|
| Women & Health 12/31/2003 V.38; N.3 3 |
|
| Why Do Professionals Disagree? The Case of Hormone Replacement Therapy and Coronary Heart Disease Prevention | |
| Author | |
| Derry, Paula S. | |
| Article | |
|
ABSTRACT. Why have professionals
disagreed about whether midlife women should be advised to use hormone
replacement therapy (HRT) for prevention of coronary heart disease (CHD)?
Because the evidence has been incomplete and could be interpreted
differently by different professionals, the question with regard to HRT
and CHD prevention thus is not "What does the evidence prove?" but rather
is "What are the decision rules by which research can be evaluated and
made sense of?" The present article attempts to clarify the problem by
cataloging dimensions along which professionals differ. These dimensions
include the weight to be given to epidemiological vs. clinical trial data;
whether a conclusion has already been drawn based on available evidence;
whether a theoretical rationale exists; whether a professional is oriented
to clinical work or research; and whether data is
distorted. KEYWORDS. Menopause, hormone replacement therapy, women's health Coronary heart disease (CHD) prevention has played a central role in hormone replacement therapy (HRT) cost/benefit analyses and has been the object of much attention from researchers CHD is the most common cause of death among older women, a medication that reduces this risk would be of great public health importance (Barrett-Connor & Grady, 1998). Because many of the possible costs and dangers of HRT use, such as breast cancer, are less common than heart disease, when a CHD benefit is assumed in a cost/benefit analysis an overall benefit for the majority of women is found (Col et al., 1997). Prior to 2002, professional opinion was divided with regard to whether HRT is a sensible strategy for CHD prevention, but the majority opinion favored HRT use. With the publication of the results of the Women's Health Initiative (WHI) clinical trial (Writing Group for the Women's Health Initiative Investigators, 2002), a sea change occurred. The majority opinion now is that HRT is ineffective and even harmful with regard to CHD prevention, although differences of opinion still remain (e.g., National Institutes of Health, 2002). How is it possible that different professionals arrive at discordant conclusions, or that a professional group will reverse itself within a few years of having adopted a policy? When evidence is incomplete and usually ambiguous, different professionals can interpret the data differently. Thus, rather than asking "What does the evidence prove?" a more fruitful approach is to ask: "What are the decision rules by which research can be evaluated and made sense of?" In this article, I examine variations in how the HRT/CHD literature was assessed prior to the WHI, and identify underlying decision-rules and values that continue to influence judgments about HRT use. WHAT DO WE KNOW? Epidemiology and Plausible Mechanisms The conclusion that HRT effectively prevents CHD was developed in the absence of clinical trial data. The primary sources of evidence about HRT and CHD prevention were a large body of epidemiological data supported by a large literature on physiological mechanisms that could plausibly underlie a beneficial effect. Additional sources of evidence included data on rates of heart disease and CHD death in oophorectomized women and monkeys and data on CHD rates in premenopausal vs. postmenopausal women. The great majority of more than 40 longitudinal, case-control, and other epidemiological studies reported large differences in CHD between HRT users and nonusers (30%-50% lower incidence of CHD) (see reviews by Barrett-Connor & Grady, 1998; Grodstein & Stampfer, 1998). Results have been reported on studies with women prescribed unopposed estrogen and estrogen with a progestin (Barrett-Connor & Grady, 1998), and for HRT prescribed for both primary and secondary prevention (Grodstein & Stampfer, 1998). The consistency of the epidemiological results was impressive, especially when compared with other literatures such as that linking HRT use to breast cancer risk, where studies may be in conflict in ways that are difficult to interpret. Even when major risk factors for CHD were controlled for, a large difference between hormone users and nonusers remained (Grodstein et al., 1996; Stampfer et al., 1991). The epidemiological work was supported by plausible physiological mechanisms that might underlie a protective effect. The beneficial effects of estrogens on cholesterol levels have been estimated to account for one quarter to one half of the CHD benefit (Bush et al., 1987). In addition, estrogens affect vascular reactivity (at least in unhealthy veins), have antioxidant properties, and have other beneficial effects (Herrington, 1997). Given the large difference in disease rates between hormone users and nonusers, the