How Many Women Use HRT? U.S. and England Data (2026)

There is no single number—and a number without its denominator is incomplete. The U.S. Food and Drug Administration states that in 2020, approximately 2 million U.S. women aged 46 to 65 received an outpatient prescription for systemic estrogen alone or combined estrogen plus progestogen. In England, NHSBSA counted 2,804,440 estimated identified patients who received at least one community-dispensed medicine on the HRT Prescription Prepayment Certificate drug list in 2024/25. Newer U.S. datasets report 3.6% of adult women seen in outpatient care with qualifying hormone-therapy use in April 2026 and 1.7% of women aged 40 and older in an insured claims database meeting a strict 180-day filled-systemic-therapy threshold in 2023. These figures are all real. They do not measure the same thing, and averaging them produces a number that means nothing.

The defensible answer is a set of source-specific numbers, not one universal total. The date matters too. The nationally representative 4.7%U.S. estimate among postmenopausal women comes from a survey cycle ending in March 2020. Separate datasets extending into 2026 show use roughly doubling from their own 2017–2018 baselines. That does not turn the older and newer figures into one continuous national series. It shows why every HRT statistic needs its population, definition, and date attached.

Scope. In this report, HRT means menopausal hormone therapy—also called MHT or HT—as each source defines it. The intended topic is hormone therapy used around perimenopause and menopause, not gender-affirming hormone therapy, hormonal contraception, fertility treatment, or oncology hormone therapy. Some administrative and real-world datasets cannot perfectly separate every indication, product, or patient group. Where that is true, the limitation stays attached to the number.

The strongest available sources report different answers because they count different populations and define “use” differently. The evidence map below preserves each denominator, therapy scope, and measurement rule instead of flattening them into one false estimate.

Data version: 2026-07-31. Comparability rule: figures may be compared over time within a source when its method remains sufficiently stable. They must not be pooled, averaged, or treated as interchangeable across sources.

The evidence supports source-specific counts, rates, and within-source trends. It does not support a worldwide total, one all-purpose U.S. percentage, or the assumption that a written prescription equals medicine continuously taken.

What these figures can support

• A count or rate quoted exactly as its source defines it
• Change over time within one source when the measurement method remains sufficiently stable
• Age, route, formulation, or geographic comparisons within the same dataset
• A distinction between prescriptions written, medicines dispensed, and sustained filled-prescription use
• Cross-country context when the denominator, period, and definition travel with every number

What these figures cannot support

• A harmonized worldwide HRT-user total
• A pooled average of the U.S. studies
• A direct ordering of countries using incompatible measures
• A causal explanation for why prescribing changed
• A claim that every recorded therapy was prescribed for menopause symptoms
• A claim that all 2,804,440 England patients were women
• A count of how many people took every prescribed dose

Is there a worldwide HRT-use count?

We reviewed the World Health Organization’s menopause fact sheet (updated October 16, 2024) and found no worldwide HRT-utilization total. WHO reports that women aged 50 and older accounted for 26% of all women and girls globally in 2021, up from 22% a decade earlier, and describes major gaps in access to menopause information and care. It does not publish a harmonized global HRT-user count.

That is a finding of this review, not a claim that no global estimate could ever be constructed. Available utilization sources are country-specific and use definitions that cannot responsibly be combined without patient-level harmonization.

We read each source in its original form, recorded its own population, denominator, therapy scope, and definition of use, and preserved those differences instead of reconciling them. No estimate was pooled, averaged, or re-weighted.

Source hierarchy

1. Official government drug-utilization and administrative statistics
2. Nationally representative peer-reviewed research
3. Peer-reviewed large-database studies
4. Primary analyses published by the original healthcare-data producer
5. Secondary reporting used only to locate or contextualize an original source

Commercial consumer surveys were excluded from every headline population estimate.

Inclusion criteria

A source entered the evidence map only if it supplied:

• A stated geography
• A stated data period
• A defined population or denominator
• A reproducible definition of therapy or use
• A numeric estimate
• An identifiable original data producer or research team
• Enough methodological detail to state the principal limitation

Calculation rules

• Source units are preserved.
• Percentages and per-1,000 equivalents appear together only when the conversion is exact.
• A source’s own reported change is used when rounded endpoints would produce a different result.
• Every derived figure is labeled The HRT Index calculation and lists its inputs.
• Cross-source estimates are not pooled.
• A count of prescriptions, items, or orders is never relabeled as a count of confirmed users.
• A result for England is never relabeled as a result for the entire United Kingdom.

