Public Health or Private Profit?

Abstract

We think the article by Sarrel et al.1 falls far short of the Journal's standards. While the post hoc analysis of the Women’s Health Initiative (WHI) results that Sarrel et al. employ is a legitimate tool for hypothesis generation, the authors present their conclusions as far more certain than that and even go so far as to recommend that their findings should inform clinical counseling. Given this, we were not surprised to read in the Acknowledgments that the lead author had been a paid consultant for a company that makes an estrogen therapy patch. We were, however, shocked and disappointed to see the nation’s leading public health journal provide a platform for such a misleading presentation of the current state of knowledge about use of estrogen therapy to prevent disease. This article purports to estimate the number of excess deaths caused by avoidance of estrogen, following the release of the WHI results. The authors ignored, however, the inconvenient fact that the WHI found no protective cardiovascular benefit for estrogen-alone compared with placebo.2 Instead, their analysis relies on an estimate derived from a post hoc subgroup analysis of the WHI, suggesting that estrogen may have a preventive health benefit for hysterectomized women aged 50 to 59 years. They then used this post hoc finding as the foundation for a mathematical model estimating how many women died because they did not take estrogen. Furthermore, the authors make claims about the benefits of estrogen that are not supported by evidence, and in some cases not even supported by a citation. For example, they state that “transdermal estradiol has been reported to be more effective than are oral estrogens in preventing cardiovascular events.”1(p1587) In addition, they never mention the fact that the WHI estrogen-only trial was stopped early because it found an increased risk of stroke and blood clots. Consequently, the authors’ conclusion that “decline in ET use since 2002 has resulted in a significant increase in mortality for hysterectomized women aged 50 to 59 years.”1(p1587) is based on a flawed study design and significantly misrepresents the state of the evidence on the health effects of taking estrogen. And because reporters rely on the Journal as a reliable source of health information, this misleading estimate was reported as a proven fact.3 We have come to expect drug company sponsored researchers to put forward sensationalistic and misleading claims about the benefits of estrogen, but we expect and deserve better from the Journal.