Combined estrogen and progestin therapy (HT) and estrogen-only therapy (ET) have been used widely by peri- and postmenopausal women for more than 40 years to manage menopausal symptoms and, more recently, for the prevention of osteoporosis and coronary heart disease (CHD).1 The vast majority of observational evidence has suggested that HT and ET reduce the incidence of CHD,2–5 bone fractures,6,7 and colorectal cancer8,9 in postmenopausal women. In the mid-1990s, an estimated 38% of postmenopausal women in the United States used HT or ET.1
The Women's Health Initiative (WHI) was designed to examine the risks and benefits of HT, ET, and diet on heart disease, bone fractures, and breast and colorectal cancer in predominantly healthy postmenopausal women.10 On May 31, 2002, the Data and Safety Monitoring Board stopped the WHI HT randomized trial (HT versus placebo for women with an intact uterus) after an average of 5.2 years of follow-up (2.5 years early) because the risks of HT were found to outweigh the benefits.11 Women who were assigned randomly to HT experienced an excess risk of invasive breast cancer, CHD, stroke, venous thromboembolism, and pulmonary embolism compared with women who were assigned randomly to placebo. The use of HT was associated with a reduced risk of colorectal cancer11 and bone fractures.12 The HT trial early-termination results were released to the general public and scientific community on July 9, 2002.11 The trial of ET for women who had hysterectomies before study enrollment also ended early, on March 4, 2004, after nearly 7 years of follow-up because there was no overall benefit on disease outcomes among women assigned randomly to ET compared with placebo.13 The observational trial of low-fat diet and calcium and vitamin D supplementation is still ongoing.
In 1998, the Heart and Estrogen and Progestin Replacement Study randomized trial14 reported no protective effect of HT among women with a previous history of coronary disease and an excess risk of coronary disease among HT users during the first year of exposure. The findings of this trial and the early termination of the WHI HT trial have led experts to conclude that HT should be used only for short-term therapy for the relief of menopausal symptoms and for the prevention of bone fractures in selected women.15,16
Using national pharmacy sales data, Hersh et al17 estimated a 38% decline in HT and ET dispensing after the termination of the WHI HT trial. However, these data did not have woman-level information and, therefore, could not be used to estimate hormone discontinuation and initiation rates. Our current observational cohort study examined woman-level data from 5 health maintenance organizations in the United States to estimate overall and age-specific prevalence of HT and ET (proportion of women using hormones in the study population) and rates of discontinuation (rate of discontinuation of hormone use among hormone users) and initiation (rate of initiation of hormones among women who do not use hormones) in the 2 years before the published results of the WHI HT trial and for 5 months after their release.
MATERIALS AND METHODS
This study was conducted within the Cancer Research Network, a consortium of nonprofit health maintenance organizations. Our observational cohort study takes advantage of the Health Maintenance Organization's Research Network's Center for Education and Research on Therapeutics18 Patient Safety Cohort to examine patterns of HT and ET use before and after the release of the WHI HT trial results (July 2002). The Patient Safety Cohort consists of a random sample of approximately 200,000 men and women of any age who were enrolled at each of 10 participating health maintenance organizations. The study population was constructed to give equal weight to each health maintenance organization and to maintain confidentiality of the participating health maintenance organizations. Patients were eligible for inclusion if they had a pharmacy prescription benefit at any time during the study period (January 1, 1999, to June 20, 2001). This study was reviewed and approved by human subjects committees from each participating institution.
We extended the period of observation for the Patient Safety Cohort through December 31, 2002. From the original sample, we constructed a dynamic cohort of 169,586 women aged 40–80 years from 5 health maintenance organizations: Group Health Cooperative, Seattle, Washington; Harvard Pilgrim Health Care, Boston, Massachusetts; Fallon Community Health Plan, Worcester, Massachusetts; Kaiser Permanente Colorado, Denver, Colorado; and Health Partners, Minneapolis, Minnesota.
We used automated pharmacy dispensing data to estimate the prevalence of HT and ET and incidence of discontinuation and initiation. Two of the study researchers (K.M.N. and M.T.C.) reviewed a list of National Drug Codes of all oral or transdermal estrogen or progestin dispensings to identify hormones used in peri- and postmenopausal women. We did not include prescriptions for vaginal or compounded estrogens in these analyses.
