Our reviewers felt the story exaggerated the study’s findings. The Associated Press story was better.
A new trial called the Kronos Early Estrogen Prevention Study (KEEPS) contradicts findings from the well-known Women’s Health Initiative (WHI), by showing that a different form of estrogen and progesterone given early after menopause did not cause harmful changes in arteries that might lead to heart attack or stroke. The main purpose of this trial of hormone therapy was to see if, when given close to menopause, it might help protect arteries from changes that could lead to future heart attack and stroke. The researchers reported a negative trial result: the hypothesized benefit wasn’t present. The study was not an endorsement “supporting” estrogen.
The results showed that there was NO effect, beneficial or harmful, on “hardening of the arteries” in the blood vessels supplying the brain and the heart. The story fails to point out the intermediate endpoints that look at risk factors for heart attack and stroke (such as cholesterol, and signs of thickening or calcium deposits in arteries) are not what we care about, which is whether heart attacks and strokes were any more or less common. Since it takes longer to detect such events, the researchers chose to look at markers for future disease rather than cardiovascular events. That’s a legitimate way to explore a hypothesis, but must be emphasized as preliminary. Don’t forget that we got into the widespread but misguided practice of prescribing estrogen for cardiovascular prevention back in the 80s based on just such assumptions, based on data that showed HT had favorable effects on cholesterol and other markers–yet in the end it caused more heart attacks and strokes.
Hormone therapy was a popular way to combat menopausal symptoms for decades. But when the WHI suggested that estrogen and progesterone treatments may increase risks for heart attacks and strokes, many women stopped taking their hormone replacements. The WHI ran for 15 years and enrolled more than 160,000 postmenopausal women, the majority of whom were older and well beyond menopause. KEEPS, on the other hand, enrolled younger postmenopausal women and used different formulations of hormone therapy. It followed 727 women for 4 years and only looked at markers of cardiovascular risk, not at actual heart attacks and strokes. The results are of value as a preliminary finding, but the limits of the study must be kept in mind.
The story does not detail costs associated with estrogen and progesterone hormone therapies. For example, are low-dose estrogen pills or transdermal estrogen patches comparable in price?
The article does a good job in cautioning the reader that the study’s sample size was too small and that the study time too short to draw firm conclusions of “no increase in breast cancer, stroke, heart attack or blood clots.” However, the article states : “Women on both forms of estrogen reported fewer hot flashes, night sweats and mood swings, as well as an improvement in bone density,” but this statement is not supported with any data in the story. Similarly, the story points out that estrogen in pill form, but not the patch, improved symptoms of depression, anxiety and tension – but again, these benefits were not quantified. In addition, how were these benefits measured by researchers?
The statement about the NIH-funded branch study on mood symptoms and cognition was particularly awkward. From the little we know of the published description of the KRONOS trial, it was not designed to look at effects of hormone therapy vs placebo specifically in women with mood and/or cognitive symptoms related to menopause. Showing a global benefit on such symptoms is not surprising, but we don’t know enough to understand whether the methods used were appropriate to the task. This study should not be interpreted as a justification for menopausal women to run to the doctor for estrogen prescriptions if they are feeling depressed or noticing some trouble remembering names–and the article doesn’t make that clear.
The article states, “Women taking estrogen in pill form saw an increase in HDL or “good” cholesterol and a decrease in LDL or “bad” cholesterol, but an increase in triglycerides.” But what does that mean? Would an increase in triglycerides be harmful to one’s health? The article does not mention any other adverse effects, but it may have been beneficial to reinforce the notion that the study lacks the absolute numbers and statistical power to claim, “No significant differences in adverse events.”
The story does a good job in pointing out some limitations of the evidence – small sample size and short study time. The article also acknowledges that these are preliminary findings presented at a conference. However, it would have been helpful to know more specifically how KEEPS differed from WHI and why the new findings should not replace current medical wisdom of hormone therapy – especially since the article emphasizes that KEEPS contradicts the results of WHI, a “much larger government-funded study.”
The story does not commit disease mongering.
No independent sources are cited in the story. Direct citations come from two sources – both of whom are researchers involved in the study. The story does bring up the fact that KEEPS was funded by the Kronos Longevity Research Institute, a nonprofit organization in Phoenix. No conflicts of interests are listed. But it’s not clear who actually paid for this substantial study. A non-profit institute does not mean that it is free of pharmaceutical industry support. The organization Kronos institiute is supported by the nonprofit “Aurora Foundation”, which gave $34 million for this study according to Indiana University Philanthropy website. Who are they?
The story does not detail the traditional treatment plan, or existing alternatives, for symptoms of menopause.
It is general knowledge that estrogen and progesterone hormone therapies are common and widely available. The story cited some data on the decline in estrogen and progesterone use in the US.
The article explains that this type of study had been conducted before with the WHI.
But we would have hoped for a little more explanatory background and context. The article doesn’t adequately address the point that it is a novel study, because it looks specifically at early menopausal women and effects of hormone therapy on blood vessels – a not unreasonable early surrogate for harder endpoints such as CV disease events. The findings are actually not novel. The WHI analysis of a subset of younger menopausal women, published in 2007, showed that there was less coronary artery calcification in women treated with Estrogen and Progesterone compared with those who took placebo; a finding the KRONOS researcher thought they would also see, but did not. (They report a nonsignificant trend toward less calcification).
The story does not appear to rely on a press release, as there is evidence of original reporting with the two quotations.