consistency of this difference across studies, the persistence of the result when known risk factors were controlled for, and the underlying biological plausibility, some epidemioiogical researchers (e.g., Grodstein & Stampfer, 1998; Stampfer et al., 1991) concluded that even without clinical trials the case for HRT was strong. It appeared implausible that confounding factors could produce such consistent results, or could be responsible for more than a small part of the benefit found. Many clinicians concluded, as a practical matter, that the best judgment that could be made, based on available evidence, was that a heart disease benefit existed (Grady et al., 1992). Some clinicians and researchers, however, remained unconvinced and argued that the case for HRT could not be proven in the absence of clinical trial data (see reviews in Barrett-Connor & Grady, 1998; Barrett-Connor & Stuenkel, 1999). Impressive as the epidemioiogical data were, correlational data cannot demonstrate cause and effect. The possibility remained that confounding factors were responsible for the differences found between hormone users and nonusers. Differences (e.g., in specific health behaviors) between hormone users and nonusers remained after controlling for major known risk factors (Barrett-Connor, 1991). Even before menopause, the long-term hormone users had more favorable risk profiles for CHD (Matthews, Kuller, Wing, Meilahn, & Plantinga, 1996). Further, only a small minority of women remains compliant with hormone use for long periods of time; researchers have documented large cardiovascular benefits, of a size comparable to the HRT benefit, for compliant (vs. noncompliant) participants in placebo groups in clinical trials of other cardiovascular medications (Petitti, 1994). High-risk women may be systematically over represented in the nonuser group. For example, for many years patient information inserts cautioned women at high risk for CHD, such as diabetics or women with high blood pressure, about possible risks to them from HRT because estrogens in birth control pills had caused circulatory problems. Evidence other than epidemiological has been similarly considered and countered. For example, it is often asserted that after menopause women lose their protection against heart disease. However, when CHD death is plotted against age, a steady acceleration with increasing age is observed with no shift associated with menopause (Barrett-Connor & Stuenkel, 1999; Bush, 1990). Another assertion was that after menopause, female rates of CHD quickly approach those of male. However, death rates of women do not approach those of men until women are in their 80s. At this age, the age-CHD curve for women is not changing whereas the curve for men is flattening (Tunstall-Pedoe, 1998). Oophorectomized women do have higher rates of heart disease, but they may differ medically from other women in unknown ways that can account for the elevated risk (Tunstall-Pedoe, 1998). Further, HRT does not completely eliminate this elevated CHD risk. Oophorectomized monkeys given a placebo exhibit greater heart disease than do those given HRT, but only when fed a diet associated with arteriosclerosis (Clarkson, 1998). Heart and Estrogen/Progestin Replacement Study (HERS) Clinical trials are considered to be the "gold standard" that establish cause and effect. Blinded, random assignment of participants to experimental and control groups is the only acceptable methodology that satisfies FDA requirements for safety and effectiveness before a new medication is approved. HRT was never approved by the FDA for heart disease prevention because no clinical trial data demonstrating a CHD benefit for HRT existed. The first large trial, which had secondary end-points, the Postmenopausal Estrogen/Progestin Intervention (PEPI) trials, appeared in 1995 (Writing Group for the PEPI Trial, 1995). The first large randomized trial with clinical outcomes was not published until 1998. This study, the Heart and Estrogen/Progestin Replacement Study (HERS) (Hulley et al., 1998; Hulley et al., 1999), failed to confirm the epidemiological results. Because of the special significance attributed to clinical trial data, the results of this study elicited strong reactions from the medical community and provide a focus for a discussion of decision-rules that might govern conclusions drawn about HRT use. HERS was a double-blind, placebo-controlled secondary prevention trial in which 2,763 women with diagnosed preexisting heart disease were randomly assigned to receive either HRT (Prempro: 0.625 mg conjugated equine estrogens plus 2.5 mg of medroxyprogesterone acetate) or placebo. After an average of 4.1 years, there were no differences in heart disease outcomes between the experimental and control groups using clinically significant, robust measures, including rate of nonfatal myocardial infarction (MI), death from heart disease, and other