Users of HT included women who received combination products or estrogen preparations with at least 1 dispensing for a progestin product. Users of ET included women who received estrogen preparations with no progestin dispensing. Progestin use was considered current for all estrogen prescriptions if there was any progestin-containing prescription dispensing during the study period, or if the prescription was for combination products.
We collected information on the highest daily conjugated equine estrogen dose dispensed in each month and organized these into the following categories: less than 0.625 mg, 0.625 mg, or more than 0.625 mg conjugated equine estrogen equivalents. We classified oral estradiol into conjugated equine estrogens as follows: less than 1.0 mg = less than 0.625 mg conjugated equine estrogen; 1.0 mg = 0.625 mg conjugated equine estrogen; and more than 1 mg = more than 0.625 mg conjugated equine estrogen.19,20 The relative equivalence was established for transdermal estradiol as follows: 0.05 mg = 0.625 mg conjugated equine estrogen; more than 0.05 mg = more than 0.625 mg conjugated equine estrogen; and less than 0.05 mg = less than 0.625 mg conjugated equine estrogen.19,20
The intended duration of each dispensing and, thus, the run-out date, was established using automated data on number of estrogen pills or patches dispensed. A woman was considered a user in any month she received at least one estrogen dispensing. We considered women to be continuous users as long as the hormone was dispensed within 60 days of the run-out date. A new run-out date was set with each successive dispensing rather than using cumulative number of pills from all dispensings.
We used automated membership data to track subjects’ enrollment in a health maintenance organization for each month. We considered women to be enrolled in the health plan until there was a lapse in membership of more than 2 months. We deleted (censored) women from the analyses at their date of disenrollment from the health maintenance organization; however, women rejoined the analysis if they reenrolled in a health maintenance organization during the study period.
We excluded the initial 7 months of data (at study start and/or after re-enrollment) for subjects to allow a sufficient run-in time for estimation of initiation rates. We examined plots of initiation and discontinuation over the course of time, and the trends were stable before June 2002 relative to the change after the release of WHI. Therefore, we pooled all the baseline data (September 1999 to June 2002) to increase the precision of the baseline estimates and generated mean monthly values. We examined ET and HT separately for all analyses.
We used log-binomial regression21 to model the incidence of initiation and discontinuation of HT and ET at baseline and for each month after June 2002, adjusting for age and health maintenance organization. We used a transition model,22 assuming a first-order Markov process, to account for the serial correlation within women that resulted from having repeated measurements on each woman. We modeled HT and ET use conditional on use/no use in the previous month, which is equivalent to modeling the incidence of initiation (use in 1 month given no use in the previous month) and discontinuation (no use in 1 month given use in the previous month). We report the relative risk and corresponding 95% confidence intervals comparing the incidence of initiation and discontinuation July 2002 through December 2002 to the baseline incidence. We also stratified the results by age group.
Age was calculated at each study month. Women contributed information for each age group they belonged to during the study period for the age-specific prevalence estimates. All analyses were conducted in SAS 8.2 (SAS Institute, Cary, NC).
There was a 12% decrease (n = 19,979) in the study population during the 39-month study period (approximately 512 per month) as the result of women aging out of the cohort, disenrollment from their health maintenance organization, or death (Table 1). Individual health maintenance organizations contributed 17.8–21.3% of the study population (data not shown). Women aged 40–49 years comprised 38% of the cohort at baseline and 34% at follow-up. We observed a small increase between baseline and follow-up in the proportion of women aged 50–59 years (29–31%) and aged 70–79 years (15–17%). The overall prevalence of hormone use was 27.2% at baseline, or 14.6% HT and 12.6% ET (Table 2). We found a 45.9% and a 27.8% reduction in the prevalence of HT and ET, respectively, after July 2002. Nearly all estrogen users received 0.625 mg of conjugated equine estrogen or its equivalent. Just more than 80% of HT and ET users took 0.625 mg conjugated equine estrogen or the equivalent at baseline; this decreased to 73.1% after July 2002. We observed a 43.7% and 18.9% decrease in the prevalence of 0.625 mg and more than 0.625 mg conjugated equine estrogen or its equivalent after July 2002, respectively, and a 5.8% increase in less than 0.625 mg use.