This is a structured evidence synthesis, not a registered systematic review or meta-analysis. The search was designed to identify the strongest current and historical utilization measures, not to catalogue every HRT study ever published.

The most recent federal absolute count identified in this review is the FDA’s estimate that approximately 2 million women aged 46 to 65 received an outpatient prescription for systemic estrogen alone or combined estrogen plus progestogen in 2020. Newer studies report percentages under different definitions rather than a directly comparable national headcount.

The FDA’s 2 million figure—read it precisely

In its November 10, 2025 statement, the FDA said its review of U.S. outpatient drug-utilization data indicated that in 2020, approximately 2 million women aged 46 to 65 received a prescription for systemic estrogen alone or combined estrogen plus progestogen. Four qualifiers travel with that number:

1. The measurement year is 2020, not 2025 or 2026.
2. The definition is systemic estrogen alone or estrogen plus progestogen. Local vaginal-only therapy and products outside that definition are not counted.
3. The age band is 46–65. Women outside it are not counted.
4. The record is a prescription. It does not prove that the medicine was dispensed or continuously taken.

The FDA placed that count beside approximately 41 million U.S. women aged 45–64 in 2020. The age bands do not match: the population context is 45–64, while the prescription count is 46–65. We report both exactly as the FDA did instead of forcing them into one prevalence calculation.

The federal headcount series before the FDA estimate

Among federal sources reviewed for this report, the last comparable multi-year headcount series identified before the FDA’s 2025 statement was AHRQ Medical Expenditure Panel Survey Statistical Brief #347, published in November 2011 with annual data through 2008.

The headcount fell 67.6%from 17.9 million to 5.8 million across the published endpoints. Total annual HRT prescriptions fell from 112.2 million to 31.8 million, and inflation-adjusted expenditures fell from $5.3 billion to $2.0 billion. The percentage calculations are The HRT Index calculations from AHRQ’s published values; AHRQ reports the rounded declines as 68%, 37%, more than 70%, and more than 60%.

The brief was published in 2011, the FDA statement in 2025—a 14-year publication gap. Their measurement years are 2008 and 2020, a 12-year data gap. They also use different age ranges and therapy definitions, so the two headcounts should not be connected as one continuous series.

The three newer U.S. rates

1.7% in 2023: Mayo Clinic Proceedings.Faubion and colleagues analyzed women aged 40 and older in a large national commercial-insurance and Medicare Advantage claims database. “Use” required at least 180 days of filled systemic HT prescriptions in a calendar year. The rate fell from 4.4% in 2007 to 1.7% in 2023. Among women aged 50–59 it was approximately 3.5%. (Journal record; Mayo Clinic summary)

3.6% in April 2026: Epic Research.Published July 30, 2026, this is the newest broad U.S. utilization estimate identified in this review. Epic analyzed 163,348,471 women aged 18–79 seen in outpatient care from January 2017 through April 2026, excluding women with a history of breast cancer, a current pregnancy, or evidence of gender-affirming care. Qualifying HT use rose from approximately 1.7% to 3.6%—a 111.8% relative increase from the published endpoints. Among women aged 45–54, use rose from approximately 24 to 68 per 1,000, or 6.8%, a 183.3% relative increase. The percentage changes are The HRT Index calculations. (Epic Research)

2.35% in February 2026: Truveta.Truveta measured estrogen-based HRT prescribing among women aged 18 and older with available prescription data. The rate rose from 11.5 to 23.5 per 1,000 between January 2018 and February 2026—a 104.8% increase reported by Truveta. Among women aged 45–54, the rate reached 49.0 per 1,000, or 4.9%. Truveta labels the findings preliminary and not peer reviewed. (Truveta)

The most recent nationally representative U.S. estimate identified specifically among postmenopausal women is 4.7%, from the NHANES 2017–March 2020 pre-pandemic cycle.The 95% confidence interval was 3.4%–6.5%. Within that cycle, use was 9.4% among women younger than 52, 4.5% among those aged 52 to under 65, and 4.3% among those 65 and older.

This is the strongest current nationwide prevalence estimate in this review for the population named in the query: postmenopausal women. It is nationally representative of the U.S. civilian noninstitutionalized population; prescription medication was collected in household interviews; and 86% of MHT containers were physically seen. It also ends in March 2020. That date must travel with the 4.7% figure.