measures. However, the HERS researchers also reported a time trend in a secondary, post-hoc analysis wherein: "Within the overall null effect, there was a statistically significant time trend, with more CHD events in the hormone group than in the placebo group in year 1 and fewer in years 4 and 5" (Hulley et al., 1998). They suggested that this time trend might be a statistical artifact, a consequence of HRT having harmful effects early on (so that only healthier participants remained later in the study), or perhaps a biologically meaningful result in which HRT has initial harmful effects that are offset over time by beneficial effects. The main result of HERS, that women who were in the HRT and placebo groups did not differ in the number of coronary problems, elicited a large number of commentaries. The results were regarded as "surprising" and "disturbing" (Pan & Boal, 1999), as raising the specter of "paradigms lost" (Herrington, 1999), or as a reminder of the limitations of epidemiological research and the need for experimental data (Pettiti, 1998). The methodology of the study was criticized and the criticisms responded to. Barrett-Connor and Stuenkel (1999) summarized criticisms of the HERS study as lacking statistical power, dealing with an atypical population, using a progestin that created problems not found with other progestins, or having no relevance for primary prevention. The study's defenders asserted that the statistical power was adequate; that the study reliably portrayed a population typical of older women with heart disease; and that the women had been given widely prescribed medications. Some authors (e.g., Nabel, 2000) cited the statistically significant time trend and asserted that HRT was associated with excess heart problems at the very beginning of the study but with benefit in later years. Others (e.g., Derry, 2002) emphasized that the benefit in later years was not statistically significant. Some clinicians became less likely to prescribe HRT for CHD prevention, whereas others dismissed the study or asserted that had it continued longer an overall benefit would have been found. Since the publication of the HERS results, the Estrogen Replacement and Arteriosclerosis trial (ERA) has been completed (Herrington et al., 2000). In this three-year clinical trial, HRT did not provide a benefit for women with preexisting heart disease when progress of disease was measured angiographically. Other researchers have reported subsidiary analyses of experimental data in which participants had excess CHD events during the first year of HRT use (e.g., Alexander et al., 2001, for post-myocardial infarction (MI) patients). Researchers from the WHI issued announcements after the second (WHI, 2000) and third (WHI, 2001) years of the study that hormone users appeared to have a greater number of cardiac problems than did placebo controls. The HRT study (but not an unopposed estrogen study) was halted in 2002 (Writing Group for the Women's Health Initiative Investigators, 2002), because HRT users had an excess risk of breast cancer. At the time the study was terminated, the increased risk of cardiac problems for HRT users was statistically significant and the overall risks of HRT use outweighed the benefits. Notwithstanding the emerging body of research beginning with HERS, the 2002 WHI results were described as "a bombshell" (e.g., Hamdy, 2002), and professionals continue to debate what conclusions to draw from the study (National Institute of Health Scientific Workshop on Menopausal Hormone Therapy, 2002). DECISION RULES Responses to data that are unexpected help to provide a window into decision-making rules for how the evidence has been assessed by professionals who have varying interpretations of the HRT/CHD literature. This paper will analyze responses to the HERS research, since this was the first clinical trial casting doubt on the benefit of HRT for CHD prevention, in order to explore how professionals arrive at different conclusions based upon the same body of evidence. Evolving responses in light of the WHI will form the subject of another paper. The American College of Cardiology (ACC) and American Heart Association (AHA) (which have issued both joint and separate recommendations with regard to HRT use), the American College of Obstetrics and Gynecology (ACOG), and the American Association of Clinical Endocrinologists (AACE) are all professional organizations that represent groups of specialists with expertise bearing on HRT use. Each group publishes opinions, guidelines, and other materials that recommend clinical practices. This paper will compare their interpretations of the pre-WHI HRT/CHD literature, and HERS in particular. Following the publication of HERS, the heart specialists (ACC/AHA Task Force on Practice Guidelines, 1999) revised their treatment guidelines for women who have had an MI. The guidelines had previously recommended that such women