Abstract
Earlier menopause is associated with increased risks of type 2 diabetes, osteoporosis, heart disease, depression, and mortality. Although menopause is typically defined as the absence of menstruation for 12 consecutive months, estimating age of menopause using repeated, prospective data can be challenging. In this data note, we describe an algorithm developed to estimate age at natural menopause using data from the Avon Longitudinal Study of Parents and Children (ALSPAC) Mothers cohort and highlight the complexities involved. Between 2008 and 2020, women were asked about their menstrual history, including when they last had a menstrual period (LMP), reasons for period cessation if relevant, and contraception and hormone replacement therapy (HRT) use at up to eight timepoints. Our algorithm estimated LMP at each timepoint using three key variables - self-reported LMP, menstruation in the last 3 months, and last 12 months - and accounted for factors affecting menstruation such as surgery, contraception, and HRT. Age at natural menopause was then derived based on the repeated LMP estimates across timepoints. Of 5,949 women included in the analysis (mean 3.6 timepoints per participant), age at natural menopause was estimated for 2,422 women. The mean estimated age was 49.4 years (SD = 4.1, range: 30-63 years, median = 50 years). This data note introduces potential users of the ALSPAC data to our algorithm and highlights key challenges when using longitudinal data to estimate menopause timing including irregular bleeding, missing data, and conflicting reports. Our aim is to support researchers wishing to use this derived variable in future studies of reproductive ageing.