Details changed to protect confidentiality. Composite of similar cases I have seen in practice.

She sat down opposite me on a Tuesday afternoon and said, before I had asked anything, “My GP thinks I am depressed. I do not think I am.” Her periods were still regular. She was tearful, not sleeping, and she had been on an SSRI for four months that had not touched it. She was frightened, because the first label she had been given was one she did not recognise in herself.

I hear a version of this conversation most weeks. A woman in her forties or early fifties, mood changed, cycles still arriving on time, already four months into a medication that was supposed to help — and it has not helped, and she is starting to wonder if the diagnosis was wrong. In November 2024, for the first time in nine years, the National Institute for Health and Care Excellence updated its menopause guideline, NG23 — NICE. Much of what that update changed speaks directly to the woman sitting opposite me on that Tuesday. This is the conversation I wish we had more time for.

Why mood arrives before the cycles do

The shorthand many GPs were taught — and many textbooks still carry — is to wait for cycle change before naming perimenopause. The 28-day rhythm goes irregular, the periods shorten or lengthen, and at that point the conversation begins. But the hormonal tide that underlies perimenopause starts moving years before the cycles visibly break rhythm. Oestrogen in particular fluctuates dramatically through the late 30s and into the 40s. Mood, sleep, and anxiety are often the first places a woman notices this — not the bleeding pattern.

A 2024 study in Nature Mental Health found that perimenopausal women are around 40% more likely than premenopausal women to experience new depressive symptoms (Exploration of first onsets of mania, schizophrenia spectrum disorders and major depressive disorder in perimenopause — Nature Mental Health, 2024). The research group also found an elevated first-onset risk for more severe mental illness during this window, which matters because the assumption that perimenopause is “mild” has underwritten a lot of dismissal.

A 2025 systematic review of UK qualitative studies gathered the accounts of 3,462 women (Women’s Menopausal Experiences in the UK — International Journal of Environmental Research and Public Health, 2025). The pattern that came through, repeatedly, was the language of dismissal: “it is just a phase,” “you are stressed,” “everyone goes through this.” A YouGov poll commissioned by the Royal College of Psychiatrists found only 28% of UK women know that perimenopause can trigger new mental illness at all (Severe mental illness and the perimenopause — BJPsych Bulletin, 2023).

The window she has come in during is exactly the window the specialist system and the 10-minute appointment miss most often.

What NICE NG23 changed in November 2024

The 2024 revision to NG23 makes two points that matter on this topic.

First, for women aged 45 and over presenting with typical perimenopausal symptoms, an FSH blood test is not required to make the diagnosis. The symptoms are the diagnosis. If your GP is “waiting for the blood test to come back before we can call it anything”, that is no longer what the guideline supports in this age group (Menopause: identification and management (NG23) — NICE).

Second, and more important for the woman in my consulting room on that Tuesday: NICE now names HRT — not antidepressants — as first-line for new-onset low mood associated with perimenopause. The British Menopause Society welcomed the update explicitly on this point (BMS statement in response to the updated NICE Menopause guideline NG23 — BMS, 2024).

That is a shift many GPs, including many who trained in the years between the 2015 guideline and now, have not yet absorbed. It is not that SSRIs are wrong for every perimenopausal woman who is low. It is that the order of the conversation has changed. HRT, first. Antidepressants, after that, if still needed.

HRT first-line for mood, in plain English

When a 46-year-old comes in with four months of low mood, no sleep, tearfulness, and cycles still broadly intact, the conversation I now try to have is this. The hormonal picture underneath her symptoms is a period of oestrogen fluctuation. Fluctuating oestrogen is a plausible mechanism for her mood. A trial of appropriate HRT, reviewed at three months, is the step the guideline recommends we try first. If in three months her mood has lifted — and in many women it will — we have our answer. If it has not, we revisit; we may add an SSRI, or we may switch approach entirely.

This is not a promise HRT will fix every perimenopausal woman’s low mood. It will not. It is a different default starting point — and one that reflects the evidence better than the starting point many of us were using five years ago.

The breast cancer numbers, in absolute terms

The reason HRT was not the default for decades is the breast cancer signal. Every woman who has been told “HRT increases your breast cancer risk” is thinking of something, and what she is thinking of is not always what the evidence actually says. The single best synthesis remains the 2019 meta-analysis in The Lancet by the Collaborative Group on Hormonal Factors in Breast Cancer (Type and timing of menopausal hormone therapy and breast cancer risk — The Lancet, 2019). The absolute numbers matter. Five years of combined (oestrogen plus daily progestagen) HRT starting at age 50 raises 20-year breast cancer risk from around 6.3% to 8.3%. That is an absolute increase of two in every hundred users. Oestrogen-only HRT raises it from 6.3% to 6.8% — half a case per hundred users.

These are real numbers and they should be named. They are also not what most women are picturing when they hear “HRT increases breast cancer risk”. In relative terms, the newspaper headline can be made to sound frightening. In absolute terms, for a woman struggling badly with perimenopausal low mood, the balance of benefit and risk looks different — and the conversation can be had honestly, in real figures, rather than in a reflex no.

The honest clinical answer is always individual. HRT is not appropriate for every woman. Duration, type, age at start, personal and family history all factor in. But the absolute-risk frame is the one the conversation should start in.

Mind and body are one system

The specialist system we have built in the UK is good at diseases and poor at people. A woman in her forties with low mood and regular periods may end up under a GP and — if the SSRI fails — a psychiatrist, without anyone asking the mind-body question at all. The perimenopausal low mood story is one of the clearest examples in medicine of why the 10-minute appointment and the specialty split fail patients. Mood and hormones are not two things. They are one system that we happen to divide into two clinical boxes for convenience.

The whole point of this channel is to sit with the connections the system misses. This one belongs at the top of that list.

The three sentences for your next appointment

If you are a woman in your forties and your mood has shifted before your cycles have, the sentences that help in a 10-minute appointment are these.

  1. My mood changed before my cycles did.
  2. Could this be perimenopausal?
  3. I would like to talk about HRT before we talk about antidepressants, based on NICE NG23.

You do not need to lecture your GP. You do need to give them the name of the guideline and the shape of the conversation you want to have. Most GPs respond well to a patient who has done this work. A small number will not, in which case the question is whether to ask for a second GP appointment, a menopause-specialist referral, or a private consultation — that choice is yours, and depends on what is available where you live.

Honest uncertainty

The individual breast cancer risk from HRT depends on type, duration, age at start, and baseline risk. The 2019 Lancet meta-analysis is the best synthesis we have, but it is observational evidence, not a randomised trial. There are women for whom HRT is not appropriate at all. Mood change in perimenopause is multifactorial; HRT helps many women, does not help all of them, and the honest answer on an individual case is always “let us try, review in three months, and be willing to change course”.

That last sentence is the whole of medicine, really. The guideline is a starting point, not a promise. Your GP’s clinical judgement still matters. What has changed is the order of the conversation.

If nothing else, I would like you to leave this piece knowing that the sentence “it is just stress” is no longer the right reflex for the 45-year-old whose mood has changed before her cycles. It may be stress. It often is not.

Hear more of the story on The Accidental Doctor.

Read more like this in the weekly Dr Michael newsletter.

Educational content only. Not personal medical advice. If you are worried about your health, speak to your GP or call NHS 111.

References

Reviewed by Dr Michael, MBBS, MRCGP, 17 April 2026.