
HR for all-cause mortality was 1.04 (95% CI, 0.97-1.10; P = .28) for CEE plus MPA
HR 0.92 (95% CI, 0.85-0.99; P = .03) for CEE alone
Age-stratified analyses
Pooled cohort HR for treatment vs placebo
    • 0.69 (95% CI, 0.51-0.94) for women aged 50 to 59 years
    • 1.04 (95% CI, 0.87-1.25) for women aged 60 to 69 years
    • 1.13 (95% CI, 0.94-1.36) for women aged 70 to 79 years
    • (P value for trend by age = 0.01)
The trends by age were not statistically significant during cumulative follow-up
Cardiovascular mortality
  • Neither treatment was significantly associated with CVD mortality during the follow-up, intervention, or post-intervention phase
  • No statistically significant trends for CVD with age were observed in either trial during any study phase.
Cancer mortality
  • For breast cancer mortality, the HRs for cumulative follow-up were 1.44 (95% CI, 0.97-2.15; P = .07) for CEE plus MPA and 0.55 (95% CI, 0.33-0.92; P = .02) for CEE alone; due to heterogeneity between risk estimates (P value for heterogeneity = 0.003), we did not pool these analyses.
  • HRs for deaths from colorectal cancer or other cancers were not statistically significant in either trial
  • Neither intervention was significantly associated with total cancer mortality
  • There was a trend by age. In the pooled cohort, the HR for total cancer mortality during the intervention phase was 0.74 (95% CI, 0.48-1.14) for women aged 50 to 59 years and ranged from 1.00 (95% CI, 0.77-1.28) for women aged 60 to 69 years to 1.24 (95% CI, 0.93-1.66) for women aged 70 to 79 years (P value for trend by age = 0.05)
Other Mortality
  • The HR in the pooled cohort for other (non-CVD, noncancer) mortality (n = 2826 deaths) did not differ between the hormone therapy group (10%) and the placebo group (10.7%) (HR, 0.95 [95% CI, 0.88-1.02]; P = .14)
  • During the intervention phase, the HR for other (non-CVD, noncancer) causes of mortality was 0.59 (95% CI, 0.39-0.90; P = .01) for CEE plus MPA vs placebo but did not differ by treatment group in the CEE-alone trial.
  • There was some small differences with respect to individual groups, including COPD and dementia
  • HRs for other (non-CVD, noncancer) mortality during the intervention phase were similar across age groups for CEE plus MPA but lower for younger women than for older women in the CEE trial (P value for trend by age = 0.002). The numbers were too small to assess cause specific differences
Additional analyses
  • In additional sensitivity analyses censoring participants who took less than 80% of study pills (active or placebo) during the first 2 years, results were generally similar to intention-to-treat results, but statistical power was reduced.
  • For cumulative follow-up, the HR for all-cause mortality was 1.03 (95% CI, 0.95-1.11) for the CEE plus MPA trial and 0.94 (95% CI, 0.86-1.03) for CEE alone. The pattern suggestive of lower mortality risks for younger women when compared with older women during the intervention phase persisted in these analyses (P value for trend by age = 0.05 in pooled analyses), but age trends were not statistically significant during cumulative follow up.


  • Heterogeneity in the results between combination therapy users and estrogen-only users with respect to breast cancer mortality limited ability to pool data. For combination hormone therapy users, the breast cancer mortality HR was 1.44 and for estrogen-alone users, the HR was 0.55. While there is a non-significant difference between combination hormone therapy users and estrogen-only users, the difference may be clinically relevant.
  • Only tested a single dose of a single formulation of the hormones.
  • High discontinuation in the treatment arm (42%) and crossover to the treatment arm (10.7%)
  • Addresses mortality and not morbidity, which may also be a clinically significant outcome for the patient.


  • National Heart, Lung, and Blood Institute
  • Medication and placebo donated by Wyeth-Ayerst Research

Further Reading

  • Manson JE, Aragaki AK, Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Chlebowski RT, Howard BV, Thomson CA, Margolis KL, Lewis CE, Stefanick ML, Jackson RD, Johnson KC, Martin LW, Shumaker SA, Espeland MA, Wactawski-Wende J, for the WHI Investigators. Menopausal Hormone Therapy and Long-term All-Cause and Cause-Specific Mortality The Women’s Health Initiative Randomized Trials. JAMA. 2017;318(10):927–938. doi:10.1001/jama.2017.11217
  • McNeil M. Menopausal Hormone Therapy Understanding Long-term Risks and Benefits. JAMA. 2017;318(10):911–913. doi:10.1001/jama.2017.11462
  • The Women’s Health Initiative Study Group. Design of the Women’s Health Initiative clinical trial and observational study. Control Clin Trials. 1998;19(1):61-109.
  • Rossouw JE, Anderson GL, Prentice RL, et al; Writing Group for the Women’s Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA. 2002;288(3):321-333.
  • Anderson GL, Limacher M, Assaf AR, et al; Women’s Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women’s Health Initiative randomized controlled trial. JAMA. 2004;291(14):1701-1712.