Why your mood shifted, why you can't sleep, and whether brain fog means something serious.
A 1-minute view of what this section covers. Tap any item to read the full discussion.
The contradiction likely reflects formulation, timing, indication, and real-world prescribing differences. A combination RCT used estradiol/dydrogesterone plus escitalopram; a UK specialist clinic reported improvement across regimens; Danish and Korean register studies captured all real-world systemic prescribing and found harm signals.
Read full discussion →Among postmenopausal women, those with sleep disturbances alone had worse quality-of-life outcomes than those with VMS alone (47,841 women). Sleep preceded hot flash awareness by ~6 months in a longitudinal study.
Read full discussion →65% of women reported cognitive decline, yet mean objective scores remained in the average-to-high-average range. Changes tend to improve during postmenopause.
Read full discussion →A WHO-commissioned systematic review found no overall significant association between MHT and dementia. Three large datasets disagree on details, but all point toward clinical decisions based on established HRT indications, not dementia prevention.
Read full discussion →"Am I going crazy?" "Why can't I sleep?" "Is this brain fog?" These are the questions women bring to this section. The evidence addresses all three — and the most important thing the research shows is that these are connected but distinct, each with its own evidence base and its own trajectory.
Between 45% and 68% of women experience heightened depressive symptoms during perimenopause — about 1.4 times more likely than during the premenopausal years. But most women remain stable. Australian longitudinal data found that only about 9% had increasing depressive symptoms linked specifically to the transition, oophorectomy, or hormone therapy changes (120).
The evidence is clear: the menopausal transition alone does not universally increase risk for depression. Specific subgroups carry the elevated risk — women with severe vasomotor symptoms, sleep disturbances, sensitivity to estrogen changes, early or surgical menopause, prior history of mental health disorders, or significant psychosocial stressors (120).
If you have a history of major depressive disorder, the numbers are different. Women with prior MDD had a 59% recurrence rate during midlife, compared to 28% in women without prior history — and 13 times more depressive symptoms during perimenopause (120). In a 19-year SWAN longitudinal study, women with consistently high depressive symptoms before their final menstrual period had an odds ratio of 6.88 for postmenopausal depression. Childhood trauma and maltreatment were enduring risk factors independent of everything else. Social support was protective (12).
Symptom prevalence varies globally and across populations. A meta-analysis of 265 studies (349,608 women) found a global depressive symptom prevalence of 43%, ranging from 22% in White women to 27% in Black women in the SWAN cohort, and reaching 82% in a Cambodian sample. Among 334 African American women, 83% reported anhedonia, 80% fatigue, 74% anxiety, and 73% feeling down or hopeless — alongside a 69% comorbidity rate for hypertension, diabetes, or heart disease (120, 110).
The symptoms that drive the experience of perimenopause are not the ones most commonly associated with it. In a study of 1,263 women (38), five symptom groups predicted the feeling of "not feeling like myself": anxiety and vigilance (β=0.220), fatigue and pain (β=0.192), brain fog (β=0.123), sexual symptoms (β=0.121), and volatile mood (β=0.083). Together they explained 41.8% of the variance in identity disruption. Hot flashes and vaginal dryness were weaker predictors. If you feel like you are losing yourself, that experience has a measurable pattern in the research.
Women describe perimenopause-related anxiety differently from clinical GAD. Rather than excessive worry occurring most days for six months or more, they describe "decreased tolerance for stress" and "insignificant things becoming major stressors." This is linked to fluctuating affecting cortisol and catecholamine pathways — a different mechanism from trait anxiety (38).
Volatile mood — irritability, sudden mood changes, sudden anger or rage — may be a standalone perimenopause symptom rather than a component of depression. The evidence suggests it is linked to different estrogen-change patterns than depressive symptoms (38).
You may encounter conflicting information about whether hormone therapy helps depression during perimenopause. Both claims have real studies behind them.
Under controlled conditions with specific formulations, hormonal therapy helped. A three-arm randomized controlled trial (n=195, 12 weeks) found that combined with significantly outperformed either treatment alone for depression and anxiety scores (124). The largest UK specialist menopause clinic (n=920) reported a 44.59% reduction in depression scores after MHT initiation or optimization, with all regimens effective — and alike (66).
In real-world prescribing data, systemic hormone therapy was associated with increased depression risk. A Danish register study following 825,238 women found systemic HT associated with about 70% higher risk of depression in the first year (HR 1.72), with risk highest in younger women aged 48-50 (14). A Korean multi-database study of 17,098 women found similar results — HRT during the symptomatic menopausal transition was associated with more than double the depression risk (HR 2.21) (24).
Why the disagreement? The positive evidence came from defined clinical contexts — oral estradiol/dydrogesterone plus escitalopram in one RCT, and MHT initiation/optimization in one specialist-clinic dataset — under controlled or specialist conditions. The register studies captured all real-world prescribing with mixed and unspecified formulations. Confounding by indication is likely: women prescribed HRT are also more likely to be experiencing symptoms that lead to a depression diagnosis. However, the Danish data showed that HT users were actually healthier at baseline, which complicates a simple confounding explanation. The answer to "does hormone therapy help depression?" depends on which hormone therapy, for whom, at what timing, and with what progestogen.