The discontinuation rates of HT increased from July 2002 to October 2002 and then stabilized or declined (Fig. 1; P < .001); this increase translated into a 5.52 higher risk of discontinuation (95% confidence interval 5.30–5.75) in October 2002 compared with baseline (P < .001; Table 3). The same trend was observed for the discontinuation of ET, although the rise in rates were not as marked (P = .001; Fig. 1). There was a significant decrease in initiation of HT and ET for all months compared with baseline (P < .001), but there was no clear trend by month. The decrease in initiation rates was greater among HT users than ET users.
The prevalence of HT was highest in women approximately 55 years of age and lowest at the extremes of younger and older age (Fig. 2). We saw a similar distribution of HT use by age during follow-up, but with a much lower prevalence across age groups. We noted a 27% decline in HT prevalence among women aged 40–44 years, a 40–50% decline in prevalence between women aged 45–65 years, and a 50–53% decline in prevalence among women aged 65–79 years.
The distribution of ET prevalence was lowest at the youngest ages and highest between women aged 59–66 years (Fig. 3). Unlike HT, the higher rate of ET use was sustained well into a woman's 70s before and after July 2002. There was less variation in the decline of ET prevalence by age than HT, with decreases ranging from 21% among women aged 65–69 years to 29% among women aged 50–54 years.
Less than 1% (0.38%) of women initiated HT per month at baseline, with the lowest proportion of initiation among women aged 40–44 years (0.09%) and the highest proportion among women aged 50–54 years (0.81%; rate data not shown). The proportion of women initiating HT was considerably lower in December 2002 (0.18% per month) than before July 2002 (P < .001), with the greatest proportion among women aged 55–59 years (0.41%; P < .001). The initiation of ET was lower than HT, with 0.28% of the cohort initiating per month before July 2002; this decreased to 0.23% after July 2002 (P < .001). Women aged 60–69 years were most likely to initiate ET before (0.44% per month) and after July 2002 (0.39%; P = .25).
Relative risks for the discontinuation of HT after July 2002 varied markedly by age; most discontinuation occurred in women aged 55–79 years, with a 5- to 6-fold increase in discontinuation rates compared with baseline rates (P < .001; Table 4). There was virtually no change in discontinuation in women aged 40–44 years (P = .90). We also observed an increase in ET discontinuation rates after July 2002 among women age 45 years or older (2- to 3-fold increase; P < .001). No linear trend in rates of ET (P = .75) or HT (P = .53) initiation by age was noted. However, we did observe twice the decline in HT initiation among the youngest and oldest women compared with women aged 55–59 years (P < .001).
Geographic diversity existed in the prevalence of HT and ET use in our cohort. The ranges of HT prevalence among health maintenance organizations were 10.2–17.1% at baseline and 4.0–10.5% at follow-up (data not shown). The prevalence of ET among health maintenance organizations ranged from 7.7% to 21.0% at baseline and 4.6% to 17% at follow-up. The same health maintenance organizations were at the same ends of the ranges for ET and HT at baseline and follow-up. We found a decrease in the prevalence of HT and ET of 46% (range 38.6–60.8%) and 28% (range 21.0–43.2%), respectively, by December 2002.
The evidence from the WHI HT trial has made it increasingly clear that the risks of using HT outweigh the benefits for women without menopausal symptoms. Our results showing an immediate reduction in HT and ET use after the termination of the WHI HT trial are supported by self-reported data from other cohorts23,24 and national pharmacy sales.17 Recent estimates of overall hormone cessation after July 2002 have ranged from 38%17 to 58%.24 There was a 37% overall decline in HT and ET use from baseline to December 2002 in our large, geographically diverse sample of women aged 40–80 years; HT prevalence declined 46% (14.6% to 7.9%) and ET declined 28% (12.6% to 9.1%). Our woman-level data most closely mirror estimates from national pharmaceutical sales data17 and reported rates of decline in Premarin (Wyeth, Philadelphia, PA) sales (Colata G. Rush to fill void in menopause-drug market. New York Times, September 1, 2002). The 2 studies reporting higher rates of discontinuation used a definition that likely overestimates true rates because they defined discontinuation as attempting to stop (stopping hormones for ≥ 1 week or reducing estrogen dosage).23,24
Other studies have found that additional information needs to be relayed to the lay public about the risks of continued HT use.23,24 In a recent study of 670 HT and ET users aged 50–69 years,23 nearly all women (93%) reported some awareness of the WHI HT trial findings, yet only 23% of women had correct knowledge of the findings. In that cohort, most of the HT users (69%) and some of the ET users (37%) attempted to stop using their hormones within the 6–8 months after July 2002.23 Among the women who attempted to stop, 74% were successful.25 We do not know whether improving knowledge of HT risks and benefits would further decrease the use of HT. Knowledge about HT was associated with successfully stopping hormones in bivariable models but was not identified as a strong predictor in multivariable models, most likely because having troublesome withdrawal symptoms was so strongly associated with women's ability to stop.25 A striking finding in our study and those cited previously is the high discontinuation rates among ET users despite the fact that the WHI ET trial was ongoing during the study period. Whether the discontinuation of ET reflects women and providers misunderstanding of the results of the trial, a generalization of the HT findings to ET, or another explanation entirely requires further investigation.
