In the mid-1990s to late 1990s, it was estimated that more than one third of postmenopausal women aged 50 to 74 years were taking hormone therapy (HT).1 Until early 2002, HT was prescribed as both a short- (management of menopausal symptoms) and long-term therapy (prevention of osteoporosis and coronary heart disease [CHD]).2 The initial release of the results from the Women's Health Initiative (WHI) trials in July 2002 raised concerns regarding the safety of HT, specifically estrogen in combination with progestin (EPT).3 The results suggested that women who received EPT had an increased risk of stroke, CHD, pulmonary embolism, deep-vein thrombosis, and breast cancer. The results of the estrogen (ET) only arm of the WHI, released in 2004, also indicated increased risk of stroke and deep-vein thrombosis but not CHD or breast cancer.4
Although these trials were not originally designed to investigate the effects of HT in treating menopausal symptoms, the findings had significant implications on the clinical management of menopause. In addition to the Food and Drug Administration, medical societies and healthcare organizations released revised menopause treatment guidelines recommending the use of HT at the lowest effective dose for the shortest duration.5-8 Numerous reports showed that in the months after the initial release of results, HT utilization dramatically decreased worldwide.2,9-13 The overwhelming media attention14 may have affected the decision of many women to discontinue HT use, dissuaded others from initiating HT, and influenced the prescribing habits of physi-cians.2,15,16 Regardless of cause, in the United States, HT utilization decreased approximately 18% within the first 3 months after the release of the WHI trial results.9 By July 2003, 12 months after the initial WHI publication, HT use had decreased between 32% and 38%,10,17 and by 18 months post-WHI, HT use was nearly 45% below mid-2002 levels.18
Evolving data affecting the benefit-risk ratio of HT,19 as well as the continued refinement of menopause treatment guidelines20,21 and the addition of new therapies to the armamentarium of HT options,22 may have further altered recent HT utilization. Although data have begun to emerge on physicians’ prescribing patterns and opinions of HT post-WHI, no recent long-term evaluation of HT prescription trends in the United States has been published. The objectives of these analyses were to assess trends in HT utilization rates by route of administration, dose strength, and prescriber specialty during the 7 years after the initial release of the WHI results in July 2002. We hypothesized a decrease in standard oral formulations and an increase in the prescribing of low-dose and transdermal formulations.
This study used the administrative database of MedImpact Healthcare Systems Inc., which provides pharmacy benefits management (PBM) services to managed care organizations, employer corporations, unions, health plans, insurance carriers, and third-party administrators, as well as local, state, and federal employee programs. This database contains information on all prescription claims for covered members. Med-Impact's researchable database consisted of 7.4 million people in 2002 and had increased to 8.9 million individuals by 2009. The vast majority of covered individuals were enrolled in commercial health maintenance organizations (HMOs), with only 1% to 2% of the total covered individuals enrolled in Medicare. All geographical regions were represented, with a roughly equal proportion of men and women. To adjust for any fluctuations in the number of covered lives in the PBM database, the number of claims per 10,000 members was reported. To determine the number of HT claims per 10,000 members, the number of claims for HT users (defined as women with ≥1 prescription HT claims) was divided by the total number of PBM members multiplied by 10,000. It should be noted that the information available for this analysis did not include individual patient disposition (eg, age or uterine status) or categorize HT usage by prescription indication.
Prescription claims for the following estrogenic agents were included in these analyses: estradiol, estradiol acetate, synthetic conjugated estrogens, conjugated estrogens, esterified estrogens, estropipate, ethinyl estradiol, estradiol/drospirenone, estradiol/norethindrone acetate, estradiol/norgestimate, conjugated estrogens/medroxyprogesterone acetate, esterified estrogens/methyl-testosterone, conjugated estrogens/methyl-testosterone and ethinyl estradiol/norethindrone acetate. The estrogen component of any coformulated medications was considered as the target hormone of interest for these analyses, and HT claims were not differentiated between EPT and ET. Prescription claims and patient volume was assessed to determine the average number of HT claims per woman on HT per quarter. The average length of supply for each HT claim, based on the days’ supply per claim per quarter was also assessed.
Claim volume by quarter was stratified by route of administration, physician specialty, and dose category. Route of administration categories included in these analyses were oral and transdermal, where transdermal included patches, gels, a spray, and a topical emulsion; other routes of administration (eg, vaginal) and estrogen-containing oral contraceptives were excluded.