be "counseled about the potential beneficial effects of ERT (estrogen replacement therapy) and offered the option of ERT if they desire it." In response to HERS, the guidelines were revised to read: "(1) Hormone replacement therapy (HRT) with estrogen plus progestin for secondary prevention of coronary events should not be given de novo to postmenopausal women after MI. (2) Postmenopausal women who are already taking HRT with estrogen plus progestin at the time of an MI can continue this therapy" [because harmful effects were found only during the first year of hormone use]. The ACC/AHA task force also suggested that more research be done and, given the weaknesses of the evidence, that the final decision be left to the patient. In 2001 the American Heart Association (Mosca et al., 2001) reiterated that HRT should not be initiated for secondary prevention and that, in addition, "[t]here are insufficient data to suggest that HRT should be initiated for the sole purpose of primary prevention of CVD (cardiovascular disease)." By contrast, the clinical endocrinology specialty group which had published guidelines after the HERS data appeared (AACE, 1999) did not modify its recommendations in response to HERS. For example, a history of MI was not on the list of AACE contraindications for HRT, and there was no recommendation that patients be informed about the HERS and its possible implications. The obstetrics and gynecology specialty group, ACOG, had published guidelines prior to the appearance of HERS and also took no action to modify its recommendations in response to the study. ACOG did not publish a letter offering a professional opinion, alter its guidelines, or include a question about a history of heart disease in their recommended perimenopausal exam. ACOG did issue a news release in response to the 2001 AHA recommendation cited above, which stated that it was "reviewing all relevant information...and any future recommendations by ACOG will be based on analysis of the available data" (ACOG, 2001). That is, neither AACE nor ACOG drew a conclusion from HERS nor gave enough weight to the study even to recommend screening patients with a history of heart problems or to provide such patients with information that future research might find HRT to be ineffective or even harmful. Weight Given Epidemiological vs. Clinical Trial Data What decision rules underlie these varying assessments and recommendations? The first decision rule is: What counts as legitimate evidence? These organizations differ in the weight given to epidemiological as opposed to clinical trial data. With regard to epidemiological work, the AHA/ACC Task Force on Preventive Cardiology (1999) stated: "The issues of [HRT] for cardiovascular disease is far from clear. Observational studies have been interpreted as indicating that [HRT] is effective in primary prevention.... Confounding factors...make it difficult to be certain of the effect of [HRT]" (p. 150). A scientific statement on preventive cardiology for women (AHA/ACC, 1999) concluded: "The recent findings from HERS have challenged previous observational data....the overall null result from HERS does not support initiation of [HRT] in older postmenopausal women with confirmed coronary disease" (p. 1752). On the other hand, the AACE (1999) guidelines state that "...many retrospective studies and several prospective studies have found that HRT was associated with a reduced recurrence of MI and [coronary artery disease] deaths.... In contrast, [HERS] found no beneficial effect of [HRT].... The differences between [HERS] and the numerous other observational studies that have shown very large decreases, remains speculative" (p. 358). The ACOG Committee on Gynecological Practice (W. Zinberg, personal communication, Nov. 17, 1999) "felt this [HERS] report flies in the face of previous findings from good observational studies. Although the Committee recognized that the HERS design is better than that of an observational study, the HERS design does have its flaws. The Committee cautioned against making sweeping conclusions on the basis of this one study." For some professionals, epidemiological data are a rich source of information about the behavior of real people in the real world, whereas clinical trials are artificial situations in which a self-chosen, possibly unusual group of participants do not experience the complexities that will be found in real life. Further, clinical trials may be inaccurate because of any of a number of methodological flaws. Some professionals believe that statistical controls compensate for lack of random assignment of participants, so that inferences about cause can reasonably be made. A professional who places great weight on the body of epidemiological data, supported by the large literature on known biological mechanisms by which HRT might well be exerting its effects, may dismiss what may be seen as "just one study" that is