Earlier menopause, earlier accumulation of chronic disease

Design

retrospective cohort study with Cox proportional hazards and multi-state transition models

Population

121,017 postmenopausal women aged 40-69 in UK Biobank, median follow-up 8.6 years

Primary outcome

multimorbidity, defined as two or more of 35 chronic conditions, and mortality

Effect

health to first chronic disease HR 1.32 (95% CI 1.28-1.36) for premature, 1.14 (1.11-1.17) early, 1.03 (1.01-1.05) relatively early menopause versus after age 50; first disease to multimorbidity 1.22, 1.13 and 1.06

Premature and early menopause are known to carry cardiovascular and bone risk. This analysis asked something more structural: where in the path from health to multimorbidity to death the association actually acts. It followed 121,017 postmenopausal UK Biobank participants aged 40 to 69 for a median 8.6 years, using multi-state models across 35 chronic conditions.

During follow-up 86,821 women developed a first chronic disease, 42,237 reached multimorbidity — two or more conditions — and 10,527 died. Compared with menopause after 50, the transition from health to first chronic disease was 32% more likely with premature menopause (HR 1.32, 95% CI 1.28-1.36), 14% with early menopause and 3% with relatively early menopause. The transition from first disease to multimorbidity showed the same gradient at 22%, 13% and 6%.

The consistent dose-response across both transitions is what makes this more than another association study, but UK Biobank participants are healthier and less diverse than the population, menopausal age is self-reported, and the analysis cannot separate menopause itself from whatever caused it to be early — smoking, deprivation, autoimmune disease, surgery. For clinic, the message is about identification rather than treatment: a woman with premature ovarian insufficiency needs a cardiovascular and bone risk plan made at that point and revisited, not a hormone conversation alone. In India, where age at menopause is on average earlier than in European cohorts, that flag applies to more women than these figures suggest.

• Record age at menopause in the notes; it is a risk marker that is almost never carried forward
• In premature ovarian insufficiency, make a cardiovascular and bone plan at diagnosis
• The gradient is dose-responsive: relatively early menopause carries a small but real excess
• The association cannot separate menopause from its causes — smoking, deprivation, surgery, autoimmunity
• Indian women reach menopause earlier on average than these European participants

The statistics, in plain English

Multi-state models follow women through defined stages rather than to a single endpoint, which is why the same exposure produces different hazard ratios for different transitions — 1.32 into first disease and 1.22 into multimorbidity for premature menopause. Very narrow intervals reflect the size of the cohort, not the accuracy of the exposure: menopausal age was self-reported, and recall error typically pulls associations towards no effect, so if anything these are conservative. A hazard ratio of 1.03 for relatively early menopause is statistically clear and clinically negligible for an individual.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.