Hersh et al17 were able to link prescription sales with physician office visit information to examine indications for hormone use. Control of menopausal symptoms was indicated by 90% of the physicians who prescribed hormones after July 2002; disease prevention was noted by 5% of physicians (4% for osteoporosis and 1% for cardiovascular disease).17 We did not have information on indication for use, which might contribute to continued use (eg, osteoporosis) or discontinuation (eg, history of coronary artery disease) among our cohort. However, we know that menopausal symptoms are a major reason for initiating ET and HT24,25 and also are the strongest predictor of not being able to stop.25 Approximately 15% of postmenopausal women have symptoms 7 years after menopause.26 We saw a decrease in use of the standard conjugated equine estrogen dose (0.625 mg) among our cohort, a finding that also was supported from national pharmacy sales data.17 Convincing data suggest that less than 0.625 mg of conjugated equine estrogen is adequate for most women to help manage menopausal symptoms.27 A shift toward reducing conjugated equine estrogen dose is not supported by the WHI HT trial results because WHI only evaluated a single dose of HT. Therefore, understanding why women and clinicians are adopting lower doses requires additional exploration. Given the hint of a plateau in discontinuation rates by the end of 2002, it is possible that further reduction in HT and ET use will only occur when effective and safe alternative therapies for controlling menopausal symptoms are available.
Collaborative agreements prevent us from providing site-specific estimates. We observed a greater range in ET prevalence estimates between sites at baseline and follow-up compared with HT, even after adjustment for age. Some of the variation in estimates might be attributed to different prescribing practices by health maintenance organization or, alternatively, to geographic variation in the prevalence of hysterectomies. Importantly, the differences we noted in the prevalence of HT and ET by region did not directly mirror previously published patterns.1 Geographic variation may account for some of the differences in our results compared with findings from New Zealand24 and Northern California.23
There are several strengths of this study. We were able to use woman-level data from a large representative sample of health maintenance organization enrollees between 40 and 80 years of age from 5 geographic sites in the United States. The reported prevalence estimates may be slight overestimates of hormone use in the United States because health maintenance organization members largely have prescription drug benefits and are believed to be wealthier and better educated than the general U.S. population. However, our findings should be generalizable to the approximately 1 in 4 U.S. residents (26.4%) enrolled in health maintenance organizations in the United States,28 if not to the larger U.S. population. We were able to estimate prevalence from pharmacy dispensings rather than from self-reported data or from written prescriptions that may not have been filled. We also were able to censor the first 7 months of data to ensure we had adequate information to estimate true incidence rates for initiation.
Despite the strengths of administrative pharmacy data, it is possible we have missed some HT and ET users. We were unable to capture hormone use in our data set if women did not fill their prescriptions at their health maintenance organization's pharmacies. However, at 1 of the participating health maintenance organizations, 95% of women aged 60 years or older reported filling all or almost all their prescriptions at their health maintenance organization's pharmacies.29 Women also may have switched from oral estrogens to vaginal or compound estrogens, which were not included in these analyses.