inconsistent with this large body of research. On the other hand, a professional who believes in the privileged status of clinical trial data to identify causal relationships accurately by eliminating the influence of confounding variables may draw a different conclusion. This clinician will be more likely to place great weight on a well-designed trial, especially one that demonstrates no benefit and some harm, even though the professional may also agree that a single study is not enough for a definitive conclusion and that more research is needed. Has a Conclusion Already Been Formed? A second decision rule, related to the first, is whether there already is conviction about what "the truth" is, i.e., whether a conclusion has already been drawn based upon existing evidence against which the new data are compared. If a scientist conducted research that demonstrated that the earth was flat, that study would be closely scrutinized and its conclusions doubted because it flies in the fact of a conclusive body of research that it contradicts. Once professionals have concluded that sufficient research exists to support a given point, discrepant research will tend to be more closely scrutinized and critiqued than research with congruent results. A greater amount of evidence is needed to dislodge a formed opinion than to sway a more fluid stance. ACOG and AACE had both drawn a conclusion with regard to HRT and CHD prevention prior to the appearance of the HERS data. ACOG guidelines for preventive and primary care during menopause (ACOG, 1996) stated that "In addition to protecting against cardiovascular disease and osteoporosis, estrogen replacement therapy is effective treatment of menopausal symptoms" (p. 103). Earlier ACOG concluded that "It has been shown that estrogen therapy protects against cardiovascular disease" (ACOG, 1995, p. 7). The AACE (1999) guidelines state that "Numerous studies attest to the postponement of the first evidence of CAD in users of HRT" (p. 358). AHA (1997), on the other hand, was more circumspect about the effectiveness of HRT for CHD prevention even before the appearance of the HERS study: "Despite the consistency and coherence of postmenopausal estrogen-associated cardioprotection, it is nevertheless possible that some or all of the association is an artifact explained by prescription, prevention, or compliance bias and differences in education and socioeconomic status" (p. 2479). Does a Theoretical Rationale Exist? A third decision rule, related to the second, is whether a coherent theory or paradigm or even a mythos exists that explains current data, against which new data are compared and within which a conclusion appears plausible, inevitable, or sensible. This underlying conviction may be tied to a judgment that the available evidence is strong enough to draw the conclusion or, alternatively, to a belief that proving this "truth" is simply a matter of collecting the right data. The clinical endocrinology group and the obstetrics/gynecology group both had such underlying frameworks. AACE (1999) guidelines assert that "menopause is a state of hormonal deficiency that should be treated" (p. 355) and that "[a]ll women should consider HRT not only as essential replacement of missing hormones but also as a type of preventive medicine" (p. 357). ACOG (1995) asserted that "Certain physiologic effects of estrogen deficiency can lead to long-term health problems such as osteoporosis and coronary heart disease" (p. 6). On the other hand, the cardiology groups did not offer a similar underlying framework about the implications of postmenopausal levels of estrogen. HRT is evaluated with the same rules for evaluating experimental data to which any clinical trial of a novel drug would be subjected. For example, HRT has a positive effect on cholesterol levels. For ACOG this was one example of a mechanism by which the beneficial effect of HRT on the cardiovascular system can be understood. On the other hand, for ACC/AHA this was only a first step. Evidence does not yet exist that HRT is effective with regard to disease end points and safety. ACC/AHA recommendations for control of cholesterol suggest that lifestyle changes be tried first. If medication is needed, statins, a class of medications for which outcome and safety data are available, are preferred, whereas HRT is a less-preferred choice. Orientation to Clinical Work vs. Research A fourth decision rule has to do with the very use of language. Clinicians must, as a practical matter, make a judgment about what is most likely to be the best course of action even when research is inconclusive or incomplete. Scientific results might be only one of many types of data upon which clinicians' judgments are based, and these judgments may draw on rules for making reliable inferences that are not based solely on scientific considerations. Whereas the FDA requires clinical trial data for proof of effectiveness, clinicians who routinely prescribe medications for off-label uses might base decisions on judgment, experience, and other criteria. Whereas a researcher may use tentative language to describe results, clinicians might make definite statements when their judgments form the basis for action; researchers who focus on clinical outcomes may use language similar to clinicians. An exchange between two professionals illustrates these points. Watts (2001) criticized a review paper on HRT and CHD prevention in his commentary entitled "Observational data do not establish cause and effect." He stated that "especially in the setting of a scientific article...