Physician specialty was determined by each prescriber's first- or second-listed specialty, using Health Market Science Prescriber MasterFile data (accessed 2009). Physicians were classified as obstetrician-gynecologist (Ob/Gyn) if the specialty was listed first or second; all other physicians were designated as other provider (Other).
Dose categories were determined as low or standard according to The North American Menopause Society classification6; low-dose was defined as less than the equivalent of 0.625 mg conjugated equine estrogens. Standard/high-dose was defined as all doses greater than the low-dose criteria. Conjugated equine estrogens equivalency was not determined for the following estrogens: estradiol acetate, estropipate, ethinyl estradiol, and ethinyl estradiol/norethindrone acetate. These estrogens were designated as “other” and were not included in the low- versus standard/high-dose comparisons.
The number of included claims and corresponding patient volume were calculated by quarter (1Q-4Q) from 1Q 2001 to 2Q 2009 from all of the MedImpact eligible members. Because of the addition of new plans as well as plan attrition during the study period, the number of claims per 10,000 members per quarter was reported to normalize for fluctuations in covered lives during the study. For each of the stratified categories, the percentage of total claims was determined using the following equation: Proportion of claims by category = (no. claims per category in NQ 200N/total claims in NQ 200N) × 100.
Overall changes in HT claims
Immediately after the publication of the WHI results in mid-2002, there was a precipitous decrease in the prescription claim volume of all HT among MedImpact's membership (Fig. 1). Just before publication of the WHI results (2Q 2002), the overall volume of HT claims was 770 claims per 10,000 members; however, in the subsequent months (3Q 2002), claims decreased by 15% to 653 claims per 10,000 members. By the end of 2002, the claims were reduced by nearly 30% (551 claims per 10,000 members) from 2Q 2002. Although the decline in HT claim volume was not as dramatic in the subsequent years, this trend continued until 2007 when only 227 HT claims per 10,000 members were reported, representing an approximate 70% drop in HT claims from 2Q 2002. During 2007 and continuing through 2009, HT claims remained constant, ranging between 200 and 230 claims per 10,000 members, a 70% to 73% decrease from 2Q 2002. We identified a 15% decrease in the ratio of claims per HT user per quarter between 2002 and 2009 (data not shown). Counter to our initial assumption that each HT user would have a single HT claim per quarter, in actuality, the average ratio was approximately two HT claims per HT user per quarter in the earlier years, with a slow evolution to approximately 1.7 claims per HT user per quarter in later years.
Changes in HT claims by specialty
In all years studied, Ob/Gyn physicians were the highest individual specialty group prescribing HT (data not shown). In the months before the WHI release, Ob/Gyn-prescribed HT claims accounted for 250 claims per 10,000 members—a total of 32% of total HT claims (Table 1). Even in the reduced HT market of 2009, Ob/Gyn continued to account for 32% of all HT claims.
The percentage change in HT claims prescribed by either Ob/Gyn physicians or other HT providers between the years of 2002 and 2009 were largely parallel. However, for each individual quarter during most of this timeframe, the percentage decrease in HT claims from 2Q 2002 prescribed by Ob/Gyn was approximately 5% to 10% less than the percentage decrease prescribed by other physicians (Fig. 2).
Changes in HT claims by route of administration
The negative impact of the WHI on prescription claims for transdermal HT was not as considerable as it was on claims for oral HT, as shown in Fig. 3. Immediately after the release of the WHI, the number of oral HT claims decreased 16% from 2Q 2002. This steep decline in oral HT claims continued for nearly 2 years post-WHI until 2Q 2004, when the claim rate had decreased by more than half. Although claims for oral HT slowed in decline, 7 years after the release of the initial WHI trial results, claims were reduced approximately 77% from 2Q 2002. This trend was not observed with transdermal HT; by 3Q 2002, only a 6% reduction was observed, and by 2Q 2009, only a 25% reduction was observed.
Throughout the observation period, oral delivery remained the most common route of administering HT; however, the use of transdermal HT steadily increased as a percentage of overall HT claims. During each year since 2002, both Ob/Gyn and other physicians have proportionally prescribed more transdermal HT formulations. In 2002, 12% of all HT claims initiated by Ob/Gyn physicians were for transdermal preparations compared with 6% of all HT claims by other physician specialties (Table 1). By 2009, these percentages increased to 29% and 17% for Ob/Gyn and others, respectively. Of note, Ob/Gyn providers accounted for most transdermal HT claims in all years considered.
