No Wonder We're All Confused:
Hormone Therapy and the Communication of Medical Information
December 7, 2005
On December 7, 2005, the Jacobs Institute convened a panel of experts to discuss hormone therapy and the communication of medical information in the wake of the Women's Health Initiative (WHI) trial results. The Capitol Hill briefing outlined the current state of knowledge on hormone therapy (HT) for menopause, physician and patient knowledge of the WHI trial results, how the media communicates medical information, and racial and ethnic differences in the experience of menopause. A recurring theme of the briefing was the challenge of communicating the difference between absolute and relative risk to health care providers and the general public. Relative risk compares the likelihood that a person taking a drug will have an adverse event to the likelihood that a person taking a placebo will have one. For example, the WHI estrogen+progestin trial found a 29% greater chance for coronary heart disease in the group taking HT. Absolute risk, on the other hand, gives an actual number of health problems that occur or are prevented because of the drug. For example, for every 10,000 women using e+p, 7 more women will have coronary heart disease each year. Absolute risk may be a more useful concept for women to use when making health care decisions.
Moderator Susan Dentzer introduced the panel and JoAnn Manson, MD, DrPH, opened the discussion by providing an overview of our current state of knowledge on hormone therapy. She reviewed findings from observational trials of hormone therapy and coronary heart disease (CHD) risk factors. Observational studies found that hormone therapy had potential benefits for CHD. Manson next discussed the findings of the WHI. The estrogen+progestin trial, which was stopped early, found more harm than benefit from HT, with increased risk of coronary heart disease, stroke, pulmonary embolism and breast cancer, and decreased risk of hip fracture and colorectal cancer. The estrogen-alone trial was also stopped early; researchers found an increased risk of stroke outweighed the decrease in hip fractures. Analyzing age groupings of women in these trials, Manson found that women in the 50 to 59 age range had fewer adverse outcomes than older women (the average age of the WHI was 63). Looking at this data and observational studies, she theorized that initiating HT early in menopause may have a cardioprotective effect for women. She argued that new clinical trials of early HT intervention are necessary and discussed the Kronos Early Estrogen Prevention Study (KEEPS), which will look at this theory. She concluded that postmenopausal HT should not be initiated or continued for the prevention of cardiovascular disease or other chronic conditions, but that HT does have a role in the treatment of moderate-to-severe hot flashes and other menopausal symptoms, the lowest effective dose should be used for the shortest duration. And she noted that additional research on the benefits and risks of HT initiated early in menopause are needed.
Stan Williams, MD, followed with a discussion of what doctors and patients know about the WHI findings. He discussed the challenges of conveying trial findings to patients and stressed the need for health care providers to present the data so that women can better understand their individual risk. Williams first discussed results of a survey of patients (n=1,076). Williams survey found that 70% of women us HT for relief of menopause symptoms and that concern of side effects and risk and their physician's recommendation were the most important factors influencing their decisions. Williams asked the women a series of questions about their knowledge of the risks and benefits of HT found by the Women's Health Initiative. Survey results demonstrated that the majority of women could not identify the correct risk. About a third of more of the women surveyed thought the risks for heart disease, stroke, venous thrombosis and breast cancer were greater than those found in the trial, and 79 percent thought there was an increased risk of death when there was none. Williams' survey of 600 physicians, practicing obstetrics/gynecology internal medicine, and family practice, revealed that doctors were not necessarily better informed than their patients. The majority of doctors knew which conditions increased or decreased in risk with taking HT, but they were not well-informed on the magnitude of the risk, with 67 percent giving overestimates and 28 percent giving the correct answers. Williams found that ob/gyns had the fewest wrong answers and that respondents were most often wrong about the risk of coronary heart disease and the majority overestimated the benefit of reduced hip fracture. Williams concluded that there is a great need for physician education about the WHI HT findings so that they can accurately counsel their patients. He also stressed the need for greater public education about the risks and benefits of HT, and focusing on the presentation of such information so that it can be easily understood by the lay public.
Kim Walsh-Childers, Ph.D., spoke next about news coverage of the WHI results. She discussed issues within the media that shape how health news is presented and what health news is covered. She noted that some researchers find the media only covers rare, dramatic or unusual health issues, other researchers find that the news media only covers issues relevant to middle class mainstream audiences. She also discussed issues of accuracy in health reporting, noting the danger of sensationalism, failures to report conflicts of interest, and failures to understand or convey the iterative nature of medical research. Looking at newspaper coverage of the WHI results, Walsh-Childers found common themes: talk to your health care provider, one size doesn't fit all, major shift in medical community's belief about HRT, and fall out for pharmaceutical companies. Regarding the communication of risk, she found that some stories did put this in terms women could understand but that absolute risk statements often appeared relatively deep in the story, often after the relative risk statements. She also noted that only half the newspapers she reviewed distinguished between the two types of hormone therapy. She recommended that absolute risk statements appear earlier in news stories, that writers chose language that is not easily misinterpreted and that stories specify which women the study results pertain to.
Sherry Sherman, Ph.D., the final speaker, discussed the Study of Women's Health Across the Nation: SWAN. The SWAN study aims to understand ovarian aging, quality of life, bone density and body composition, and cardiovascular risk factors. One objective of SWAN is to study the natural history of menopause and the decline in ovarian function in five different racial/ethnic groups. The study included Chinese, Japanese, Hispanic, African American and Caucasian women. SWAN found that not all women experience menopause at the same age. It also found that some symptoms attributable to menopause were prevalent before menopause. For example, 46% of women in the cross sectional survey (n=16,000) complained of stiffness/soreness premenopause and 55% complained about this post menopause, for difficulty sleeping, 31% complained premenopause and 40% post, and for forgetfulness, 31% pre and 42% post. Interestingly, the survey found that a high percentage of women (49%) complain of hot flashes/sweats post menopause, compared with 19% premenopause and 57% in late perimenopause. Looking at racial and ethnic differences in symptom frequency, the SWAN study found that a higher percentage of African Americans suffer from hot flashes than other groups. In the longitudinal study (n=3,150), researchers found that body mass index and race/ethnicity relate to hormone levels. Sherman concluded with a discussion of bone mass density. She noted that BMD is higher among African American women. However, when BMD is adjusted for weight (it increases with weight), and women under 70 kgs are compared, African American, Japanese and Chinese women have similar bone density, and white women have lower BMD that these groups.
JoAnn Manson, MD, DrPH, Harvard Medical School, Brigham and Women's Hospital
R. Stan Williams, MD, Harry Prystowsky Professor of Reproductive Medicine and Associate Chairman, Department of Obstetrics and Gynecology and Chief, Reproductive Endocrinology and Fertility, University of Florida
Kim B. Walsh-Childers, PhD, Professor of Journalism, University of Florida College of Journalism and Communications
Sherry Sherman, PhD, Program Director, Clinical Aging and Reproductive Hormone Research, National Institute on Aging, NIH