The Women’s Health Initiative Postmenopausal Hormone Trials: Overview and Baseline Characteristics of Participants MARCIA L. STEFANICK, PHD, BARBARA B. COCHRANE, PHD, JUDITH HSIA, MD, DAVID H. BARAD, MD, JAMES H. LIU, MD, AND SUSAN R. JOHNSON, MD Ann Epidemiol 2003;13:S78–S86.  2003 Elsevier Inc. All rights reserved. KEY WORDS: Hormone Replacement Therapy, Estrogen, Progestin, Randomized Clinical Trial, Disease Prevention, Coronary Heart Disease, Breast Cancer, Osteoporosis, Women’s Health, Postmenopausal Women. INTRODUCTION The postmenopausal hormone therapy (PHT) component of the Women’s Health Initiative (WHI) is composed of two randomized, placebo-controlled, double-blind trials in postmenopausal women aged 50 to 79 years at initial screening, testing the effects of estrogen alone (E-alone) and estrogen plus progestin (E + P) on coronary heart disease (CHD) as the primary outcome, hip and other fractures and colorectal cancer as secondary outcomes, and pulmonary embolism, breast and endometrial cancers as potential risks. The design and rationale of the PHT trials, including general eligibility and exclusion criteria and considerations regard- ing sample size and statistical power, have been described previously (1). Postmenopausal hormones have been initiated in meno- pausal women for the treatment of vasomotor symptoms, mood disturbances, vaginal dryness, and prevention of rapid bone loss for several decades. Despite a paucity of data on effects of initiating hormone use in older women, postmeno- pausal hormones have also been promoted for the preven- tion of CHD, osteoporotic fractures, and other diseases that occur years after menopause (2). It is generally recom- mended (2) that women with a uterus be prescribed a combi- nation of estrogen and progestin to prevent endometrial From the Stanford Center for Research in Disease Prevention, Stanford University, Palo Alto, CA (M.L.S.); Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, WA (B.B.C.); Lipid Research Center, George Washington University, Washington, DC (J.H.); Department of Obstetrics & Gynecology, Albert Einstein College of Medi- cine, Bronx, NY (D.H.B.); Department of Obstetrics & Gynecology, University of Cincinnati, Cincinnati, OH (J.H.L); and College of Medi- cine, University of Iowa, Iowa City, IA (S.R.J). Address correspondence to: Dr. B.B. Cochrane, Cancer Prevention Research, Public Health Sciences, 1100 Fairview Avenue North, P.O. Box 19024, Seattle, WA 98109, USA. Tel.: (206) 667-6814; Fax: (206) 667- 4142. E-mail: barbc@whi.org Received December 20, 2002.  2003 Elsevier Inc. All rights reserved. 1047-2797/03/$–see front matter 360 Park Avenue South, New York, NY 10010 doi:10.1016/S1047-2797(03)00045-0 hyperplasia or cancer, whereas women with a hysterectomy receive unopposed estrogen. The purported benefits of estro- gen are assumed to be similar for combined hormones, al- though relatively few studies have included long-term estrogen plus progestin users, particularly those taking con- tinuous progestin. Reports of greater risk of breast cancer with cyclic estrogen/progestin combinations vs. unopposed estrogen (3, 4) highlight the need to determine the risks and benefits for both estrogen and combined hormones in appropriate clinical populations, including older women. None of the clinical trials of postmenopausal hormones for cardiovascular endpoints completed previously, e.g., the PEPI study (5), HERS (6), ERA trial (7), or WEST (8), have provided information on the role of hormones in pri- mary prevention of heart disease, nor was there clinical trial evidence that hormones prevent osteoporotic hip fractures (9) or increase breast cancer. A large randomized, controlled trial of postmenopausal hormones involving predominantly women without prior CHD or osteoporosis is needed to determine overall benefits and risks of long-term hormone use. WHI set out to randomize 27,500 ethnically diverse women into such a program for an 8.5-year period. Because women with a uterus were assigned to placebo or estrogen plus progestin, whereas women who had a hysterec- tomy were assigned to placebo or estrogen alone, the WHI hormone component is designed as two separate trials. Data are therefore presented for the total hormone component, as well as for the two distinct cohorts, i.e., those participating in the E + P trial and those participating in the E-alone trial. METHODS Eligibility Criteria and Screening Details regarding eligibility criteria and the screening pro- cess, including hormone component-specific reasons for excluding participants, appear in Hays’ article in this issue.