How Long Can You Stay on HRT? Duration, Safety, and When to Stop

What the Guidelines Actually Say About HRT Duration

The Menopause Society 2022 position statement and NICE guideline NG23 both reject an arbitrary time limit on hormone therapy. Duration should be individualized, reassessed each year, and driven by symptom burden, quality of life, fracture risk, cardiovascular profile, and personal cancer history. Neither document recommends stopping at 5 years as a blanket rule.

The 5-year caution that became embedded in clinical practice traced back to the 2002 Women's Health Initiative (WHI) trial, which randomized 16,608 postmenopausal women (mean age 63) to conjugated equine estrogen 0.625 mg plus medroxyprogesterone acetate 2.5 mg orally or placebo. The trial was stopped early when the data safety monitoring board detected a small absolute increase in breast cancer, coronary events, stroke, and pulmonary embolism in the active arm [1]. The absolute excess was 8 additional breast cancers, 7 additional coronary events, 8 additional strokes, and 8 additional pulmonary emboli per 10,000 women-years [1]. Critics have since pointed out that the WHI population was older (average 12 years post-menopause), predominantly oral, and used synthetic MPA rather than micronized progesterone, making its risk estimates poorly applicable to perimenopausal women starting transdermal therapy today [2].

The 2022 Menopause Society position states: "For women who are aged younger than 60 years or who are within 10 years of menopause onset and have no contraindications, the benefit-risk ratio is favorable for treatment of bothersome vasomotor symptoms and for prevention of bone loss" [3]. That favorable ratio does not come with a printed expiration date.

For women over 60 or more than 10 years post-menopause, additional cardiovascular and venous thromboembolism caution applies. Continuation is still possible but requires explicit individualized discussion documented in the chart.

How Fast Does HRT Work?

Vasomotor symptoms, specifically hot flushes and night sweats, typically improve within 4 weeks of reaching therapeutic estradiol levels. Most women notice meaningful reduction in flush frequency by week 8 and near-maximal benefit by week 12. Vaginal symptoms, including dryness and dyspareunia, respond more slowly. Full mucosal restoration can take 3 to 6 months with systemic estrogen and up to 12 weeks with local vaginal estradiol preparations [4].

Sleep quality, mood, and cognitive fog often improve in parallel with vasomotor control. A 2021 Cochrane review of 12 trials found that HRT reduced flush frequency by roughly 75% compared with placebo at 12 weeks [5]. Bone density improvement is measurable by DEXA at 12 to 24 months and continues incrementally for years thereafter [6].

Route matters for speed. Transdermal estradiol patches (0.05 mg/day or 0.1 mg/day) reach steady-state serum estradiol of roughly 40 to 80 pg/mL within 3 to 5 days of the first application. Oral estradiol 1 or 2 mg daily achieves comparable serum levels but with higher hepatic first-pass metabolism. Estradiol gel applied once daily to skin behaves similarly to patches in pharmacokinetic studies [7].

If symptoms are not adequately controlled by week 12, the correct step is dose adjustment or route change rather than abandoning therapy. Serum estradiol measurement (target 40 to 200 pg/mL for symptom control, depending on individual response) guides titration.

Long-Term Safety: Parsing the Real Risks

The headline figures from the WHI are frequently cited out of context. Three points help clarify actual risk for the typical HealthRX patient.

First, route modifies thrombotic risk substantially. Oral estrogen increases hepatic clotting factor synthesis, raising venous thromboembolism (VTE) risk approximately 2-fold compared with non-users. Transdermal estradiol does not produce the same hepatic effect. A large UK cohort study of 80,396 women published in the BMJ found that transdermal estrogen was not associated with increased VTE risk at any dose, while oral estrogen was (adjusted relative risk 1.58 to 95% CI 1.24 to 2.02) [8].

Second, progestogen type changes breast cancer signal. The E3N French cohort study (N=80,377, follow-up 8.1 years) found that combined estrogen plus synthetic progestin raised breast cancer risk (relative risk 1.69 to 95% CI 1.50 to 1.91), while estrogen plus micronized progesterone did not exceed baseline risk (relative risk 1.00 to 95% CI 0.83 to 1.22) [9]. These findings support use of body-identical micronized progesterone (Utrogestan 100 or 200 mg) over synthetic MPA in women without contraindications.

