The Women's Health Initiative (WHI) clinical trials generated definitive answers to the specific questions they were designed to answer. Menopausal hormone therapy in the form of oral conjugated equine estrogens (CEE) alone or CEE plus medroxyprogesterone acetate (CEE+MPA) should not be used for primary prevention of chronic diseases among postmenopausal women because the risks outweigh the benefits. Before the WHI clinical trials were stopped early and those results were announced, there was a general, although not universal, consensus in the scientific literature and public at large that menopausal hormone therapy's benefits outweighed its risks. In the ten years since the surprising early termination of those two WHI clinical trials, substantial discussion, critique, reanalysis, and opinion has been offered to reconcile the differences between the WHI clinical trial results and a diverse set of a priori and a posteriori expectations. Some of that assessment has focused on epidemiologic studies, which had provided much of the data on which the original decision to launch the WHI trials was based. This review discusses a number of potential lessons that current and future epidemiology could take from the WHI. Epidemiologic observational studies should more often emulate the big-picture perspective of randomized clinical trials. Even apparently conflicting epidemiologic study observations and clinical trial results may have similar underlying data. Creative use of both intervention and observational study designs and data, for both menopausal hormone therapy and other important exposures, is essential to generating the research, clinical, and translational findings that advance public health. This article is part of a Special Issue entitled 'Menopause'.
Keywords: Epidemiology; Observational study; Randomized clinical trial; Study design; Timing hypothesis.
Copyright © 2013. Published by Elsevier Ltd.