Under 60 or Within 10 Years: ACOG's Menopause Hormone Window, in Numbers

ACOG calls hormone therapy for hot flashes a "strong" recommendation. It uses the same word for the rule about when to start it, and for two families of nonhormonal drugs. Then it hangs a different evidence grade on each: moderate for the hormones, low-to-moderate for the timing rule, moderate-to-high for the alternatives. That mismatch is the most useful thing in Clinical Practice Guideline No. 13, published Thursday, Oct. 15 in Obstetrics & Gynecology.

In ACOG's scheme, "strong" means "Benefits clearly outweigh harms and burdens. Most patients should receive the intervention." The evidence grade is separate: how sure the authors are about the size of the benefit and risk. In ACOG's scheme, only "high" evidence is described as unlikely to change with further research; "moderate" means randomized trials with some limitations. Put together, that reads as a call that favors treatment for most eligible people, with real room for the numbers to move. (That reading is an interpretation, not ACOG's wording.)

So who is it for? A person with bothersome hot flashes or night sweats, who is under 60 or within 10 years of the start of natural menopause, and who has none of the conditions on the exclusion list. For her, ACOG's strong recommendation favors hormone therapy, after a shared decision about her own risks. For people outside the window it calls for an individualized decision; for people with a listed contraindication, hormones are generally not recommended. This is general information, not medical advice, and your own history decides.

The guideline covers hot flashes, the sleep they cost and mood. Vaginal and urinary symptoms wait for separate guidance ACOG says is coming in 2027. And it makes no recommendation to use hormones to prevent heart disease, dementia or any other chronic disease. It replaces a January 2014 Practice Bulletin.

The window, and who it leaves out

The rule reads: "When MHT is selected for the management of VMS and vasomotor-related sleep disturbance, ACOG recommends initiation of therapy before age 60 years or within 10 years of the start of natural menopause in individuals without contraindications to hormone use." (MHT is menopausal hormone therapy; VMS means hot flashes and night sweats.)

Note the "or." Either half qualifies, and the clock runs from natural menopause, which averages 49.9 in the United States. For most women the two halves overlap until about 60.

Outside the window, hormones are not forbidden. The guideline calls for an individualized decision, because baseline risk of clots, stroke and heart disease is already higher and starting hormones then "may further increase the risk," particularly of clots and stroke. In a Cochrane review of 19 trials, hormones started more than 10 years after menopause carried a clot risk ratio of 1.96 and a stroke ratio of 1.21.

Inside the window the same review found a lower risk of coronary heart disease (ratio 0.52) but still more clots (1.74). That heart finding is a subgroup result; the Cochrane authors' overall verdict was that hormones for preventing cardiovascular disease have "little if any benefit." The timing rule carries the weakest evidence of any of ACOG's strong calls.

The exclusion list is long and specific. It names unexplained vaginal bleeding; past or current breast cancer; certain uterine and ovarian cancers; coronary heart disease; a past or current blood clot; stroke or heart attack; known clotting disorders; liver disease; and pregnancy. (NPR's list leaves out breast cancer and coronary disease. The guideline names both.)

For people who cannot or will not take hormones, ACOG strongly recommends two drug families, on the higher evidence grade. SSRIs and SNRIs are antidepressants that also quiet hot flashes; low-dose paroxetine is the only one FDA has approved for it, and paroxetine and fluoxetine should be avoided with tamoxifen. Neurokinin receptor antagonists block brain signals involved in temperature control: fezolinetant (Veozah, approved May 2023) and elinzanetant (Lynkuet, approved Oct. 24, 2025).

Both come with a liver catch. FDA added a boxed warning to fezolinetant on Dec. 16, 2024, after trial data showed liver-enzyme elevations above three times normal in 12 of 589 women on the drug (2.0%) against six of 583 on placebo (1.0%). Elinzanetant's label calls for liver tests before treatment and at three months. In the 52-week OASIS-3 trial, Contemporary OB/GYN reported, moderate-to-severe hot flashes fell by week 12 from 6.7 to 1.6 a day on elinzanetant and from 6.8 to 3.4 on placebo. The guideline's comparison with hormones is indirect, across trials.

