If you've read that hot flushes last two or three years and you're four years in, you have probably concluded that something is wrong with you.
Nothing is wrong with you. The two-to-three-year figure is out of date, and it was never based on much.
The short version
- The largest study to date found a median total duration of 7.4 years for women with frequent flushes.
- They persisted for a median of 4.5 years after the final period.
- When yours started is the strongest predictor. Early starters have them for much longer.
- Older guidance said six months to two years. That has been overturned by better data.
What the evidence actually shows
The study is SWAN — the Study of Women's Health Across the Nation — analysed by Avis and colleagues and published in JAMA Internal Medicine in 2015. It followed 3,302 women across seven US sites from 1996 to 2013, with the duration analysis covering the 1,449 who reported frequent flushes or night sweats, meaning six or more days in the previous fortnight.
Among those women, the median total duration was 7.4 years, and symptoms persisted a median of 4.5 years after the final menstrual period.
Median means half the women had them for longer. Some had them for around 14 years.
For context on how far that shifted things: many patient information sources at the time said three to five years, and clinical guidelines said six months to two years. This study is the reason that changed. If you're reading something that still quotes the old figures, it hasn't been updated.
The bit that actually predicts your answer
Here is the useful finding, and it's the one most summaries skip.
When your flushes started predicts how long they'll run, and the difference is large:
- Started while still having regular periods, or in early perimenopause → median 11.8 years total, with around 9.4 of those years after the final period
- Started only after periods had stopped → median 3.4 years
So the woman who began flushing at 44 while her cycle was still fairly regular is on a very different trajectory from the woman whose first flush came a year after her last period — roughly three times the length.
That's counter-intuitive. You might reasonably assume an early start means an early finish. It's the opposite.
Ethnicity makes a difference too
In SWAN, African American women had the longest total duration — a median of 10.1 years.
The NHS also notes that ethnic background can affect symptoms, giving the example that women from a Black ethnic background are more likely to have hot flushes that are severe and continue for longer.
This matters because most menopause content is written as though everyone's experience is the same length and shape. If yours is running longer than the figures you keep reading, that may not be unusual at all — and it's worth saying in an appointment rather than assuming you're an outlier who should just cope.
What to be careful about with these numbers
Two honest caveats.
SWAN is a US study. There's no reason to think British women's biology differs, but the healthcare context, treatment rates and reporting do. Treat the figures as the best available estimate rather than a British measurement.
It's observational. It followed women and recorded what happened. It doesn't establish cause, and it can't tell you what your duration will be — only what the distribution looked like across 1,449 women.
What it does do is demolish the two-year figure, which was the point.
You don't have to wait it out
This is the part that matters more than the arithmetic. Seven years is a long time to endure something treatable, and flushes are treatable.
HRT. NICE guidance says to offer HRT to people with vasomotor symptoms associated with menopause. It remains the main medicine treatment.
Menopause-specific CBT. Since the 2024 update, NICE recommends considering it as an option for vasomotor symptoms — in addition to HRT, or as an alternative for people for whom HRT is contraindicated or who prefer not to take it. It works by reducing the distress and disruption the flushes cause rather than by stopping them outright, which sounds like a lesser goal until you've had a flush during a meeting.
Fezolinetant. A newer non-hormonal option. NICE recommends it for moderate to severe vasomotor symptoms when HRT is unsuitable — see NICE's technology appraisal TA1143. Worth knowing it exists if HRT isn't an option for you, because it's recent enough that not every article mentions it.
What NICE says not to do first: SSRIs, SNRIs and clonidine should not routinely be offered as first-line treatment for vasomotor symptoms alone. If that's what you're offered before HRT or CBT has been discussed, asking about the alternatives is reasonable.
The reframe
Most women arrive at this question wanting to know how much longer they have to hold on.
The better question is why you're holding on at all. The duration figures matter mainly as an argument against waiting — if the median is 7.4 years and you started early, "it'll pass soon" is not a plan.
Take the numbers to an appointment rather than to the internet.
- Published
- 5 August 2026
- Evidence last checked
- 5 August 2026
Sources
- Avis NE, Crawford SL, Greendale G, et al. Duration of Menopausal Vasomotor Symptoms Over the Menopause Transition. JAMA Internal Medicine, 2015 — SWAN cohort, 1,449 women with frequent symptoms; median total duration, post-FMP persistence, onset-stage and ethnicity findings. pmc.ncbi.nlm.nih.gov (opens in a new tab)
- NICE guideline NG23, Menopause: identification and management (published 12 November 2015; last updated 15 April 2026) — HRT for vasomotor symptoms, menopause-specific CBT, fezolinetant, position on SSRIs and SNRIs. nice.org.uk (opens in a new tab)
- NHS, Symptoms of menopause and perimenopause — effect of ethnic background on symptom severity and duration. nhs.uk (opens in a new tab)
Ann Suleman
Ann Suleman is going through menopause herself. She isn't a nutritionist or a doctor — she reads the evidence carefully, cites it, and writes about what she finds.
Before you go — the free nutrition planner
A printable weekly planner built around what the evidence supports for perimenopause and menopause. Yours by email, no charge.