We are confident that women who received multiple dispensings of estrogen were using their prescription drugs. However, women with hormone dispensings may not necessarily have used the complete amount; therefore, estimates of exposure may be higher than actual use. Our previous work has found that some women fill hormone prescriptions erratically—it is unknown whether erratic use is to contain cost, decrease dose, or for some other reason. The best way to minimize misclassification of current hormone use is to allow a lag-time beyond the prescription run-out.30 We used a 60-day lag-time to ensure we captured users who were not taking their hormones exactly as prescribed. Our lag-time effectively captured women who may have been using different methods for tapering their medication. We may have underestimated true discontinuation rates during the first 2–3 months after July 2002 given the variability in actual dosing. Despite this, we still saw a rapid decline in use by study month with a peak in October 2002, only 2.5 months after the termination of the WHI HT trial.
More detailed articles have been published since the initial WHI HT trial findings.12,31–35 There may be an additional impact on use resulting from the release of the black box warnings on Prempro (Wyeth, Philadelphia, PA)36 and Premarin (Wyeth)37 in January 2003. Whether the subsequent WHI reports and new black-box labels have further contributed to declines in HT and ET use requires additional analyses beyond the scope of our data collection.
The diffusion of the WHI HT trial results had an immediate impact on HT and ET discontinuation and is likely responsible for the 46% and 28% decline, respectively, in initiation of these therapies. However, substantial HT use still exists among women who are likely not using HT for short-term therapy for the relief of menopausal symptoms. Long-term HT use has important public health consequences. There are an estimated 57.6 million women aged 40–80 years in the United States.38 Applying age-specific HT prevalence estimates from our cohort reveals that as many as 4.3 million women aged 40–80 years in the United States may still be using HT. If we apply our HT prevalence estimates to observed rates from the WHI HT trial,11 we could expect 8,167 additional adverse events per year attributable to HT use given the risks (CHD; stroke; pulmonary emboli; hip fracture; breast or endometrial cancer; or deaths) and benefits (colorectal cancer) of HT. Although admittedly crude, these estimates give a flavor for the potential magnitude of the effect of continued HT use in the United States. Further exploration of why women continue to use HT and identification of alternative methods for addressing those reasons for continued use are indicated.
1. Keating NL, Cleary PD, Rossi AS, Zaslavsky AM, Ayanian JZ. Use of hormone replacement therapy by postmenopausal women in the United States. Ann Intern Med 1999;130:545–53.
2. Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med 1991;20:47–63.
3. Grady D, Rubin SM, Petitti DB, Fox CS, Black D, Ettinger B, et al. Hormone therapy to prevent disease and prolong life in postmenopausal women. Ann Intern Med 1992;117:1016–37.
4. Rijpkema AH, van der Sanden AA, Ruijs AH. Effects of post-menopausal oestrogen-progestogen replacement therapy on serum lipids and lipoproteins: a review. Maturitas 1990;12:259–85.
5. Adams MR, Kaplan JR, Manuck SB, Koritnik DR, Parks JS, Wolfe MS, et al. Inhibition of coronary artery atherosclerosis by 17-beta estradiol in ovariectomized monkeys: lack of an effect of added progesterone. Arteriosclerosis 1990;10:1051–7.
6. Weiss NS, Ure CL, Ballard JH, Williams AR, Daling JR. Decreased risk of fractures of the hip and lower forearm with postmenopausal use of estrogen. N Engl J Med 1980;303:1195–8.
7. Genant HK, Baylink DJ, Gallagher JC, Harris ST, Steiger P, Herber M. Effect of estrone sulfate on postmenopausal bone loss. Obstet Gynecol 1990;76:579–84.
8. Kampman E, Potter JD, Slattery ML, Caan BJ, Edwards S. Hormone replacement therapy, reproductive history, and colon cancer: a multicenter, case-control study in the United States. Cancer Causes Control 1997;8:146–58.
9. Newcomb PA, Storer BE. Postmenopausal hormone use and risk of large-bowel cancer [published erratum appears in J Natl Cancer Inst 1995;87:1416]. J Natl Cancer Inst 1995;87:1067–71.
10. Design of the Women's Health Initiative clinical trial and observational study. The Women's Health Initiative Study Group. Control Clin Trials 1998;19:61–109.
11. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321–33.
12. Cauley JA, Robbins J, Chen Z, Cummings SR, Jackson RD, LaCroix AZ, et al. Effects of estrogen plus progestin on risk of fracture and bone mineral density: the Women's Health Initiative randomized trial. JAMA 2003;290:1729–38.