The highest NHANES point estimate is 29.5%, not 26.9%

The paper’s headline endpoint comparison starts at 26.9% in 1999–2000 and ends at 4.7% in 2017–March 2020. Its complete Table 1 contains a higher point estimate in the next cycle.

The highest point estimate in the published series is 29.5% in 2001–2002. Measured from 29.5% to 4.7%, the decline is 84.1%. Measured from the paper’s 1999–2000 endpoint of 26.9%, it is 82.5%. Among women aged 52 to under 65, the decline from 38.5% to 4.5% is 88.3%. These are The HRT Index calculations from the published table. The paper is not wrong to compare its first and last cycles. The useful correction is narrower: 26.9% is the study’s starting endpoint, not its highest published point estimate.

The racial disparity in the published point estimates

The paper’s discussion also notes that non-Hispanic Black women are more likely to experience severe menopausal symptoms and less likely to receive treatment.

What formulations counted in the latest NHANES cycle?

Among MHT users in 2017–March 2020, estrogen-only therapy accounted for 52.8%, estrogen plus progestogen 36.1%, progestogen-only 10.5%, and estrogen plus testosterone 0.6%. These are shares of MHT use within the study, not prevalence percentages among all postmenopausal women.

NHS Business Services Authority recorded 2,804,440 estimated identified patients who received at least one community-dispensed HRT-list item in England during 2024/25, alongside 14,725,133 prescribed items. The patient count was 6.4% higher than in 2023/24. The data are England-only and are not restricted by gender, age, or menopause indication.

We did not identify a current, directly comparable U.K.-wide absolute headcount. A separate U.K. primary-care study reported prescribing prevalence and first-prescription incidence, not a national patient total.

The complete England series

Estimated identified patients were 2.73 times the 2015/16 level, while prescribed items were 4.71 times the 2015/16 level. From 2020/21 to 2024/25, the patient count increased 102.6%. In 2024/25 there were 5.25 prescribed items per estimated identified patient at the aggregate level. These are The HRT Index calculations.

Why “estimated identified patients” is the accurate wording

NHSBSA defines a patient as a unique NHS number captured from a prescription form or electronic prescription message. It identifies NHS numbers for most, but not all, prescription items, so its patient totals are estimates of identified patients rather than a complete census. The methodology also states that the publication does not distinguish patients by gender, includes all age groups, and does not receive the clinical indication. “HRT” is a catch-all term for drugs on the HRT PPC list, not one universal clinical definition.

• Accurate: “2,804,440 estimated identified patients in England”
• Accurate when rounded: “approximately 2.8 million NHS HRT-list patients in England”
• Not supported: “exactly 2.8 million women”
• Not supported: “2.8 million people across the U.K.”

Who appears in the England patient count?

NHSBSA also reported 787,159 estimated identified patients in the least deprived quintile and 343,599 in the most deprived quintile—a 2.29-fold raw-count difference. This is not an age-standardized prescribing-rate comparison. Across Integrated Care Boards, the 2024/25 rate ranged from 72.3 estimated identified patients per 1,000 residents to 21.3 per 1,000—a 3.39-fold spread against an England rate of 48 per 1,000.

The publication excludes private prescriptions, hospital and prison prescribing, prescriptions issued but never dispensed, and specified over-the-counter use such as Gina vaginal tablets. Those omissions mean the 2.8 million figure is not a complete count of every HRT transaction in England. But the included dataset is also not restricted to women, menopause-age patients, or menopause indications. The net difference cannot be calculated from these data.

Both descriptions are supported, depending on the source and time window.Long-run U.S. measures show a steep decline from the early 2000s. Epic and Truveta show more recent increases from 2017–2018 through early 2026. England’s estimated identified-patient count rose in every financial year except 2020/21. These are separate series, not one curve.

The long decline

The recent rise

Epic describes its overall rate as roughly flat near 1.8%–1.9% through 2022, then rising more sharply from 2023. Truveta’s increase runs across the full January 2018–February 2026 period.

The product and route mix is changing

Epic found that growth was concentrated in single-hormone products and non-oral routes. Progestin-only use increased about 3.3-fold (from roughly 0.3% to 1.1%); estrogen-only use about 2.3-fold (from roughly 1.1% to 2.6%). Transdermal use increased about 3.5-fold (0.33% to 1.17%) and vaginal use rose from 0.45% to 1.26%.

Truveta’s separate dispensing cohort also recorded a shift toward non-oral estrogen: patch dispensing rose from 6.2 to 19.9 per 1,000, vaginal cream from 2.8 to 14.3 per 1,000, and oral use fell from 10.4 to 8.6 per 1,000—a 16.5% decrease reported by Truveta. Both data producers show growth in non-oral estrogen delivery. That agreement supports the direction of the route shift; it does not make their absolute rates interchangeable.