Changes in HT claims by dose
Most (94%) of the HT prescriptions in 2002 were considered standard or high-dose HT. Claim volume for standard/high-dose HT (all routes of administration) was 726 claims per 10,000 members in 2Q 2002. In time, the use of standard/high-dose HT declined to 156 claims per 10,000 members (Fig. 4), close to one fifth of the 2Q 2002 rate. Conversely, low-dose HT increased from approximately 6% in 2002 to nearly 25% of all HT claims in 2009. Therefore, the proportional growth of low-dose HT was at the expense of standard/high doses. Furthermore, the prescription trends of low-dose HT were notably different from those of standard/high HT; low-dose HT claims in 2Q 2002 were 44 claims per 10,000 members; in 3Q 2002, a total of 41 claims per 10,000 members were identified; therefore, unlike standard/high-dose HT, no abrupt drop in claim volume was observed immediately after the release of the WHI results. Furthermore, prescription claims for low-dose HT remained relatively consistent in time, with only a small increase in claims from mid-2004 to early 2005. Furthermore, we analyzed the reduction in EPT versus ET alone between 2Q 2002 and 2Q 2006 and found similar percentage reductions in HT claims (data not shown); as such, we chose not to analyze separately these forms of HT. To properly analyze the claims rate for EPT and ET alone, individual patient data would be necessary, which were not available for analysis.
Changes in HT claims by both dose and route of administration
As detailed in Table 1, standard/high doses of oral HT formulations remain the most commonly prescribed HT therapy. Overall, in 2002, standard/high oral HT accounted for almost 9 of 10 claims, but 7 years later, standard/high oral HT claims accounted for less than two in three HT claims, and this decrease was not specific to physician specialty. Between the years of 2002 and 2009, as a percentage of total HT claims, overall standard/high-dose transdermal claims increased nearly threefold, with a relative increase similar between Ob/Gyn and other physicians. Furthermore, HT claims for low-dose preparations (both oral and transdermal) have shown the largest relative increases in all HT formulations since 2002. Although low-dose oral HT claims have increased fourfold as a proportion of total HT claims, prescription claims for low-dose transdermal formulations have increased more than 10-fold.
The results from these analyses reveal a major shift in prescription claims of HT that began after the publication of the initial WHI results in mid-2002. In the following months, HT claims dropped drastically, and by the end of 2002, HT claims had decreased by nearly one third. In the next 5 years, HT claims continued to decline, albeit more slowly, with claims in the subsequent years stabilizing at about one quarter of the 2002 claim volume. With this reduced market, claims for both low-dose and transdermal formulations have proportionally increased since 2002. The largest proportional increase in claims was for low-dose transdermal HT, yet by 2009, these transdermal formulations accounted for less than 5% of all HT claims. Furthermore, the trends in HT prescription claims did not differ between Ob/Gyn and other physician specialties. These data are consistent with other short-term reports from pharmacy claims that describe the decrease in HT use in the months after the initial release of the WHI results.2,9-11,17,23 During the last 5 months of 2002, HMO claims for EPT were reduced by almost 50%, and ET claims decreased by nearly 30%.2 Using national pharmacy sales data, Hersh et al10 estimated a 38% overall decline in HT within the year after the publication of the WHI trial. Similarly, Wysowski and Governale17 reported that 1 year after the publication of the WHI results, prescriptions of oral ET and EPT had declined about 32% from their peak.
Our data demonstrated similar reductions in HT claims between Ob/Gyn and other physician specialties, which is in contrast with data from a small, retrospective study of prescription data from the United States Military healthcare system in the state of Hawaii between July 1999 and July 2005.23 The authors concluded that HT prescription claims rates were influenced by physician specialty; during the 3 years after publication of the WHI results, primary care physicians showed the greatest decrease in HT prescription claims rates (77%) compared with an approximate 57% decrease for Ob/Gyn and 40% decrease for physicians in all surgical specialties. We speculate that study design may have contributed to the difference in results between studies. The military study was considerably smaller, with a total of approximately 71,000 HT prescriptions during the 6-year study, with most prescriptions (approximately 48,000) written before July 2002. In our study, the number of included HT prescription claims in the second quarter of 2002 alone was more than 7.7 million, with almost 2.5 million prescription claims from Ob/ Gyn prescribers. Furthermore, the results of the military claims database may not be an adequate representation of overall US prescribing habits because the prescriptions were written only by physicians at military facilities in the state of Hawaii and the period of data collection post-WHI in the Army study was nearly half (3 y) the post-WHI timeframe as reported here (7 y).