Gabapentin and cognitive behavioral therapy earn conditional grades. Soy isoflavones and herbal supplements earn none: the evidence "does not support the efficacy or safety" of either.

What does hormone therapy fix beyond hot flashes? Less than the coverage implied. On sleep, a meta-analysis of 15 trials found a small gain in self-reported sleep quality, "but not in objective sleep parameters measured by polysomnography" (lab-measured sleep). On mood, the guideline finds the evidence "insufficient to recommend for or against" hormones for depression. Antidepressants and cognitive behavioral therapy remain first-line for major depression.

The risks, per 1,000 women

The guideline's own risk numbers come from the Women's Health Initiative (WHI) trials, as summarized by Cochrane. Read the chart as extra cases over placebo, per 1,000 women.

In human terms: 10 per 1,000 means one extra clot among 100 women over about five and a half years. Five per 1,000 means one extra case among 200. On estrogen alone, taken by women without a uterus for a median of 7.2 years, invasive breast cancer did not rise, though strokes did (eight extra per 1,000, or 1 in 125). The five extra clots with estrogen alone carry a confidence interval that touches 1.00, so chance is not ruled out. See an earlier look at blood clots and DVT.

Breast cancer is where the honest answer is a range, because three measurements exist and they are not the same kind of number. The 2002 randomized trial found about one extra invasive case per 1,250 women a year. A 2017 Cochrane review put the risk after 5.6 years of combined therapy at 20 to 30 per 1,000, up from 19. A 2019 Lancet analysis of 108,647 cancers estimated that five years of estrogen with a daily progestogen, started at 50, might add about one case in 50 users by age 70.

The Lancet figure is the highest. It comes from observational data, which ACOG's guideline excluded, and it is a cumulative figure to age 69, with the authors' own caveat that it holds "if these associations are largely causal." Its estrogen-only estimate was about one in 200. The guideline points to validated tools for a woman's own baseline risk: the American College of Cardiology calculator for cardiovascular disease, and the NCI Breast Cancer Risk Assessment Tool and Tyrer-Cuzick model for breast cancer.

Age changes the arithmetic more than any single risk. The table comes from the final WHI report in JAMA in 2013, covering 27,347 women across both trials. Its "global index" combines heart disease, stroke, pulmonary embolism, breast and other cancers, hip fracture and death, counted per 10,000 women per year. The regimens were oral conjugated equine estrogens, with or without medroxyprogesterone.

A separate analysis in women with hot flashes shows why ACOG drew a line: from 50 to 69, cardiovascular disease did not rise significantly, but at 70 and older it did, by 217 to 382 excess events per 10,000 person-years depending on the regimen. The guideline is also frank about who was in the trials:

"nearly 67% of the WHI participants were aged 60 years or older, which makes the findings difficult to generalize to younger, healthy perimenopausal or recently postmenopausal women, who have a lower absolute risk of chronic diseases such as stroke, coronary heart disease (CHD), and VTE and are the most likely to use and benefit from hormone therapy."

ACOG Clinical Practice Guideline No. 13, Obstetrics & Gynecology, November 2026

The risks at 50 to 59 are lower than the headline trial numbers, but the guideline offers no figure for them, saying only that later analyses show "a very low absolute risk" at that age.

Why a 2002 trial still sets the numbers

On July 9, 2002, NIH announced it had stopped the WHI estrogen-plus-progestin trial, in 16,608 women aged 50 to 79, after a mean of 5.2 years. Breast cancer had crossed the safety board's stopping boundary. FDA put a boxed warning on estrogen products in 2003. Per 10,000 women a year, there were eight more strokes and eight more pulmonary emboli, against five fewer hip fractures.

Twenty-four years later the numbers still lean on that trial. A Cochrane review of long-term use found that "nearly 70%" of its data came from two studies, and that none "focused on perimenopausal women." The guideline adds that observational data suggest the WHI's oral regimen may carry more clot risk than transdermal estradiol.

The boxed-warning story is also muddled in the coverage. FDA announced on Monday, Nov. 10, 2025 that it was "initiating the removal" of broad warnings. It approved the first six revised labels on Thursday, Feb. 12, 2026, 94 days later. (FDA's Feb. 12 announcement of the first six revised labels has the detail.) The endometrial-cancer boxed warning on estrogen-alone products stays.

