Evidence for a secular trend in menopausal age: a population study of women in Gothenburg.

OBJECTIVE: To describe secular trends in age of natural menopause.
DESIGN: A prospective study based on a random sample of the total female population in Gothenburg, Sweden, started in 1968 with follow-ups in 1974-75, 1980-81, 1992-93, and 2000-02. PARTICIPANTS: 1,462 women born in 1930, 1922, 1918, 1914, and 1908 (participation rate, 90.1%) representative of women of the same ages in the general population. Information regarding menopausal age was provided by 1,373 of the 1,462 women (93.9%). The number was further reduced to 1,017 after exclusion of women who had taken hormones, undergone a surgical menopause, or both.
RESULTS: The mean age at natural menopause showed a steady increase across birth cohorts. Trends were similar in women who had smoked and women who had never smoked, even after adjusting for different covariates. The upward trend was 0.1 years per birth year (SE 0.020, P < 0.0001). Interestingly, women with earlier menarche had a somewhat earlier age at menopause, independent of the cohort effect. When hormone users were included in the sample, the cohort effect was also found to be independent of oral contraceptive use and hormone therapy.
CONCLUSIONS: This study has shown that, independent of variations in socioeconomic status, smoking status, oral contraceptive use, or hormone therapy use, as well as other potential confounders, there was a highly significant secular trend of increase in menopausal age. The observation of a positive association between menarche and menopausal age has, to our knowledge, not previously been described.