[w]e must avoid even suggesting unwarranted conclusions from observational studies" (p. 772). He objected to such language as "a reduction in incidence" of CHD or "beneficial effects" of HRT when reporting on observational studies, rather than using the more accurate, noncausal phrase "lower incidence" (p. 772). Mosca (2001), the author of the review paper Watts criticized, defended her use of language: "I agree...that observational studies do not provide definitive answers. Unfortunately, observational studies are all that clinicians have in the area.... While they do not enable us to draw definitive conclusions, they do enable us to make provisional inferences.... This is common practice in translating findings from epidemiological research into decisions in clinical practice.... It is for this reason that I described a reduction in the incidence of CHD rather than a lower incidence of CHD in postmenopausal users of HRT" (p. 774). Referring to the use of words such as "beneficial" as "standard terminology," she continued: "I do not believe that my phrase implies causality, although as an epidemiologist and clinician, I respect both viewpoints" (p. 774). Thus, different experts may use the same words but mean different things by them; practitioners may use language that implies causality without requiring causal proofs when this language provides an orientation for action in an imperfect world. Although Mosca (2001) referred to these as "provisional inferences" (p. 774), they may nevertheless consolidate a clinician's conclusion about the import of the research and thus influence evaluation of discrepant information. DISTORTIONS OF DATA Closely scrutinizing and critiquing the work of others is an intrinsic part of scientific and clinical work. However, sometimes results, especially those discordant with expectations, may be discounted, criticized, or ignored in ways that appear excessive, unreasonable, unscientific, or simply erroneous. Different standards of evidence may be applied to results of different studies, or a professional may simply reject or ignore a result. The methodology of studies that support one conclusion may be closely scrutinized and critiqued, whereas no methodological observations may be made about studies that support the opposite conclusion. Simple human error, enthusiasm, excessive rigor in scrutinizing the unexpected, preconceptions, biases, promoting one's own research, and financial associations with pharmaceutical companies may all influence the way research is interpreted and reported. Although the professionals' motives are unknown, examples of these distortions of data can easily be found. The National Committee for Quality Assurance (NCQA) is a private organization that creates standards by which health organizations are evaluated. NCQA began in 2000 to survey whether women, beginning at age 45, received menopause counseling from their health care providers, with the stated goal of enabling women to make informed choices (NCQA, 2001). With regard to CHD and HRT, NCQA (2001) stated that "Long-term risks of estrogen deficiency include the development of osteoporosis and coronary artery disease.... Many studies have concluded that estrogen protects against the development of CAD" (p. 154). The situation with regard to secondary prevention is "less clear" (p. 154). About HERS, NCQA stated: "In a recent prospective clinical trial of women with CAD, women taking HRT had no reduction in risk in the first eight months but had fewer cardiac events in years three, four and five" (p. 154). The main result of HERS, that no differences existed overall between hormone users and nonusers, was not reported. Nor were we told that the result reported was from a secondary, post hoc analysis and was an interpretation rather than primary data, that the benefit in later years was not statistically significant, or that the HRT group had a statistically significant increase in risk in the first year of the study. We were not told that professionals disagree about the HRT/CHD relationship. The counseling guidelines developed for NCQA (Jacobs Institute, 2000) erroneously listed HRT as an FDA-approved treatment for heart disease. CONCLUSION CHD prevention has great significance in HRT decision-making. If no reduction