Previous reports of HT use after the release of the WHI results indicated that neither low-dose (<0.625 mg) nor transdermal HT formulations declined as much as standard/high-dose HT formulations.2,10,18 Buist et al2 showed that low-dose use increased 5.8% between 1999 and 2003.2 Similarly, Hersh et al10 reported a 6% increase in low-dose HT during the second half of 2002; however, this increase was not sustained in 2003.10 Our study finds that, as a proportion of total HT claims, low-dose oral formulations increased nearly fourfold between the years of 2002 and 2009, whereas the use of low-dose transdermal HT increased more than 10-fold. We ascribe the substantial time-related shifts toward lower doses and transdermal preparations partly to emerging clinical data and revised treatment guidelines. Randomized clinical trials and meta-analyses have suggested that transdermal HT may have less of an impact than oral HT on serum triglycerides, lipoprotein profiles, and gallbladder disease and may improve the benefit-to-risk ratio of HT, especially with respect to incidence of venous thromboembolism.24-29 In fact, the American Association of Clinical Endocrinologists has included a preference for transdermal HT in women with hypertension and/or hypertriglyceridemia or increased risk of cholelithiasis in their clinical guidelines.30 In addition, in the intervening years between 2002 and 2009, launches and marketing of newer HT formulations22 could have increased both provider and patient awareness of these potentially safer and more tolerable low-dose alternatives.31 Although our data support a continuing growth trend for use of low-dose and transdermal formulations, standard/high-dose oral ET remains the most common menopause therapy in the United States.
MedImpact Healthcare Systems Inc. serves approximately 36 million lives nationwide, with full pharmacy benefits to nearly 9 million individuals. The data obtained from this large database offer a robust measure of the evolution in HT prescribing trends with the capability to determine changes within both a short and long time frame. Using these claims, however, limited our ability to investigate the patient's disposition or to categorize HT usage by prescription indication. MedImpact Healthcare System's PBM membership is composed primarily of commercial HMOs and Medicaid, with only a small percentage (1%-2%) of Medicare members. As such, age and sex were relatively constant during the timeframe of our observation and would not impact the outcome of these analyses. Furthermore, we opted to combine all estrogen-containing formulations after observing similar prescription trends for EPT and ET. Individual patient data would also have been necessary to properly analyze the difference in EPT versus ET alone. The similar changes in prescription claims rates for EPT and ET alone are in contrast with short-term data reported by Hersh et al10 who found that prescription claims for ET decreased by one third and that prescriptions for EPT decreased by two thirds.10 An additional limitation of our analyses of claims data is the inaccurate representation of the numbers of women using HT. Counter to our initial assumption that the length of supply would be constant in time, the length of supply has slightly but progressively increased from 2002 to 2009. As such, the decrease in average quarterly refills may have led to a slight over-estimation of the reduction in claims volume.
These data confirm that the short-term studies conducted immediately post-WHI, which evaluated HT trends between 2002 and 2003, correctly predicted subsequent trends in overall HT prescription claims; however, these studies did not accurately predict the increase in low dose or transdermal HT in recent years. However, even with these increases, the overall HT market has not recovered from the release of WHI trial results. Some interesting and important prescribing trends were identified for lower dose and transdermal formulations between the years of 2002 and 2009. Although these trends have been fueled by new guidelines, emerging research, changing expert opinion, and increased availability of new formulations, by 2009, only one in four prescriptions was for low-dose therapy. Together, these claims data suggest that further clinician education may be necessary to increase awareness of low-dose and transdermal HT options.