Third, duration effects accumulate differently by formulation. Estrogen-only HRT (in hysterectomized women) was actually associated with a reduction in breast cancer incidence in the WHI estrogen-alone trial at 7 years follow-up [10]. The risk picture for combined HRT increases with years of use, which is why annual review matters practically rather than just procedurally.

Absolute annual risk magnitudes help frame shared decision-making: the approximate 8 extra breast cancer cases per 10,000 women-years of combined HRT is comparable to drinking 1 to 2 units of alcohol daily or having a BMI over 30, both of which are widely accepted lifestyle exposures. Communicating risk in absolute terms, not relative terms, is recommended by NICE NG23 [11].

When and How to Stop HRT

Stopping HRT is a clinical decision, not an administrative one. Reasons to discontinue include new diagnosis of hormone-receptor-positive breast cancer, unexplained vaginal bleeding (pending investigation), active VTE or arterial thrombotic event, or patient preference after full risk-benefit discussion.

Abrupt discontinuation ("cold turkey") is physiologically possible and not medically dangerous in most women. However, rebound vasomotor symptoms affect roughly 50% of women who stop suddenly, frequently at an intensity equal to or greater than pre-treatment symptoms [12]. A structured taper over 3 to 6 months reduces this rebound. One practical approach is halving the estradiol dose every 6 to 8 weeks (for example, dropping from 0.1 mg/day patch to 0.05 mg/day, then to 0.025 mg/day before stopping), while simultaneously reducing progestogen frequency. Symptoms should be reassessed at each step.

Bone density declines after stopping. A 1998 study published in the Annals of Internal Medicine tracking 122 women found that bone mineral density fell to non-user levels within 2 to 3 years of HRT cessation regardless of how long therapy had been used [13]. Women stopping HRT who have a fracture risk indication should be evaluated for alternative bone-protective therapy, such as alendronate 70 mg weekly or denosumab 60 mg every 6 months.

Cardiovascular protection from HRT also attenuates after stopping, though the time course is less well characterized than bone effects. Women who gained cardiovascular benefit from early initiation (the timing hypothesis) should be counseled that cessation, particularly abrupt cessation, may allow vasomotor instability to return and may remove any lipid-favorable effects of estradiol.

HRT and Birth Control: Two Different Hormones, Two Different Purposes

HRT doses of estradiol are too low to suppress ovulation. A standard HRT patch delivering 0.05 to 0.1 mg/day of estradiol produces serum levels of 40 to 80 pg/mL, well below the sustained supraphysiologic levels needed to inhibit the hypothalamic-pituitary-ovarian axis for contraception. Women in perimenopause who are still ovulating, even sporadically, remain at risk of unintended pregnancy when using HRT without additional contraception [14].

The Faculty of Sexual and Reproductive Healthcare (FSRH) guideline recommends continuing contraception until at least 12 months after the last spontaneous period in women under 50, and at least 6 months after the last spontaneous period in women 50 and over [15]. FSH levels are unreliable for confirming permanent ovarian failure in women already using HRT, because exogenous estrogen suppresses FSH and can produce falsely reassuring low values.

Practical contraceptive options compatible with HRT include:

• The levonorgestrel IUS (Mirena 52 mg) inserted uterine device, which also fulfills the progestogen arm of HRT in women with an intact uterus while providing contraception.
• Barrier methods (condoms, diaphragm) for women who prefer non-hormonal options.
• The progestogen-only pill (desogestrel 75 mcg daily, brand name Cerazette in the UK) alongside a transdermal estradiol-only preparation.
• Combined oral contraceptives containing ethinyl estradiol 20 to 35 mcg, which provide both contraception and symptom relief in perimenopausal women under 50 without contraindications, per FSRH guidance [16].

Combined OCs are not considered HRT and carry their own VTE risk profile (approximately 3 to 5-fold increase over non-use for most formulations) [17]. The switch from a COC to conventional low-dose HRT is typically made around age 50 to 51, once menopause is confirmed.

HRT and Pregnancy: Can It Happen?

Pregnancy during standard postmenopausal HRT is not possible once true menopause (12 consecutive months of amenorrhea) is confirmed. Ovarian function is absent and follicular activity has ceased. No documented cases of spontaneous pregnancy from intact ovaries exist in confirmed postmenopausal women on HRT.

The risk applies in perimenopause. Follicular reserve declines but is not zero. Ovulation can occur unpredictably even with cycle irregularity or several months of amenorrhea. A woman aged 47 with 8 months of missed periods may still ovulate and conceive without contraception [18]. HRT does not protect against this.