in CHD exists, then for the average woman overall known risks equal or outweigh overall known benefits of long-term HRT use (Barrett-Connor & Grady, 1998; Writing Group for the Women's Health Initiative Investigators, 2002). With the announcement of the WHI HRT clinical trial results, a sea change occurred with regard to professional opinion about the use of HRT for long-term disease prevention, but differences of opinion remain. Most professionals no longer recommend HRT for long-term disease prevention; however, some professionals assert that research is needed to establish whether some HRT formulations or doses are beneficial rather than harmful for some populations of women, and professionals disagree about the safe duration of use when HRT is prescribed for relief of distressing symptoms (e.g., National Institutes of Health, 2002). Many health care decisions are made when knowledge of the facts is incomplete. Underlying decision rules, rather than the facts alone, thus influence judgments and recommendations. In the case of HRT for long-term disease prevention, a medication was prescribed for large numbers of women for long periods of time based on decision rules that influenced initial judgments and the evaluation of additional evidence (e.g., HERS and WHI data). It is important for midlife women and professionals who are making health care decisions to consider the decision-making rules that underlie how conclusions are formed and recommendations made. Whether epidemiological or clinical research is relied on, whether a firm conclusion has been formed, whether judgments follow from metatheories about what menopause is, and whether data has been distorted, are all important. These decision-making rules may influence the recommendation made and the estimates of possible benefit and harm. In addition, professionals may state as truths what in reality are professional or clinical judgments. Uncertainty about what the facts are means that both efficacy and risk of a practice are uncertain. Attitudes towards the use of a medication with uncertain efficacy is thus also an important factor when making decisions. REFERENCES Alexander, K., Newby, L., Hellkamp, A., Harrington, R., Peterson, E., Kopecky, S. et al. (2001). Initiation of hormone replacement therapy after acute myocardial infarction is associated with more cardiac events during follow-up. Journal of the American College of Cardiology, 38, 1-7. American Association of Clinical Endocrinologists. (1999). AACE medical guidelines for clinical practice for management of menopause. Endocrine Practice, 5, 355-366. American College of Obstetrics and Gynecology. (1995). ACOG Technical Bulletin Number 210: Health maintenance for the perimenopausal woman. Washington, DC: ACOG. American College of Obstetrics and Gynecology. (1996). Guidelines for women's health care: Primary and preventive care, menopause. Washington, DC: ACOG. American College of Obstetrics and Gynecology. (2001). Statement of the American College of Cardiology on the American Heart Association's position on hormone replacement therapy and cardiovascular disease. July 25, 2001. Retrieved from: www.acog.org/from_home/publications/press_releases/n07-31-01-4.htm. American College of Cardiology/American Heart Association Task Force on Practice Guidelines. (1999). ACC/AHA guidelines for the management of patients with acute myocardial infarction. Available at: www.acc.org/clinical/guidelines/nov96/edits/jac1716pVla.htm. American Heart Association. (1997). AHA medical/scientific statement: Cardiovascular disease in women. Circulation, 96, 2468-2482. American Heart Association/American College of Cardiology. (1999). AHA/ACC scientific statement: Consensus panel statement -- guide to preventive cardiology for women. Journal of the American College of Cardiology, 33, 1751-1755. Barrett-Connor, E. (1991). Postmenopausal estrogen and prevention bias. Annals of Internal Medicine, 115, 455-456. Barrett-Connor, E., & Grady, D. (1998). Hormone replacement therapy, heart disease, and other considerations. Annual Review of Public Health, 19, 55-72. Barrett-Connor, E., & Stuenkel, C. (1999). Hormones and heart disease in women: Heart and estrogen/progestin replacement study in perspective. Journal of Clinical Endocrinology and Metabolism, 84, 1848-1853. Bush, T. (1990). The epidemiology of cardiovascular disease in postmenopausal women. Annals of the New York Academy of Sciences, 592, 263-271. Bush, T., Barrett-Connor, E., Cowan, L., Criqui, M., Wallace, R., Suchindran, C. et al. (1987). Cardiovascular mortality and noncontraceptive use of estrogen in women: Results from the Lipid Research Clinics Program follow-up study. Circulation, 75, 1102-1109. Clarkson, T. (1998, November). Estrogens, phytoestrogens, SERMs and postmenopausal coronary artery atherosclerosis. Paper presented at the Women's Health Research Symposium, University of