Effects of Hormone Therapy on survival, cancer, cardiovascular and dementia risks in 7 million menopausal women over age 65: a retrospective observational study

Background The long-term influence of menopausal hormone therapy remains unanswered due to the termination of randomized clinical trials and discordant findings from observational studies. Methods From 2007-2019 enrollment records of 100% Medicare beneficiaries, we identified 7 million female enrollees aged 65 or more. We identified type, route and strength of estrogen based on their prescription drug utilization records. Using vital status record and encounter records, we defined the first onset of thirteen patient outcomes; all-cause mortality; 5 cancers (breast, lung, endometrial, colorectal, ovarian cancers); 6 CV conditions (ischemic heart diseases, heart failure, venous thromboembolism, stroke, atrial fibrillation, acute myocardial infarction); and dementia. Then, we implemented an extended Cox regression analysis to examine the effects of type, route, and strength of estrogens on each of 13 study outcomes. Findings Estrogen monotherapy (ET) exhibited a significant, 20% (aHR=0.80; 95% CI 0.78-0.82), relative risk reduction of mortality. The reduction was greater with estradiol (aHR=0.78; 95% CI 0.75-0.80) than conjugated estrogen (aHR=0.86; 95% CI 0.85-0.88), and with vaginal (aHR=0.69; 95% CI 0.65-0.74) than oral (aHR=0.89; 95% CI 0.87-0.90) and transdermal (aHR=0.78; 95% CI 0.75-0.81) preparations. ET also exhibited significant risk reductions for all study cancers, breast (aHR=0.82; 95% CI 0.80-0.84), lung (aHR=0.87; 95% CI 0.84-0.90), endometrial (aHR=0.65; 95% CI 0.62-0.69), colorectal (aHR=0.86; 95% CI 0.82-0.90) and ovarian (aHR=0.83; 95% CI 0.79-0.88). ET slightly increased risks of ischemic heart diseases (aHR=1.03; 95% CI 1.01-1.04). However, such risk was not observed with low dose ET (aHR=0.98; 95% CI 0.97-0.99). Both combination therapy (aHR=1.11; 95% CI 1.08-1.14) and progestogen monotherapy (aHR=1.09; 95% CI 1.05-1.13) exhibited a significantly increased risk of breast cancer. Oral HT exhibited a moderately increased risk of dementia. Conclusions Among female Medicare beneficiaries aged ≥65, the effect of menopausal hormone therapy varies by type, route, and strength but overall estrogen seemed beneficial.