Comparison of SWAN and WISE menopausal status classification algorithms.

BACKGROUND: Classification of menopausal status is important for epidemiological and clinical studies as well as for clinicians treating midlife women. Most epidemiological studies, including the Study of Women's Health Across the Nation (SWAN), classify women based on self-reported bleeding history.
METHODS: The Women's Ischemia Syndrome Evaluation (WISE) study developed an algorithm using menstrual and reproductive history and serum hormone levels to reproduce the menopausal status classifications assigned by the WISE hormone committee. We applied that algorithm to women participating in SWAN and examined characteristics of women with concordant and discordant SWAN and WISE classifications.
RESULTS: Of the 3215 SWAN women with complete information at baseline (1995-1997), 2466 (76.7%) received concordant classifications (kappa = 0.52); at the fifth annual follow-up visit, of the 1623 women with complete information, 1154 (72.7%) received concordant classifications (kappa = 0.57). At each time point, we identified subgroups of women with discordant SWAN and WISE classifications. These subgroups, ordered by chronological age, showed increasing trends for menopausal symptoms and follicle-stimulating hormone (FSH) and a decreasing trend for estrogen (p < 0.001).
CONCLUSIONS: The WISE algorithm is a useful tool for studies that have access to blood samples for hormone data unrelated to menstrual cycle phase, with or without an intact uterus, and no resources for adjudication. Future studies may want to combine aspects of the SWAN and WISE algorithms by adding hormonal measures to the series of bleeding questions in order to determine more precisely where women are in the perimenopausal continuum.