Cover Image


View/Hide Left Panel

steady increase in HT use since the early 1980s, with a prevalence of about 38 to 40 percent among women ages 50 to 74 in 1995. Data on numbers of prescriptions for all forms of HT reveal a continued increase from 1995 to 2001, with the annual number of prescriptions peaking at 92 million in 2000 (Hersh, Stefanick, and Stafford, 2004).2

The plausibility of the hypothesis that HT use underlies the U.S. survival disadvantage rests heavily on the assumption that it is a risk factor for overall mortality. If this assumption is valid, the strength of evidence implicating HT use depends on a second assertion, namely that its use has been more prevalent in the United States than in other Western populations. In this chapter, I evaluate the evidence for these assumptions.

Observational studies have generally reported substantially lower rates of heart disease—on the order of 35 to 50 percent—for long-term users of postmenopausal HT compared with nonusers (Grodstein, Manson, and Stampfer, 2006; Grodstein et al., 2000; Manson and Bassuk, 2007; Prentice and Anderson, 2008). For example, in 1992, a meta-analysis based on 32 observational studies concluded that ever-users of estrogen had a relative risk of coronary heart disease of 0.65 compared with never-users, a finding generally in line with two previous meta-analysis estimates of 0.55 and 0.58 (Grady et al., 1992). However, during the past decade, randomized controlled trials (RCTs) have challenged findings from observational studies regarding the harm and benefits of long-term use of HT. Although both types of studies are in general agreement about its benefits for colorectal cancer and hip fracture and the increased risk associated with breast cancer, they have produced widely discrepant estimates for coronary heart disease or CHD (Grodstein, Clarkson, and Manson, 2003; Nelson et al., 2002).

Data from the Women’s Health Initiative (WHI), a well-publicized randomized trial that administered the two dominant forms of hormone therapy to healthy U.S. women ages 50 to 79,3 suggested an increased incidence of CHD and thromboembolic events, as well as all cardiovascular disease combined, based on an average of 5.2 years of use of the estrogen-progestin regimen (Nelson et al., 2002; Writing Group for the Women’s Health Ini-


The two most common forms of HT in the United States are (1) a mixture of conjugated equine estrogens, typically referred to as unopposed estrogens, and (2) a combination of estrogen and progestin (a synthetic substance with effects similar to progesterone). In light of evidence suggesting an increased risk of uterine cancer associated with unopposed estrogens, women with an intact uterus have been using estrogen-progestin combinations. The majority of prescriptions in the United States have been for orally administered hormone therapy; transdermal and vaginal preparations have been used much less frequently (Hersh et al., 2004).


Initiated in 1992, the WHI enrolled about 27,500 postmenopausal women into two parallel clinical trials designed to evaluate the consequences of hormone therapy for disease prevention (particularly CHD and fractures). Women without a uterus were randomly assigned to receive unopposed estrogens or a placebo, and women with an intact uterus were randomly assigned to receive an estrogen-progestin combination or a placebo.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement