Compounded Bioidentical Hormone Therapy: Identifying Use Trends and Knowledge Gaps Among US Women

JoAnn V. Pinkerton, MD; Nanette Santoro, MD


Menopause. 2015;22(9):926-936. 

In This Article


Our analyses and calculations conservatively estimate that between 1 million and 2.5 million US women aged 40 years or older are currently using CHT, accounting for approximately 28% to 68% of HT prescriptions filled annually and costing US$1 billion to US$2 billion each year. We believe that this is the first report in the literature to estimate the number of US women using CHT and the amount spent on CHT annually in the United States.

To support our extrapolated findings on CHT use, we searched the literature for recent studies reporting HT use among US women that might corroborate the 5% rate of current HT use in the Rose survey. Sprague et al[4] reported that 4.7%

(95% CI, 3.3-6.1) of US women aged 40 years or older who participated in the National Health and Nutrition Examination Survey (NHANES; n=10,107) used oral HT (estrogen or progestin alone or combined) in 2009-2010. Steinkellner et al[5] evaluated the 2009 pharmacy claims for approximately 10 million insured women aged 50 years or older and similarly estimated current use of oral estrogen-based or progestogenbased HT at 4.8%. The rate of current use climbed to 8.8% when all HT delivery methods were considered.[5]


Jewett et al[8] interpolated data from NHANES to estimate that 1.8% of US women aged 40 to 84 years used combined oral estrogen-progestin therapy in 2010. They did not estimate the prevalence of oral estrogen-only or nonoral formulations.[8] However, Sprague et al[4] reported that 2.9% of women aged 40 years or older in NHANES used oral estrogen-only formulations in 2009-2010, and Steinkellner et al[5] stated that 4.3% of women aged 50 years or older used nonoral HT in 2009. Taken together, the findings of these three studies—all of which applied a more limited definition of HT and two of which targeted a narrower age range than the Rose survey— suggest that the 5% rate of current HT use observed in the Rose survey may underestimate the extent of HT use in the United States.[4,5,8]

The Source Healthcare Analytics PHAST 2.0 prescription data report, which showed that 36 million annual prescriptions for FDA-approved HT were prescribed in 2012, agrees closely with a 2011 report by Stagnitti and Lefkowitz,[6] which showed that adult US women filled 32 million prescriptions for FDA-approved HT in 2008. In addition, the average US$49 cost per product from the Rose survey that we used in calculating the prevalence of CHT use seems reasonably consistent with real-world pricing for these agents.[19,22,23] Although the mean use of HT exceeded 1 year in the Rose and Harris surveys, we assumed minimal use of 9 months and maximal use of 12 months in estimating the annual prevalence of CHT to ensure that the range accounted for those women who are not fully adherent. According to a Mayo Clinic study, 54% of CHT users reported that they had used CHT longer than 1 year, suggesting that our estimate of 9 to 12 months of use is reasonable.[19]

The Rose and Harris surveys showed that at least three quarters of women experienced symptoms during the menopausal transition, with symptom onset in both surveys at about age 47 years. More than 91% of perimenopausal and postmenopausal women in the Harris poll had a history of VMS, which 67% self-judged as moderate to severe. These findings are consistent with other representative population-based studies of VMS during the menopausal transition, such as the Study of Women's Health Across the Nation, Melbourne Women's Midlife Health Project, and Penn Ovarian Aging Study, which reported VMS rates ranging from 40% to 97%.[24,25,26]

HT is indicated for and considered the most effective treatment of menopause-related moderate to severe VMS.[1,27] It is also indicated for moderate to severe vulvovaginal atrophy that is unresponsive to vaginal moisturizers or lubricants.[27] In relation to the high prevalence of symptoms in the Rose and Harris surveys, the rate of HT use in the surveys was low, consistent with the literature. For example, only 123 Harris completers had ever used HT, implying that more than half of women who developed moderate to severe hot flashes during the menopausal transition may not have received effective treatment.

