Dólleman · Human reproduction (Oxford, England) 2014 · prospective cohort study · n=314

Anti-Mullerian hormone is a more accurate predictor of individual time to menopause than mother's age at menopause.

Cited 73 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective prognostic cohort study evaluating biomarker-based prediction models.

PubMed 24435779 · doi:10.1093/humrep/det446 · record verified 2026-08-26

What was done

The authors evaluated predictors of time to menopause using data from two population-based cohorts. Group 1 included 164 mother-daughter pairs from the DOM cohort to test the predictive value of mother's age at natural menopause. Group 2 consisted of 150 regularly cycling women pooled from two independent studies with baseline anti-Müllerian hormone (AMH) and mother's age at natural menopause followed prospectively over a 12-year period. Cox proportional hazards regression assessed the predictive capacity of female age, mother's age at natural menopause, and AMH for time to menopause, with discrimination measured by C-statistics and incremental value measured by the net reclassification index.

What was found

A model with female age and mother's age at natural menopause had C-statistics of 79% in Group 1 and 85% in Group 2, with both variables significantly predicting time to menopause (Group 1: HR 1.54 for age and HR 0.93 for mother's age; Group 2: HR 1.59 for age and HR 0.89 for mother's age; p < 0.001 for all). In Group 2, adding AMH to the multivariable model increased the C-statistic to 92%, and only female age (HR 1.41, p < 0.0001) and AMH (HR 0.06, p < 0.0001) remained significant predictors, while mother's age did not (HR 0.93, p = 0.08). The mean weighted net reclassification index showed a 47% improvement in predictive accuracy when adding AMH to the model of age and mother's age at natural menopause.

Why it matters

Anti-Müllerian hormone is a substantially more accurate predictor of remaining time to menopause than maternal age at menopause, providing meaningful incremental prognostic value over family history.

Limits

The cohort evaluating AMH was relatively small (150 women) and pooled from different studies. Individual prediction certainty remains too low for immediate clinical application, and the study was restricted to women with regular menstrual cycles.

Cited by