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
2. Buist DS, Newton KM, Miglioretti DL, et al. Hormone therapy
prescribing patterns in the United States. Obstet Gynecol
3. Rossouw JE, Anderson GL, Prentice RL, 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
4. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative
randomized controlled trial. JAMA
5. American College of Obstetricians and Gynecologists. Vasomotor symptoms. Obstet Gynecol
6. The North American Menopause Society. Treatment of menopause-associated vasomotor symptoms: position statement of The North American Menopause Society. Menopause
7. Naftolin F, Schneider HP, Sturdee DW. Guidelines for hormone treatment of women in the menopausal transition and beyond. Maturitas
8. Hartsfield CL, Connelly MT, Newton KM, Andrade SE, Wei F, Buist DS. Health system responses to the Women's Health Initiative
findings on estrogen and progestin: organizational response. J Natl Cancer Inst Monogr
9. Haas JS, Kaplan CP, Gerstenberger EP, Kerlikowske K. Changes in the use of postmenopausal hormone therapy
after the publication of clinical trial results. Ann Intern Med
10. Hersh AL, Stefanick ML, Stafford RS. National use of postmenopausal hormone therapy
: annual trends and response to recent evidence. JAMA
11. Hillman JJ, Zuckerman IH, Lee E. The impact of the Women's Health Initiative
on hormone replacement therapy in a Medicaid program. J Womens Health
12. Mishra G, Kok H, Ecob R, Cooper R, Hardy R, Kuh D. Cessation of hormone replacement therapy after reports of adverse findings from randomized controlled trials: evidence from a British Birth Cohort. Am J Public Health
13. Newton KM, Buist DS, Miglioretti DL, et al. The impact of comorbidities on hormone use. After the 2000 release of the Women's Health Initiative
. J Gen Intern Med
14. Haas JS, Geller B, Miglioretti DL, et al. Changes in newspaper coverage about hormone therapy
with the release of new medical evidence. J Gen Intern Med
15. Ettinger B, Grady D, Tosteson AN, Pressman A, Macer JL. Effect of the Women's Health Initiative
on women's decisions to discontinue post-menopausal hormone therapy
. Obstet Gynecol
16. Wei F, Miglioretti DL, Connelly MT, et al. Changes in women's use of hormones after the Women's Health Initiative
estrogen and progestin trial by race, education, and income. J Natl Cancer Inst Monogr
17. Wysowski DK, Governale LA. Use of menopausal hormones in the United States, 1992 through June, 2003. Pharmacoepidemiol Drug Saf
18. Majumdar SR, Almasi EA, Stafford RS. Promotion and prescribing of hormone therapy
after report of harm by the Women's Health Initiative
19. Taylor HS, Manson JE. Update in hormone therapy
use in menopause. J Clin Endocrinol Metab
20. US Preventive Services Task Force. Hormone therapy
for the prevention of chronic conditions in postmenopausal women: recommendations from the U.S. Preventive Services Task Force. Ann Intern Med
21. The North American Menopause Society. Estrogen and progestogen use in peri- and postmenopausal women: March 2007 position statement of The North American Menopause Society. Menopause
22. Carroll N. A review of transdermal nonpatch estrogen therapy for the management of menopausal symptoms. J Womens Health (Larchmt)
23. Parente L, Uyehara C, Larsen W, Whitcomb B, Farley J. Long-term impact of the Women's Health Initiative
on HRT. Arch Gynecol Obstet
24. Spencer C, Crook D, Ross D, Cooper A, Whitehead M, Stevenson J. A randomised comparison of the effects of oral versus transdermal 17β-oestradiol, each combined with sequential oral norethisterone acetate, on serum lipoprotein levels. Br J Obstet Gynaecol
25. Sendag F, Karadadas N, Ozsener S, Bilgin O. Effects of sequential combined transdermal and oral hormone replacement therapies on serum lipid and lipoproteins in postmenopausal women. Arch Gynecol Obstet
26. Ouyang P, Michos ED, Karas RH. Hormone replacement therapy and the cardiovascular system lessons learned and unanswered questions. J Am Coll Cardiol
27. Canonico M, Plu-Bureau G, Lowe GD, Scarabin PY. Hormone replacement therapy and risk of venous thromboembolism in postmenopausal women: systematic review and meta-analysis. BMJ
28. Liu B, Beral V, Balkwill A, Green J, Sweetland S, Reeves G. Gallbladder disease and use of transdermal versus oral hormone replacement therapy in postmenopausal women: prospective cohort study. BMJ
29. Cirillo DJ, Wallace RB, Rodabough RJ, et al. Effect of estrogen therapy on gallbladder disease. JAMA
30. AACE Menopause Guidelines Revisions Task Force. American Association of Clinical Endocrinologists medical guidelines for clinical practice for the diagnosis and treatment of menopause. Endocr Pract
31. Ettinger B. Vasomotor symptom relief versus unwanted effects: role of estrogen dosage. Am J Med