Completers of the Harris and Rose surveys were asked about their experience with specific types of HT. The pool of responses revealed that some prescribing of HT is inconsistent with evidence-based practices.[1,9,27,28] In the Rose survey, 25% of products taken by current HT users were hormone formulations that the FDA has not approved to treat menopausal symptoms, including testosterone-based regimens, hormone pellets, and vaginal and injected progestogens (alone or with estrogen). Among Rose completers who stated that they had undergone hysterectomy, 95 were currently using a progestogen-containing regimen. Furthermore, several current HT users in the Rose survey stated that they had undergone salivary testing to measure levels of sex steroid hormones. Despite the lack of evidence and national recommendations against using salivary testing to titrate HT,[9,12,29,30] its use has increased in recent years, concurrent with the rising prevalence of CHT use.[11]

Our data support the general consensus that the CHT market is growing.[9,11,18] Conversely, the market for FDA-approved HT shrank dramatically in the past decade or so.[4,5,6,31] Some women who are experiencing menopausal symptoms may choose CHT over FDA-approved HT because of Internet-perpetrated myths that compounded hormones are more natural and thus safer than commercially manufactured ones.[32] Findings from the Harris survey suggest that few women understand that CHT products are not FDA-approved, and the results of the Rose survey indicate that many women are unsure whether the HT they are taking is compounded. Other surveys have revealed that postmenopausal women have inaccurate information on CHT.[19,33,34,35,36,37] The Mayo Clinic study by Iftikhar et al[19] reported that 46% of women surveyed were familiar with CHT, and 67% of these women considered CHT safer than FDA-approved HT. Sayakhot et al[36] administered a questionnaire on menopause to 114 Australian women aged 40 to 51 years who had breast cancer, which showed that 79% were confused about the risks and benefits of CHT. Pharmacy compounders in the United States have never been required to label CHT or to provide detailed prescribing information,[38] which may contribute to US women's general lack of accurate knowledge of these products.

Most recognized women's health organizations in the United States, including the American Association of Clinical Endocrinologists, the American College of Obstetricians and Gynecologists, the American Society for Reproductive Medicine, The Endocrine Society, The North American Menopause Society, the US Preventive Services Task Force, and the Women's Health Practice and Research Network of the American College of Clinical Pharmacy, have position statements on HT that warn about the lack of scientific evidence supporting the efficacy and safety of CHT.[1,9,28,39,40,41] International organizations such as the International Menopause Society, the Australian Menopause Society, and the British Menopause Society have issued similar guidance.[42,43,44] Both The North American Menopause Society and the American College of Obstetricians and Gynecologists recommend that physicians who are considering prescribing CHT first determine whether an appropriate FDA-approved alternative is available.[1,9,11] Some medical and legal experts have expressed concern that physicians who prescribe compounded drugs for reasons other than medical need may be held liable if a woman for whom they prescribed CHT experiences harm from the treatment.[45,46]

Strengths and Limitations of the Surveys

The Rose and Harris surveys have several limitations. Although the sample for each survey was matched to the US population of adult women by age, race, income, education, and region, the survey design limits the generalizability of the data to the population at large. Respondents were accrued from a nonprobability opt-in panel of registered survey takers rather than from a randomized sample of the general population.

Participation rates for both surveys were low but consistent with rates commonly reported for online surveys.[20] Participation has dropped sharply across all survey methods in the past decade; for example, the Pew Research Center[47] reported that the response rate for telephone surveys declined from 36% in 1997 to 9% in 2012. The link between low participation in online surveys and nonresponse bias is unclear, with some data showing no relationship.[20,48] The qualification rate for the Rose survey was numerically low owing to the restrictive eligibility criteria (current and past HT users) and is not a limitation of the survey.

Women received an incentive only if they completed the survey, which may have influenced how they answered screening questions. In addition, all respondents had Internet access, and differences may exist between Internet users and nonusers. The women who took the survey may differ from women in the general population in other meaningful ways that the survey was not designed to capture. The subset of HT users in the Harris study was small, which further limits our ability to extrapolate these data to the general population. The self-reported nature of the questionnaire introduces the potential for recall bias, although self-reporting of HT use has generally been found to be concordant with prescription reimbursement data.[49,50,51]

The wording of the survey questions may have affected responses. It is not clear to what extent women interpreted terms such as "bioidentical," "natural," "individualized," or "compounded" to mean CHT. In addition, we did not ask questions about some demographic characteristics or personal behaviors that have been associated with how women experience menopausal symptoms.

Despite these limitations, as a result of the surveys' use of large sample sizes and standard methods for reducing nonresponse bias (such as purposive systematic sampling, quotas, and weighting), we believe that they have yielded helpful information on menopausal experience and the prevalence of CHT use among US women. The Rose survey, which was central to our extrapolations, accrued more than 17,000 women whose age-based composition mirrored that of the US population. The relative agreement between our observations on the overall use of HT among postmenopausal women and the results of other studies increases confidence in our findings. Data on HT use from other published peer-reviewed surveys and studies suggest that our conclusions may even underestimate the magnitude of CHT use in the United States. Given the substantial use of CHT by US women, larger studies determining what women know about these products could be useful in developing strategies for ensuring that providers have the necessary tools to educate women about CHT.