Charting rule: do not connect NHANES, MEPS, Mayo, Epic, Truveta, and NHSBSA in one line. Separate charts preserve the truth. One merged line would imply continuity the data do not have.

Because “use” and “women” are operational definitions, not self-explanatory variables. A prescription written, a product dispensed, 180 days of filled supply, and a qualifying outpatient record are four different events. A rate among postmenopausal women is not interchangeable with a rate among all adult women who used outpatient care.

Definition crosswalk

Denominator crosswalk

Once both the denominator and the use definition are attached, 1.7%, 3.6%, and 4.7% stop looking like contradictory answers. They are measurements of different populations and events.

The FDA placed three figures side by side and concluded that menopausal hormone therapy may be under-used among women likely to benefit. Those figures show a large gap in scale, but they cannot support a precise treatment-gap percentage because their age ranges, populations, indications, and therapy definitions do not align.

It is tempting to apply the 34% survey estimate to the 41 million Census count and compare the result with the 2 million prescription estimate. This report does not publish that calculation because the resulting percentage would be methodologically unsound: the 41 million count covers all women aged 45–64; the 34% estimate applies to postmenopausal U.S. women aged 40–65; the prescription count is for ages 46–65; and a prescription does not prove symptom response, preference for treatment, or clinical appropriateness.

The FDA’s own conclusion is the defensible one: its utilization data suggest MHT may be under-used among women likely to benefit. The available inputs do not establish how many untreated women should receive it, so this report does not publish a treatment-gap count or percentage.

The national NHANES and FDA utilization figures describe 2020 or earlier, while Epic and Truveta extend into 2026.The U.S. labeling environment also changed in 2025–2026, but the recent rise began before those actions. The utilization data cannot establish that the FDA panel or label changes caused the increase.

Two timeline corrections matter:

• November 10, 2025 was a request for labeling changes, not proof that every product’s warning changed that day.
• The recent utilization rise predates the labeling actions: Epic’s sharper climb begins in 2023, and Truveta’s series rises across 2018–2026.

A 2020 national-survey estimate and a 2026 outpatient estimate are separated by time, method, and regulatory context. None of those differences should be erased.

No source measures every woman, every formulation, the reason for every prescription, and day-to-day medication use in real time. This report presents several valid but non-interchangeable estimates rather than manufacturing one definitive total.

• No harmonized global utilization total was identified. Available sources are country-specific and methodologically incompatible.
• The nationally representative U.S. postmenopausal data end in March 2020. NHANES also inferred menopause status for 5.7% of participants with missing reproductive information.
• The FDA absolute count is a 2020 outpatient prescription estimate for women aged 46–65 and specified systemic therapies. The statement does not publish a downloadable utilization series.
• The Mayo 1.7% estimate uses a strict 180-day filled-prescription threshold in an insured claims population. It intentionally measures a narrower form of sustained systemic use.
• Epic’s denominator is women using outpatient care, not all U.S. women. Menopausal status is approximated with age bands, and the analysis was published by the original EHR data producer rather than in a peer-reviewed journal.
• Truveta’s findings are preliminary and not peer reviewed. Menopause is not confirmed, and the prescribing analysis covers estrogen-based products.
• NHSBSA’s England data are administrative and not restricted by gender, age, or indication. Patient identification is high but not complete, and private, hospital, prison, unfilled, and specified OTC use are outside scope.
• AHRQ’s MEPS series ends in 2008 and uses a broad legacy HRT drug classification. It should not be treated as a direct predecessor to the FDA count.
• The U.K. primary-care study and Australian PBS study are database samples, not current national headcounts.
• No prescription or dispensing database proves adherence. A written order, a pharmacy fill, and medicine taken as directed are different events.
• Therapy scope differs. Vaginal estrogen, progestogen-only products, combination products, and testosterone-containing products enter some sources and not others.
• Rounded endpoints can change derived percentages. Source-reported changes take priority when rounding creates a mismatch.
• This is a structured evidence synthesis, not a meta-analysis or clinical guideline.

This report describes utilization. It does not decide whether hormone therapy is appropriate for any individual. Starting, continuing, changing, or stopping hormone therapy is a clinical decision. Current FDA-approved prescribing information and a qualified clinician are the appropriate sources for an individual treatment decision.