If a perimenopausal woman on HRT suspects pregnancy, a urine or serum beta-hCG should be obtained immediately. Estradiol and progesterone at HRT doses are not teratogenic based on available data, but intentional continuation of HRT during a confirmed intrauterine pregnancy is not indicated and the prescribing physician should be contacted without delay [19].

Donor egg IVF is a separate clinical scenario where exogenous high-dose estrogen and progesterone are used to prepare the uterine lining in menopausal women. This is distinct from menopause HRT and is managed entirely within a reproductive medicine framework.

Annual Review: What the Appointment Should Cover

The Menopause Society and NICE both specify that women on HRT should have a formal review at least once per year. This review is not a formality. The clinician should reassess:

1. Symptom status: Are vasomotor symptoms controlled? Has vaginal atrophy resolved? Has sleep improved?
2. Current dose and route adequacy: Serum estradiol measurement where symptom control is suboptimal.
3. New diagnoses or risk factors: New hypertension, new VTE, new breast cancer diagnosis, significant weight gain, or liver disease all change the risk calculation.
4. Mammography and cervical screening: HRT use does not alter standard screening intervals, but breast density increase on mammography with HRT use is well documented and radiologists should be informed [20].
5. Cardiovascular risk markers: Lipid panel, blood pressure, and fasting glucose.
6. Patient preference: Many women wish to continue indefinitely for quality of life. Others wish to trial cessation. Both are valid starting points for the conversation.

Women who are doing well on HRT at annual review and have no new contraindications should be offered continuation. No guideline currently recommends a compulsory stop at 5 or 10 years. The British Menopause Society states explicitly: "It is not necessary to limit the duration of HRT use, as it can be continued in those in whom benefits outweigh risks" [21].

Special Populations: Over 60, Surgical Menopause, and Premature Ovarian Insufficiency

Women who underwent surgical menopause (bilateral oophorectomy) before natural menopause face a different risk calculus. Abrupt surgical castration removes the protective effect of endogenous estrogens decades ahead of schedule, increasing lifetime cardiovascular risk, osteoporosis risk, and all-cause mortality if HRT is not initiated [22]. The Menopause Society recommends HRT continuation at minimum until the average age of natural menopause (51 years) in surgically menopausal women.

Women with premature ovarian insufficiency (POI), defined as ovarian failure before age 40 and affecting roughly 1 in 100 women, should receive HRT until at least age 51 according to NICE guideline NG23 [11]. The standard of care is estradiol doses higher than typical postmenopausal HRT (often 100 mcg/day transdermal or 2 mg/day oral) with cyclical or continuous progestogen. POI carries a 50% higher risk of hip fracture and significantly elevated cardiovascular risk compared with age-matched controls; hormone replacement is the most effective single intervention to reduce both [23].

Women over 60 starting HRT for the first time face a less favorable benefit-risk ratio due to older vascular age and longer estrogen-free interval. Oral estrogen in this group increases coronary risk, as demonstrated in the WHI subgroup analysis of women over 70 (hazard ratio 1.71 for CHD, 95% CI 1.01 to 2.89) [1]. Transdermal preparations at lower doses remain a reasonable option in carefully selected older women with severe refractory vasomotor symptoms or high fracture risk, provided cardiovascular and thrombotic risk has been formally assessed.

Monitoring Tests During Long-Term HRT

Long-term HRT does not require exotic monitoring. The following schedule is supported by guideline evidence:

• Blood pressure: At every annual review.
• Mammography: Per standard national screening schedule (every 2 to 3 years in most countries for ages 50 to 74). Inform the radiologist of HRT use because combined HRT increases mammographic density in up to 75% of users, which may reduce sensitivity [20].
• Bone density (DEXA): Baseline in women with POI or high fracture risk, then every 2 years while adjusting therapy or post-cessation.
• Lipid panel and fasting glucose: Every 1 to 2 years, especially in women with metabolic risk factors.
• Serum estradiol: Not routinely required in asymptomatic women on standard doses, but useful when symptoms persist or when transdermal absorption is suspected to be variable (for example, in women with high BMI or significant skin conditions) [3].
• Endometrial surveillance: Not routinely indicated for women on continuous combined HRT with a normal uterus. Unexplained vaginal bleeding at any stage requires transvaginal ultrasound to measure endometrial thickness; a threshold of <4 mm in postmenopausal women on HRT makes endometrial pathology unlikely [24].