One route-dependent finding adds perspective: in the same Danish data, was associated with lower depression risk in women over 54 (HR 0.80) — the opposite direction from systemic HT. One plausible explanation is that treating genitourinary symptoms may improve quality of life, which could affect mood (14).
ADHD and perimenopause share substantial symptom overlap — brain fog, difficulty concentrating, mood swings, emotional dysregulation. The key distinction is timing: ADHD symptoms are typically present from childhood, while perimenopause-linked cognitive and mood symptoms emerge in midlife (125). See the Special Populations section for the full ADHD evidence, including medication and hormone therapy considerations.
CBT is specifically recommended by NICE guidelines for depressed mood during menopause. Mindfulness-based therapy has shown improvements in menopause-specific quality of life and reductions in anxiety, depression, and stress, though some of these effects need further validation. Vitamin D supplementation may improve anxiety more than depression in the available evidence, though certainty is limited (120). See the Treatment Options section for exercise dosing and the full non-pharmacological evidence.
Sleep disruption during the menopausal transition is not a secondary symptom of hot flashes — the evidence establishes it as an independent health concern with its own trajectory, its own cardiovascular consequences, and its own treatment evidence.
Insomnia prevalence was 31-42% at any given SWAN assessment visit, with late perimenopause carrying about 30% higher odds than early perimenopause (1). In a German survey of 26,338 peri- and postmenopausal women, 90.3% had at least some sleep disturbance and 48% met criteria for clinical insomnia (93).
Among postmenopausal women, those with sleep disturbances alone had worse quality-of-life outcomes than those with vasomotor symptoms alone (p<0.001) — in a study of 47,841 women across the US and Europe (95). Sleep is not just a side effect of hot flashes. It has its own impact.
CBT-I is the behavioral treatment with the strongest evidence — see the Treatment Options section for remission rates and delivery format.
Combination plus outperformed either alone for insomnia in a parallel 3-arm trial (n=166, 12 weeks) — the same dual-pathway pattern seen in the depression trial (65).
Mind-body therapies produced a large pooled effect on sleep (SMD -0.86, 18 RCTs, 1,572 women), with yoga and Qigong showing the most stable signals for sleep specifically and mindfulness and music therapy better for emotional regulation (102). These effect sizes are promising but come with high heterogeneity (I²=88%) and most modality comparisons rest on 1-2 studies — symptom-driven and feasibility-driven choices are more appropriate than a strict hierarchy.
Dual orexin receptor antagonists (suvorexant, daridorexant, lemborexant) are emerging as pharmacological options that target the orexin system directly (129).
Cognitive changes during perimenopause — difficulty concentrating, word-finding problems, feeling mentally slower — are real, hormonally linked, and documented in objective testing. But for most women, they reflect relative decline from their own baseline, not absolute impairment.
In a study that tracked over 1,500 women in the US and Canada, forgetfulness was the single most bothersome symptom — rated higher than hot flashes, anxiety, or pain — for women in both late reproductive stage and menopausal transition (38).
Interestingly, several symptoms women associate with menopause — fatigue, headache, anxiety, brain fog — are common at similar rates across pre-, peri-, and postmenopausal women (50). This does not mean they are not real during perimenopause. It means the transition may amplify an experience that is not exclusive to it.
The relationship between hormone therapy and long-term dementia risk is genuinely uncertain. Three large datasets disagree, and the disagreement is informative.
[EVIDENCE] A WHO-commissioned systematic review and meta-analysis (10 studies, over 1 million women) found no overall significant association between hormone therapy and dementia or mild cognitive impairment. The only included randomized trial (WHIMS, women 65 and older at initiation) found combined therapy associated with higher probable dementia risk (HR 1.76), but this was a modest absolute excess of about 7 per 1,000 women. Several very-low-certainty subgroup signals pointed in different directions depending on formulation, duration, and timing. There was insufficient data to evaluate the effects of route, estrogen type, progestogen type, or dose. (PMID 41448220, ref 106)
A Danish nationwide study (5,589 dementia cases, 55,890 controls, 19-year follow-up) found that combined — predominantly with (83.5%) — was associated with higher all-cause dementia risk that increased with duration. Progestin-only therapy and vaginal estrogen-only were not significantly associated. The study had no data on (19).
A Korean register study (1,399,256 women) found formulation-stratified signals: and were each associated with slightly elevated Alzheimer's risk; was not significantly associated, but the transdermal subgroup was small (n=2,081) with wide confidence intervals — making this weak negative evidence rather than positive evidence of safety (26).
What this means for you: The current evidence does not support using hormone therapy for the purpose of preventing dementia. The European Society of Human Reproduction and Embryology recommends HT for dementia prevention in premature ovarian insufficiency, but the WHO-commissioned review found no studies supporting that specific indication. The established evidence base for hormone therapy is stronger for other indications — vasomotor symptoms, genitourinary health, and bone protection — than for dementia prevention.