Maryland, Baltimore, MD. Col, N., Eckman, M., Karas, R., Pauker, S., Goldberg, R., Ross, E. et al. (1997). Patient-specific decisions about hormone replacement therapy in postmenopausal women. Journal of the American Medical Association, 277, 1140-1147. Derry, P. (2002). Times trends in the HERS secondary prevention trial: Much ado about nothing? Journal of the American Medical Women's Association, 57, 215-216. Grady, D., Rubin, S., Petitti, D., Fox, C., Black, D., Ettinger, B. et al. (1992). Hormone therapy to prevent disease and prolong life in postmenopausal women. Annals of Internal Medicine, 117, 1016-1037. Grodstein, F., & Stampfer, M. (1998). The cardioprotective effects of estrogens. In J. Studd (Ed.), The management of the menopause (pp. 211-219). New York: Parthenon. Grodstein, F., Stampfer, M., Manson, J., Colditz, G., Wilier, W., Rosner, B. et al. (1996). Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. New England Journal of Medicine, 335, 453-451. Hamdy. R. (2002). Lessons learned from the Women's Health Initiative. Southern Medical Journal, 95, 951-965. Herrington, D. (1997). Sex hormones and normal cardiovascular physiology in women. In D. Julian & N. Wenger (Eds.), Women and heart disease (pp. 243-264). London: Martin Dunitz. Herrington, D. (1999). The HERS trial results: Paradigms lost? Annals of Internal Medicine, 131, 463-466. Herrington, D., Reboussin, D., Brosnihan, B., Sharp, P., Shumaker, S., & Snyder, T. (2000). Effects of estrogen replacement in the progression of coronary-artery atherosclerosis. New England Journal of Medicine, 343, 522-529. Hulley, S., Grady, D., Bush, T., Furberg, C., Herrington, D., Riggs, B. et al. (1998). Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Journal of the American Medical Association, 280, 605-613. Hulley, S., Grady D., Vittinghoff, E., Bush, T., Furberg, C., & Riggs, B. (1999). Hormone replacement therapy for secondary prevention of coronary heart disease [Letter]. Journal of the American Medical Association, 281, 797. Jacobs Institute of Women's Health Expert Panel on Menopause Counseling. (2000). Guidelines for counseling women on the management of menopause. Washington, DC: Jacobs Institute. Matthews, K., Kuller, L., Wing, R., Meilahn, E., & Plantinga, P. (1996). Prior to use of estrogen replacement therapy, are users healthier than nonusers? American Journal of Epidemiology, 143, 971-977. Mosca, L. (2001). In reply [to Watts]. Archives of Internal Medicine, 161, 774. Mosca, L., Collins, P., Herrington, D., Mendelsohn, M., Pasternak, R., Robertson, R. et al. (2001). Hormone replacement therapy and cardiovascular disease: A statement for healthcare professionals from the American Heart Association. Circulation, 104, 499-503. National Committee for Quality Assurance. (2001). HEDIS 2001. Specifications for survey measures (vol. 3). Washington, DC: National Committee for Quality Assurance. National Institutes of Health. (2002). Scientific Workshop on Menopausal Hormone Therapy. Bethesda, MD, October 23-24, 2002. Nabel, E. (2000). Coronary heart disease in women -- an ounce of prevention. New England Journal of Medicine, 343, 572-574. Pan, C., & Boal, J. (1999). Hormone replacement therapy for secondary prevention of coronary heart disease. [Letter]. Journal of the American Medical Association, 281, 794. Pettiti, D. (1994). Coronary heart disease and estrogen replacement therapy: Can compliance bias explain the results of observational studies? Annals of Epidemiology, 4, 115-118. Petitti, D. (1998). Hormone replacement therapy and heart disease prevention: Experimentation trumps observation. Journal of the American Medical Association, 260, 650-652. Stampfer, M., Colditz, G., Willett, W., Manson, J., Rosner, B., Speizer, F. et al. (1991). Postmenopausal estrogen therapy and cardiovascular disease. New England Journal of Medicine, 325, 756-762. Tunstall-Pedoe, H. (1998). Myth and paradox of coronary risk and the menopause. Lancet, 351, 1425-1427. Watts, N. (2001). Observational data do not establish cause and effect. Archives of Internal Medicine, 161, 772. Writing Group for the PEPI Trial. (1995). Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. Journal of the American Medical Association, 273, 199-208. Writing Group for the Women's Health Initiative Investigators. (2002). Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Journal of the American Medical Association, 288. 321-333. Women's Health Initiative. (2000). WHI HRT update. Retrieved from: www.nhlbi.nih.gov/whi/HRT update2000en.pdf. Women's Health Initiative. (2001). WHI HRT update. Retrieved from: www.nhlbi.nih.gov/whi/HRTUpdate2001.pdf. Article copyright The Haworth Press, Inc. |
| Publication Name: | Women & Health | ||
| Volume & Number: | V.38; N.3 | ||
| Publication Date: | 12/31/2003 | ||
| Article Title: | Why Do Professionals Disagree? The Case of Hormone Replacement Therapy and Coronary Heart Disease Prevention | ||
| Page: | 3 | ||
| Author: | Derry, Paula S. |