13. Anderson GL, Limacher M, Assaf AR, Bassford T, Beresford SA, Black H, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. JAMA 2004;291:1701–12.
14. Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. JAMA 1998;280:605–13.
15. Fletcher SW, Colditz GA. Failure of estrogen plus progestin therapy for prevention. JAMA 2002;288:366–8.
16. Nelson HD, Humphrey LL, Nygren P, Teutsch SM, Allan JD. Postmenopausal hormone replacement therapy: scientific review. JAMA 2002;288:872–81.
17. Hersh AL, Stefanick ML, Stafford RS. National use of postmenopausal hormone therapy: annual trends and response to recent evidence. JAMA 2004;291:47–53.
18. Platt R, Davis R, Finkelstein J, Go AS, Gurwitz JH, Roblin D, et al. Multicenter epidemiologic and health services research on therapeutics in the HMO Research Network Center for Education and Research on Therapeutics. Pharmacoepidemiol Drug Saf 2001;10:373–7.
19. Menopause core curriculum study guide. 2nd ed. Cleveland (OH): North American Menopause Society; 2002. p. 126.
20. A new conjugated estrogen. Med Lett Drugs Ther 1999;41:67–8.
21. Wacholder S. Binomial regression in GLIM: estimating risk ratios and risk differences. Am J Epidemiol 1986;123:174–84.
22. Diggle PJ, Heagerty P, Liang KY, Zeger SL. Analysis of longitudinal data. 2nd ed. New York (NY): Oxford University Press; 2002.
23. Ettinger B, Grady D, Tosteson AN, Pressman A, Macer JL. Effect of the Women's Health Initiative on women's decisions to discontinue postmenopausal hormone therapy. Obstet Gynecol 2003;102:1225–32.
24. Lawton B, Rose S, McLeod D, Dowell A. Changes in use of hormone replacement therapy after the report from the Women's Health Initiative: cross sectional survey of users. Br Med J 2003;327:845–6.
25. Grady D, Ettinger B, Tosteson AN, Pressman A, Macer JL. Predictors of difficulty when discontinuing postmenopausal hormone therapy. Obstet Gynecol 2003;102:1233–9.
26. Avis NE, McKinlay SM. The Massachusetts Women's Health Study: an epidemiologic investigation of the menopause. J Am Med Womens Assoc 1995;50:45–9, 63.
27. Utian WH, Shoupe D, Bachmann G, Pinkerton JV, Pickar JH. Relief of vasomotor symptoms and vaginal atrophy with lower doses of conjugated equine estrogens and medroxyprogesterone acetate. Fertil Steril 2001;75:1065–79.
28. Centers for Disease Control and Prevention. Health maintenance organizations (HMOs) and enrollment, according to model type, geographic region, and Federal program: United States, selected years 1976-2002. Available at: www.cdc.gov/nchs/data/hus/tables/2003/03hus132.pdf
. Retrieved September 9, 2004.
29. Buist DSM, LaCroix AZ, Brenneman S, Abbott T. A population-based osteoporosis screening program: who does not respond and what are the consequences? J Am Geriatr Soc 2004;52:1130–7.
30. Newton KM, LaCroix AZ. The association between body mass index and reinfarction and survival after first myocardial infarction in women. J Women's Health 1996;5:433–44.
31. Wassertheil-Smoller S, Hendrix SL, Limacher M, Heiss G, Kooperberg C, Baird A, et al. Effect of estrogen plus progestin on stroke in postmenopausal women: the Women's Health Initiative: a randomized trial. JAMA 2003;289:2673–84.
32. Chlebowski RT, Hendrix SL, Langer RD, Stefanick ML, Gass M, Lane D, et al. Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women: the Women's Health Initiative Randomized Trial. JAMA 2003;289:3243–53.
33. Hays J, Ockene JK, Brunner RL, Kotchen JM, Manson JE, Patterson RE, et al. Effects of estrogen plus progestin on health-related quality of life. N Engl J Med 2003;348:1839–54.
34. Shumaker SA, Legault C, Rapp SR, Thal L, Wallace RB, Ockene JK, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2651–62.
© 2004 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
35. Rapp SR, Espeland MA, Shumaker SA, Henderson VW, Brunner RL, Manson JE, et al. Effect of estrogen plus progestin on global cognitive function in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2663–72.