FDA's November release also claimed that starting within 10 years of menopause might cut cardiovascular disease by as much as 50% and Alzheimer's disease by 35%. CNN reported that the 50% heart-disease claim in an HHS fact sheet rested on decades-old research reviews, and that a separate dementia claim (65%) rested on a small 2005 Danish study of 343 women. ACOG's guideline claims none of it; it treats symptoms only. As JoAnn Manson, the WHI's lead investigator, told NPR in 2013: The recommendation is, don't use hormones for chronic disease prevention, but do consider whether younger women [could use them] for treatment of moderate to severe symptoms of menopause.

Patch, pill or gel, and a pharmacy that is out

The guideline issues no formal recommendation on route, only a tilt. A network meta-analysis of 32 trials ranked transdermal estradiol plus a progestogen most effective for hot flash frequency, and sleep benefit was largest with transdermal estrogen. Clinicians are told to check clot history and "consider the use of transdermal forms."

Independent evidence points the same way, with a caveat. A 2019 BMJ study of 80,396 UK women who had a clot and 391,494 matched controls tied oral therapy to higher risk (adjusted odds ratio 1.58) and transdermal to none (0.93). It was observational: association, not proof.

If you have a uterus, estrogen needs a partner, because it can thicken the uterine lining and raise endometrial cancer risk. That partner is a progestogen, or the bazedoxifene combination (Duavee), which the guideline rates conditional with low-to-moderate evidence and no follow-up beyond two years. A 52-mg levonorgestrel intrauterine device is an off-label way to supply the progestogen. After a hysterectomy, estrogen goes alone. Bleeding is common in the first three months and should be evaluated if it continues after six.

That brings the patch problem. In February, NBC reported that menopause clinics fielded "multiple messages a day" from patients who could not fill patch prescriptions. The American Society of Health-System Pharmacists counted 14 brands or doses in shortage by May 10. CBS News reported on Sept. 3 that patch prescriptions rose from about 594,000 in June 2024 to 1.6 million in May 2026, 2.7 times in 23 months.

Sources disagree. HHS told Healthline in April that "Currently, it is not in shortage," while ASHP and clinicians report shortages. FDA said, in an update CBS reported, that patches remain available, though availability varies by pharmacy, and its shortage database listed no estradiol entry in a single check on Saturday, Oct. 17.

Practical steps are modest: talk to the prescriber and the pharmacist. Gels, sprays and oral tablets are the alternatives clinicians name, though switching is not always smooth. The guideline points to a separate ACOG consensus on compounded bioidentical hormones; background on compounding pharmacies is here. Do not change what you take without your prescriber.

Dr. Catherine Cansino, a guideline coauthor, told NPR what the patch gap reveals:

"I think the shortage of transdermal estrogen therapy highlights the fact that we have all of this attention about hormone therapy, and not necessarily enough about other treatment options."

Dr. Catherine Cansino, guideline coauthor and chair of ACOG's Committee on Clinical Practice Guidelines

How long can you stay on hormones? Dr. Robert Kauffman, another coauthor, told Endocrinology Advisor that ACOG has "taken away the cap," and that a small subset of women may need hormones for many years beyond 60. The guideline asks for a check at three months, then every 12 months, and contraception assessed until age 55. Stopping is individualized, and symptoms may return.

All of that turns into questions you can carry into the room:

• Where am I against the window: my age, and how many years since my last period?
• What is my own baseline risk of heart disease, breast cancer and clots, and does my blood pressure or cholesterol change the answer?
• Does anything on the exclusion list apply, from unexplained bleeding to a clot, a stroke or liver disease?
• Patch, gel or pill, given my clot history, and what is the fallback if my pharmacy runs out?
• I have a uterus: which progestogen, or the bazedoxifene combination, or an intrauterine device?
• When do we check back, what bleeding is expected, and when do we talk about stopping?
• If hormones are not right for me, which nonhormonal option fits, and do I need liver tests?

ACOG's definition of "strong" ends with "Most patients should receive the intervention." That is not "every patient," and it is not whatever happens to be on this month's pharmacy shelf. It is a decision to be made with numbers on the table, and made again at three months and every year after.